diff --git a/3rdparty/taglib/CMakeLists.txt b/3rdparty/taglib/CMakeLists.txt
index cacd58f54..fe2cf4f8b 100644
--- a/3rdparty/taglib/CMakeLists.txt
+++ b/3rdparty/taglib/CMakeLists.txt
@@ -7,6 +7,15 @@ math(EXPR TAGLIB_SOVERSION_MAJOR "${TAGLIB_SOVERSION_CURRENT} - ${TAGLIB_SOVERSI
 math(EXPR TAGLIB_SOVERSION_MINOR "${TAGLIB_SOVERSION_AGE}")
 math(EXPR TAGLIB_SOVERSION_PATCH "${TAGLIB_SOVERSION_REVISION}")
 
+include(TestBigEndian)
+test_big_endian(IS_BIG_ENDIAN)
+
+if(NOT IS_BIG_ENDIAN)
+  add_definitions(-DSYSTEM_BYTEORDER=1)
+else()
+  add_definitions(-DSYSTEM_BYTEORDER=2)
+endif()
+
 configure_file(taglib_config.h.cmake ${CMAKE_CURRENT_BINARY_DIR}/taglib_config.h)
 
 set(CMAKE_INCLUDE_CURRENT_DIR ON)
@@ -21,6 +30,7 @@ include_directories(
   ${CMAKE_CURRENT_SOURCE_DIR}/mp4
   ${CMAKE_CURRENT_SOURCE_DIR}/ogg/vorbis
   ${CMAKE_CURRENT_SOURCE_DIR}/ogg/speex
+  ${CMAKE_CURRENT_SOURCE_DIR}/ogg/opus
   ${CMAKE_CURRENT_SOURCE_DIR}/mpeg/id3v2
   ${CMAKE_CURRENT_SOURCE_DIR}/mpeg/id3v2/frames
   ${CMAKE_CURRENT_SOURCE_DIR}/mpeg/id3v1
@@ -60,6 +70,8 @@ set(tag_HDRS
   toolkit/tmap.h
   toolkit/tmap.tcc
   toolkit/tpropertymap.h
+  toolkit/trefcounter.h
+  toolkit/tdebuglistener.h
   mpeg/mpegfile.h
   mpeg/mpegproperties.h
   mpeg/mpegheader.h
@@ -94,6 +106,8 @@ set(tag_HDRS
   ogg/flac/oggflacfile.h
   ogg/speex/speexfile.h
   ogg/speex/speexproperties.h
+  ogg/opus/opusfile.h
+  ogg/opus/opusproperties.h
   flac/flacfile.h
   flac/flacpicture.h
   flac/flacproperties.h
@@ -114,6 +128,7 @@ set(tag_HDRS
   riff/aiff/aiffproperties.h
   riff/wav/wavfile.h
   riff/wav/wavproperties.h
+  riff/wav/infotag.h
   asf/asffile.h
   asf/asfproperties.h
   asf/asftag.h
@@ -230,6 +245,11 @@ set(speex_SRCS
   ogg/speex/speexproperties.cpp
 )
 
+set(opus_SRCS
+  ogg/opus/opusfile.cpp
+  ogg/opus/opusproperties.cpp
+)
+
 set(trueaudio_SRCS
   trueaudio/trueaudiofile.cpp
   trueaudio/trueaudioproperties.cpp
@@ -255,6 +275,7 @@ set(aiff_SRCS
 set(wav_SRCS
   riff/wav/wavfile.cpp
   riff/wav/wavproperties.cpp
+  riff/wav/infotag.cpp
 )
 
 set(mod_SRCS
@@ -290,6 +311,8 @@ set(toolkit_SRCS
   toolkit/tfilestream.cpp
   toolkit/tdebug.cpp
   toolkit/tpropertymap.cpp
+  toolkit/trefcounter.cpp
+  toolkit/tdebuglistener.cpp
   toolkit/unicode.cpp
 )
 
@@ -297,7 +320,7 @@ set(tag_LIB_SRCS
   ${mpeg_SRCS} ${id3v1_SRCS} ${id3v2_SRCS} ${frames_SRCS} ${ogg_SRCS}
   ${vorbis_SRCS} ${oggflacs_SRCS} ${mpc_SRCS} ${ape_SRCS} ${toolkit_SRCS} ${flacs_SRCS}
   ${wavpack_SRCS} ${speex_SRCS} ${trueaudio_SRCS} ${riff_SRCS} ${aiff_SRCS} ${wav_SRCS}
-  ${asf_SRCS} ${mp4_SRCS} ${mod_SRCS} ${s3m_SRCS} ${it_SRCS} ${xm_SRCS}
+  ${asf_SRCS} ${mp4_SRCS} ${mod_SRCS} ${s3m_SRCS} ${it_SRCS} ${xm_SRCS} ${opus_SRCS}
   tag.cpp
   tagunion.cpp
   fileref.cpp
diff --git a/3rdparty/taglib/ape/apefile.cpp b/3rdparty/taglib/ape/apefile.cpp
index cb6522427..3cb9d9a70 100644
--- a/3rdparty/taglib/ape/apefile.cpp
+++ b/3rdparty/taglib/ape/apefile.cpp
@@ -47,7 +47,7 @@ using namespace TagLib;
 
 namespace
 {
-  enum { ApeAPEIndex, ApeID3v1Index };
+  enum { ApeAPEIndex = 0, ApeID3v1Index = 1 };
 }
 
 class APE::File::FilePrivate
@@ -90,14 +90,16 @@ APE::File::File(FileName file, bool readProperties,
                 Properties::ReadStyle propertiesStyle) : TagLib::File(file)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 APE::File::File(IOStream *stream, bool readProperties,
                 Properties::ReadStyle propertiesStyle) : TagLib::File(stream)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 APE::File::~File()
@@ -129,12 +131,9 @@ void APE::File::removeUnsupportedProperties(const StringList &properties)
 
 PropertyMap APE::File::setProperties(const PropertyMap &properties)
 {
-  if(d->hasAPE)
-    return d->tag.access<APE::Tag>(ApeAPEIndex, false)->setProperties(properties);
-  else if(d->hasID3v1)
-    return d->tag.access<ID3v1::Tag>(ApeID3v1Index, false)->setProperties(properties);
-  else
-    return d->tag.access<APE::Tag>(ApeAPEIndex, true)->setProperties(properties);
+  if(d->hasID3v1)
+    d->tag.access<ID3v1::Tag>(ApeID3v1Index, false)->setProperties(properties);
+  return d->tag.access<APE::Tag>(ApeAPEIndex, true)->setProperties(properties);
 }
 
 APE::Properties *APE::File::audioProperties() const
@@ -236,6 +235,16 @@ void APE::File::strip(int tags)
   }
 }
 
+bool APE::File::hasAPETag() const
+{
+  return d->hasAPE;
+}
+
+bool APE::File::hasID3v1Tag() const
+{
+  return d->hasID3v1;
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // private members
 ////////////////////////////////////////////////////////////////////////////////
diff --git a/3rdparty/taglib/ape/apefile.h b/3rdparty/taglib/ape/apefile.h
index 8b187f6a6..f7b509f40 100644
--- a/3rdparty/taglib/ape/apefile.h
+++ b/3rdparty/taglib/ape/apefile.h
@@ -59,7 +59,7 @@ namespace TagLib {
     //! An implementation of TagLib::File with APE specific methods
 
     /*!
-     * This implements and provides an interface APE WavPack files to the
+     * This implements and provides an interface for APE files to the
      * TagLib::Tag and TagLib::AudioProperties interfaces by way of implementing
      * the abstract TagLib::File API as well as providing some additional
      * information specific to APE files.
@@ -84,20 +84,22 @@ namespace TagLib {
       };
 
       /*!
-       * Contructs an WavPack file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an APE file from \a file.  If \a readProperties is true the
+       * file's audio properties will also be read.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(FileName file, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs an WavPack file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an APE file from \a stream.  If \a readProperties is true the
+       * file's audio properties will also be read.
        *
        * \note TagLib will *not* take ownership of the stream, the caller is
        * responsible for deleting it after the File object.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(IOStream *stream, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
@@ -128,10 +130,11 @@ namespace TagLib {
 
       /*!
        * Implements the unified property interface -- import function.
-       * As for the export, only one tag is taken into account. If the file
-       * has no tag at all, APE will be created.
+       * Creates an APEv2 tag if necessary. A pontentially existing ID3v1
+       * tag will be updated as well.
        */
       PropertyMap setProperties(const PropertyMap &);
+
       /*!
        * Returns the APE::Properties for this file.  If no audio properties
        * were read then this will return a null pointer.
@@ -149,27 +152,38 @@ namespace TagLib {
       /*!
        * Returns a pointer to the ID3v1 tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
+       * If \a create is false (the default) this may return a null pointer
        * if there is no valid ID3v1 tag.  If \a create is true it will create
-       * an ID3v1 tag if one does not exist. If there is already an APE tag, the
-       * new ID3v1 tag will be placed after it.
+       * an ID3v1 tag if one does not exist and returns a valid pointer.
        *
-       * \note The Tag <b>is still</b> owned by the APE::File and should not be
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an ID3v1 tag.  Use hasID3v1Tag() to check if the file 
+       * on disk actually has an ID3v1 tag.
+       *
+       * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasID3v1Tag()
        */
       ID3v1::Tag *ID3v1Tag(bool create = false);
 
       /*!
        * Returns a pointer to the APE tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
+       * If \a create is false (the default) this may return a null pointer
        * if there is no valid APE tag.  If \a create is true it will create
-       * a APE tag if one does not exist.
+       * an APE tag if one does not exist and returns a valid pointer.
        *
-       * \note The Tag <b>is still</b> owned by the APE::File and should not be
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an APE tag.  Use hasAPETag() to check if the file 
+       * on disk actually has an APE tag.
+       *
+       * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasAPETag()
        */
       APE::Tag *APETag(bool create = false);
 
@@ -183,6 +197,20 @@ namespace TagLib {
        */
       void strip(int tags = AllTags);
 
+      /*!
+       * Returns whether or not the file on disk actually has an APE tag.
+       *
+       * \see APETag()
+       */
+      bool hasAPETag() const;
+
+      /*!
+       * Returns whether or not the file on disk actually has an ID3v1 tag.
+       *
+       * \see ID3v1Tag()
+       */
+      bool hasID3v1Tag() const;
+
     private:
       File(const File &);
       File &operator=(const File &);
diff --git a/3rdparty/taglib/ape/apefooter.cpp b/3rdparty/taglib/ape/apefooter.cpp
index 688649016..539832baf 100644
--- a/3rdparty/taglib/ape/apefooter.cpp
+++ b/3rdparty/taglib/ape/apefooter.cpp
@@ -177,19 +177,19 @@ void APE::Footer::parse(const ByteVector &data)
 
   // Read the version number
 
-  d->version = data.mid(8, 4).toUInt(false);
+  d->version = data.toUInt(8, false);
 
   // Read the tag size
 
-  d->tagSize = data.mid(12, 4).toUInt(false);
+  d->tagSize = data.toUInt(12, false);
 
   // Read the item count
 
-  d->itemCount = data.mid(16, 4).toUInt(false);
+  d->itemCount = data.toUInt(16, false);
 
   // Read the flags
 
-  std::bitset<32> flags(TAGLIB_CONSTRUCT_BITSET(data.mid(20, 4).toUInt(false)));
+  std::bitset<32> flags(TAGLIB_CONSTRUCT_BITSET(data.toUInt(20, false)));
 
   d->headerPresent = flags[31];
   d->footerPresent = !flags[30];
diff --git a/3rdparty/taglib/ape/apeitem.cpp b/3rdparty/taglib/ape/apeitem.cpp
index 9d9647ec3..3490173a2 100644
--- a/3rdparty/taglib/ape/apeitem.cpp
+++ b/3rdparty/taglib/ape/apeitem.cpp
@@ -125,6 +125,7 @@ void APE::Item::setBinaryData(const ByteVector &value)
 {
   d->type = Binary;
   d->value = value;
+  d->text.clear();
 }
 
 ByteVector APE::Item::value() const
@@ -137,31 +138,35 @@ ByteVector APE::Item::value() const
 
 void APE::Item::setKey(const String &key)
 {
-    d->key = key;
+  d->key = key;
 }
 
 void APE::Item::setValue(const String &value)
 {
-    d->type = Text;
-    d->text = value;
+  d->type = Text;
+  d->text = value;
+  d->value.clear();
 }
 
 void APE::Item::setValues(const StringList &value)
 {
-    d->type = Text;
-    d->text = value;
+  d->type = Text;
+  d->text = value;
+  d->value.clear();
 }
 
 void APE::Item::appendValue(const String &value)
 {
-    d->type = Text;
-    d->text.append(value);
+  d->type = Text;
+  d->text.append(value);
+  d->value.clear();
 }
 
 void APE::Item::appendValues(const StringList &values)
 {
-    d->type = Text;
-    d->text.append(values);
+  d->type = Text;
+  d->text.append(values);
+  d->value.clear();
 }
 
 int APE::Item::size() const
@@ -200,7 +205,10 @@ StringList APE::Item::values() const
 
 String APE::Item::toString() const
 {
-  return isEmpty() ? String::null : d->text.front();
+  if(d->type == Text && !isEmpty())
+    return d->text.front();
+  else
+    return String::null;
 }
 
 bool APE::Item::isEmpty() const
@@ -229,18 +237,20 @@ void APE::Item::parse(const ByteVector &data)
     return;
   }
 
-  uint valueLength  = data.mid(0, 4).toUInt(false);
-  uint flags        = data.mid(4, 4).toUInt(false);
+  const uint valueLength  = data.toUInt(0, false);
+  const uint flags        = data.toUInt(4, false);
 
   d->key = String(data.mid(8), String::UTF8);
 
-  d->value = data.mid(8 + d->key.size() + 1, valueLength);
+  const ByteVector value = data.mid(8 + d->key.size() + 1, valueLength);
 
   setReadOnly(flags & 1);
   setType(ItemTypes((flags >> 1) & 3));
 
-  if(Text == d->type)
-    d->text = StringList(ByteVectorList::split(d->value, '\0'), String::UTF8);
+  if(Text == d->type) 
+    d->text = StringList(ByteVectorList::split(value, '\0'), String::UTF8);
+  else
+    d->value = value;
 }
 
 ByteVector APE::Item::render() const
diff --git a/3rdparty/taglib/ape/apeitem.h b/3rdparty/taglib/ape/apeitem.h
index f7fd05e3c..0588d1850 100644
--- a/3rdparty/taglib/ape/apeitem.h
+++ b/3rdparty/taglib/ape/apeitem.h
@@ -97,7 +97,7 @@ namespace TagLib {
 
       /*!
        * Returns the binary value.
-       * If the item type is not \a Binary, the returned contents are undefined
+       * If the item type is not \a Binary, always returns an empty ByteVector.
        */
       ByteVector binaryData() const;
 
@@ -152,8 +152,9 @@ namespace TagLib {
       int size() const;
 
       /*!
-       * Returns the value as a single string. In case of multiple strings,
-       * the first is returned.
+       * Returns the value as a single string.  In case of multiple strings,
+       * the first is returned.  If the data type is not \a Text, always returns 
+       * an empty String.
        */
       String toString() const;
 
@@ -163,7 +164,8 @@ namespace TagLib {
 #endif
 
       /*!
-       * Returns the list of text values.
+       * Returns the list of text values.  If the data type is not \a Text, always 
+       * returns an empty StringList.
        */
       StringList values() const;
 
diff --git a/3rdparty/taglib/ape/apeproperties.cpp b/3rdparty/taglib/ape/apeproperties.cpp
index b4fcec6ad..4940edead 100644
--- a/3rdparty/taglib/ape/apeproperties.cpp
+++ b/3rdparty/taglib/ape/apeproperties.cpp
@@ -125,10 +125,10 @@ void APE::Properties::read()
 
   // Then we read the header common for all versions of APE
   d->file->seek(offset);
-  ByteVector commonHeader=d->file->readBlock(6);
+  ByteVector commonHeader = d->file->readBlock(6);
   if(!commonHeader.startsWith("MAC "))
     return;
-  d->version = commonHeader.mid(4).toUInt(false);
+  d->version = commonHeader.toUShort(4, false);
 
   if(d->version >= 3980) {
     analyzeCurrent();
@@ -182,7 +182,7 @@ void APE::Properties::analyzeCurrent()
   // Read the descriptor
   d->file->seek(2, File::Current);
   ByteVector descriptor = d->file->readBlock(44);
-  uint descriptorBytes = descriptor.mid(0,4).toUInt(false);
+  const uint descriptorBytes = descriptor.toUInt(0, false);
 
   if ((descriptorBytes - 52) > 0)
     d->file->seek(descriptorBytes - 52, File::Current);
@@ -191,14 +191,14 @@ void APE::Properties::analyzeCurrent()
   ByteVector header = d->file->readBlock(24);
 
   // Get the APE info
-  d->channels = header.mid(18, 2).toShort(false);
-  d->sampleRate = header.mid(20, 4).toUInt(false);
-  d->bitsPerSample = header.mid(16, 2).toShort(false);
+  d->channels      = header.toShort(18, false);
+  d->sampleRate    = header.toUInt(20, false);
+  d->bitsPerSample = header.toShort(16, false);
   //d->compressionLevel =
 
-  uint totalFrames = header.mid(12, 4).toUInt(false);
-  uint blocksPerFrame = header.mid(4, 4).toUInt(false);
-  uint finalFrameBlocks = header.mid(8, 4).toUInt(false);
+  const uint totalFrames      = header.toUInt(12, false);
+  const uint blocksPerFrame   = header.toUInt(4, false);
+  const uint finalFrameBlocks = header.toUInt(8, false);
   d->sampleFrames = totalFrames > 0 ? (totalFrames -  1) * blocksPerFrame + finalFrameBlocks : 0;
   d->length = d->sampleRate > 0 ? d->sampleFrames / d->sampleRate : 0;
   d->bitrate = d->length > 0 ? ((d->streamLength * 8L) / d->length) / 1000 : 0;
@@ -207,13 +207,13 @@ void APE::Properties::analyzeCurrent()
 void APE::Properties::analyzeOld()
 {
   ByteVector header = d->file->readBlock(26);
-  uint totalFrames = header.mid(18, 4).toUInt(false);
+  const uint totalFrames = header.toUInt(18, false);
 
   // Fail on 0 length APE files (catches non-finalized APE files)
   if(totalFrames == 0)
     return;
 
-  short compressionLevel = header.mid(0, 2).toShort(false);
+  const short compressionLevel = header.toShort(0, false);
   uint blocksPerFrame;
   if(d->version >= 3950)
     blocksPerFrame = 73728 * 4;
@@ -221,10 +221,11 @@ void APE::Properties::analyzeOld()
     blocksPerFrame = 73728;
   else
     blocksPerFrame = 9216;
-  d->channels = header.mid(4, 2).toShort(false);
-  d->sampleRate = header.mid(6, 4).toUInt(false);
-  uint finalFrameBlocks = header.mid(22, 4).toUInt(false);
-  uint totalBlocks = totalFrames > 0 ? (totalFrames - 1) * blocksPerFrame + finalFrameBlocks : 0;
+  d->channels   = header.toShort(4, false);
+  d->sampleRate = header.toUInt(6, false);
+  const uint finalFrameBlocks = header.toUInt(22, false);
+  const uint totalBlocks 
+    = totalFrames > 0 ? (totalFrames - 1) * blocksPerFrame + finalFrameBlocks : 0;
   d->length = totalBlocks / d->sampleRate;
   d->bitrate = d->length > 0 ? ((d->streamLength * 8L) / d->length) / 1000 : 0;
 }
diff --git a/3rdparty/taglib/asf/asfattribute.cpp b/3rdparty/taglib/asf/asfattribute.cpp
index 2cfada7b0..4ee2d0a18 100644
--- a/3rdparty/taglib/asf/asfattribute.cpp
+++ b/3rdparty/taglib/asf/asfattribute.cpp
@@ -23,12 +23,9 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <taglib.h>
 #include <tdebug.h>
+#include "trefcounter.h"
 #include "asfattribute.h"
 #include "asffile.h"
 
diff --git a/3rdparty/taglib/asf/asffile.cpp b/3rdparty/taglib/asf/asffile.cpp
index 455631f82..241998ca8 100644
--- a/3rdparty/taglib/asf/asffile.cpp
+++ b/3rdparty/taglib/asf/asffile.cpp
@@ -23,12 +23,9 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <tdebug.h>
 #include <tbytevectorlist.h>
+#include <tpropertymap.h>
 #include <tstring.h>
 #include "asffile.h"
 #include "asftag.h"
@@ -186,7 +183,8 @@ ByteVector ASF::File::FilePropertiesObject::guid()
 void ASF::File::FilePropertiesObject::parse(ASF::File *file, uint size)
 {
   BaseObject::parse(file, size);
-  file->d->properties->setLength((int)(data.mid(40, 8).toLongLong(false) / 10000000L - data.mid(56, 8).toLongLong(false) / 1000L));
+  file->d->properties->setLength(
+    (int)(data.toLongLong(40, false) / 10000000L - data.toLongLong(56, false) / 1000L));
 }
 
 ByteVector ASF::File::StreamPropertiesObject::guid()
@@ -197,9 +195,9 @@ ByteVector ASF::File::StreamPropertiesObject::guid()
 void ASF::File::StreamPropertiesObject::parse(ASF::File *file, uint size)
 {
   BaseObject::parse(file, size);
-  file->d->properties->setChannels(data.mid(56, 2).toShort(false));
-  file->d->properties->setSampleRate(data.mid(58, 4).toUInt(false));
-  file->d->properties->setBitrate(data.mid(62, 4).toUInt(false) * 8 / 1000);
+  file->d->properties->setChannels(data.toShort(56, false));
+  file->d->properties->setSampleRate(data.toUInt(58, false));
+  file->d->properties->setBitrate(data.toUInt(62, false) * 8 / 1000);
 }
 
 ByteVector ASF::File::ContentDescriptionObject::guid()
@@ -372,14 +370,16 @@ ASF::File::File(FileName file, bool readProperties, Properties::ReadStyle proper
   : TagLib::File(file)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 ASF::File::File(IOStream *stream, bool readProperties, Properties::ReadStyle propertiesStyle)
   : TagLib::File(stream)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 ASF::File::~File()
@@ -401,6 +401,21 @@ ASF::Tag *ASF::File::tag() const
   return d->tag;
 }
 
+PropertyMap ASF::File::properties() const
+{
+  return d->tag->properties();
+}
+
+void ASF::File::removeUnsupportedProperties(const StringList &properties)
+{
+  d->tag->removeUnsupportedProperties(properties);
+}
+
+PropertyMap ASF::File::setProperties(const PropertyMap &properties)
+{
+  return d->tag->setProperties(properties);
+}
+
 ASF::Properties *ASF::File::audioProperties() const
 {
   return d->properties;
diff --git a/3rdparty/taglib/asf/asffile.h b/3rdparty/taglib/asf/asffile.h
index 3a7eef565..5ccf2fdeb 100644
--- a/3rdparty/taglib/asf/asffile.h
+++ b/3rdparty/taglib/asf/asffile.h
@@ -48,30 +48,27 @@ namespace TagLib {
     public:
 
       /*!
-       * Contructs an ASF file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an ASF file from \a file.
        *
        * \note In the current implementation, both \a readProperties and
-       * \a propertiesStyle are ignored.
-       *
-       * \note TagLib will *not* take ownership of the stream, the caller is
-       * responsible for deleting it after the File object.
+       * \a propertiesStyle are ignored.  The audio properties are always
+       * read.
        */
-      File(FileName file, bool readProperties = true, Properties::ReadStyle propertiesStyle = Properties::Average);
+      File(FileName file, bool readProperties = true, 
+           Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs an ASF file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an ASF file from \a stream.
        *
        * \note In the current implementation, both \a readProperties and
-       * \a propertiesStyle are ignored.
+       * \a propertiesStyle are ignored.  The audio properties are always
+       * read.
        *
        * \note TagLib will *not* take ownership of the stream, the caller is
        * responsible for deleting it after the File object.
        */
-      File(IOStream *stream, bool readProperties = true, Properties::ReadStyle propertiesStyle = Properties::Average);
+      File(IOStream *stream, bool readProperties = true, 
+           Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
        * Destroys this instance of the File.
@@ -90,6 +87,22 @@ namespace TagLib {
        */
       virtual Tag *tag() const;
 
+      /*!
+       * Implements the unified property interface -- export function.
+       */
+      PropertyMap properties() const;
+
+      /*!
+       * Removes unsupported properties. Forwards to the actual Tag's
+       * removeUnsupportedProperties() function.
+       */
+      void removeUnsupportedProperties(const StringList &properties);
+
+      /*!
+       * Implements the unified property interface -- import function.
+       */
+      PropertyMap setProperties(const PropertyMap &);
+
       /*!
        * Returns the ASF audio properties for this file.
        */
@@ -103,7 +116,6 @@ namespace TagLib {
       virtual bool save();
 
     private:
-
       int readBYTE(bool *ok = 0);
       int readWORD(bool *ok = 0);
       unsigned int readDWORD(bool *ok = 0);
diff --git a/3rdparty/taglib/asf/asfpicture.cpp b/3rdparty/taglib/asf/asfpicture.cpp
index c36ffa3a9..999f92043 100644
--- a/3rdparty/taglib/asf/asfpicture.cpp
+++ b/3rdparty/taglib/asf/asfpicture.cpp
@@ -23,12 +23,9 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <taglib.h>
 #include <tdebug.h>
+#include "trefcounter.h"
 #include "asfattribute.h"
 #include "asffile.h"
 #include "asfpicture.h"
@@ -149,7 +146,7 @@ void ASF::Picture::parse(const ByteVector& bytes)
     return;
   int pos = 0;
   d->type = (Type)bytes[0]; ++pos;
-  uint dataLen = bytes.mid(pos, 4).toUInt(false); pos+=4;
+  const uint dataLen = bytes.toUInt(pos, false); pos+=4;
 
   const ByteVector nullStringTerminator(2, 0);
 
diff --git a/3rdparty/taglib/asf/asfproperties.cpp b/3rdparty/taglib/asf/asfproperties.cpp
index 835cbdf98..b82c131a7 100644
--- a/3rdparty/taglib/asf/asfproperties.cpp
+++ b/3rdparty/taglib/asf/asfproperties.cpp
@@ -23,10 +23,6 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <tdebug.h>
 #include <tstring.h>
 #include "asfproperties.h"
diff --git a/3rdparty/taglib/asf/asftag.cpp b/3rdparty/taglib/asf/asftag.cpp
index b7b541fcb..c07abe94d 100644
--- a/3rdparty/taglib/asf/asftag.cpp
+++ b/3rdparty/taglib/asf/asftag.cpp
@@ -23,10 +23,7 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
+#include <tpropertymap.h>
 #include "asftag.h"
 
 using namespace TagLib;
@@ -196,3 +193,162 @@ bool ASF::Tag::isEmpty() const
          d->attributeListMap.isEmpty();
 }
 
+static const char *keyTranslation[][2] = {
+  { "WM/AlbumTitle", "ALBUM" },
+  { "WM/Composer", "COMPOSER" },
+  { "WM/Writer", "WRITER" },
+  { "WM/Conductor", "CONDUCTOR" },
+  { "WM/ModifiedBy", "REMIXER" },
+  { "WM/Year", "DATE" },
+  { "WM/OriginalReleaseYear", "ORIGINALDATE" },
+  { "WM/Producer", "PRODUCER" },
+  { "WM/ContentGroupDescription", "GROUPING" },
+  { "WM/SubTitle", "SUBTITLE" },
+  { "WM/SetSubTitle", "DISCSUBTITLE" },
+  { "WM/TrackNumber", "TRACKNUMBER" },
+  { "WM/PartOfSet", "DISCNUMBER" },
+  { "WM/Genre", "GENRE" },
+  { "WM/BeatsPerMinute", "BPM" },
+  { "WM/Mood", "MOOD" },
+  { "WM/ISRC", "ISRC" },
+  { "WM/Lyrics", "LYRICS" },
+  { "WM/Media", "MEDIA" },
+  { "WM/Publisher", "LABEL" },
+  { "WM/CatalogNo", "CATALOGNUMBER" },
+  { "WM/Barcode", "BARCODE" },
+  { "WM/EncodedBy", "ENCODEDBY" },
+  { "WM/AlbumSortOrder", "ALBUMSORT" },
+  { "WM/AlbumArtistSortOrder", "ALBUMARTISTSORT" },
+  { "WM/ArtistSortOrder", "ARTISTSORT" },
+  { "WM/TitleSortOrder", "TITLESORT" },
+  { "WM/Script", "SCRIPT" },
+  { "WM/Language", "LANGUAGE" },
+  { "MusicBrainz/Track Id", "MUSICBRAINZ_TRACKID" },
+  { "MusicBrainz/Artist Id", "MUSICBRAINZ_ARTISTID" },
+  { "MusicBrainz/Album Id", "MUSICBRAINZ_ALBUMID" },
+  { "MusicBrainz/Album Artist Id", "MUSICBRAINZ_ALBUMARTISTID" },
+  { "MusicBrainz/Release Group Id", "MUSICBRAINZ_RELEASEGROUPID" },
+  { "MusicBrainz/Work Id", "MUSICBRAINZ_WORKID" },
+  { "MusicIP/PUID", "MUSICIP_PUID" },
+  { "Acoustid/Id", "ACOUSTID_ID" },
+  { "Acoustid/Fingerprint", "ACOUSTID_FINGERPRINT" },
+};
+
+PropertyMap ASF::Tag::properties() const
+{
+  static Map<String, String> keyMap;
+  if(keyMap.isEmpty()) {
+    int numKeys = sizeof(keyTranslation) / sizeof(keyTranslation[0]);
+    for(int i = 0; i < numKeys; i++) {
+      keyMap[keyTranslation[i][0]] = keyTranslation[i][1];
+    }
+  }
+
+  PropertyMap props;
+
+  if(!d->title.isEmpty()) {
+    props["TITLE"] = d->title;
+  }
+  if(!d->artist.isEmpty()) {
+    props["ARTIST"] = d->artist;
+  }
+  if(!d->copyright.isEmpty()) {
+    props["COPYRIGHT"] = d->copyright;
+  }
+  if(!d->comment.isEmpty()) {
+    props["COMMENT"] = d->comment;
+  }
+
+  ASF::AttributeListMap::ConstIterator it = d->attributeListMap.begin();
+  for(; it != d->attributeListMap.end(); ++it) {
+    if(keyMap.contains(it->first)) {
+      String key = keyMap[it->first];
+      AttributeList::ConstIterator it2 = it->second.begin();
+      for(; it2 != it->second.end(); ++it2) {
+        if(key == "TRACKNUMBER") {
+          if(it2->type() == ASF::Attribute::DWordType)
+            props.insert(key, String::number(it2->toUInt()));
+          else
+            props.insert(key, it2->toString());
+        }
+        else {
+          props.insert(key, it2->toString());
+        }
+      }
+    }
+    else {
+      props.unsupportedData().append(it->first);
+    }
+  }
+  return props;
+}
+
+void ASF::Tag::removeUnsupportedProperties(const StringList &props)
+{
+  StringList::ConstIterator it = props.begin();
+  for(; it != props.end(); ++it)
+    d->attributeListMap.erase(*it);
+}
+
+PropertyMap ASF::Tag::setProperties(const PropertyMap &props)
+{
+  static Map<String, String> reverseKeyMap;
+  if(reverseKeyMap.isEmpty()) {
+    int numKeys = sizeof(keyTranslation) / sizeof(keyTranslation[0]);
+    for(int i = 0; i < numKeys; i++) {
+      reverseKeyMap[keyTranslation[i][1]] = keyTranslation[i][0];
+    }
+  }
+
+  PropertyMap origProps = properties();
+  PropertyMap::ConstIterator it = origProps.begin();
+  for(; it != origProps.end(); ++it) {
+    if(!props.contains(it->first) || props[it->first].isEmpty()) {
+      if(it->first == "TITLE") {
+        d->title = String::null;
+      }
+      else if(it->first == "ARTIST") {
+        d->artist = String::null;
+      }
+      else if(it->first == "COMMENT") {
+        d->comment = String::null;
+      }
+      else if(it->first == "COPYRIGHT") {
+        d->copyright = String::null;
+      }
+      else {
+        d->attributeListMap.erase(reverseKeyMap[it->first]);
+      }
+    }
+  }
+
+  PropertyMap ignoredProps;
+  it = props.begin();
+  for(; it != props.end(); ++it) {
+    if(reverseKeyMap.contains(it->first)) {
+      String name = reverseKeyMap[it->first];
+      removeItem(name);
+      StringList::ConstIterator it2 = it->second.begin();
+      for(; it2 != it->second.end(); ++it2) {
+        addAttribute(name, *it2);
+      }
+    }
+    else if(it->first == "TITLE") {
+      d->title = it->second.toString();
+    }
+    else if(it->first == "ARTIST") {
+      d->artist = it->second.toString();
+    }
+    else if(it->first == "COMMENT") {
+      d->comment = it->second.toString();
+    }
+    else if(it->first == "COPYRIGHT") {
+      d->copyright = it->second.toString();
+    }
+    else {
+      ignoredProps.insert(it->first, it->second);
+    }
+  }
+
+  return ignoredProps;
+}
diff --git a/3rdparty/taglib/asf/asftag.h b/3rdparty/taglib/asf/asftag.h
index 6b2d2bf35..f68579d88 100644
--- a/3rdparty/taglib/asf/asftag.h
+++ b/3rdparty/taglib/asf/asftag.h
@@ -176,6 +176,10 @@ namespace TagLib {
        */
       void addAttribute(const String &name, const Attribute &attribute);
 
+      PropertyMap properties() const;
+      void removeUnsupportedProperties(const StringList& properties);
+      PropertyMap setProperties(const PropertyMap &properties);
+
     private:
 
       class TagPrivate;
diff --git a/3rdparty/taglib/fileref.cpp b/3rdparty/taglib/fileref.cpp
index 5c42ed4c1..4403a5fbf 100644
--- a/3rdparty/taglib/fileref.cpp
+++ b/3rdparty/taglib/fileref.cpp
@@ -27,13 +27,10 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <tfile.h>
 #include <tstring.h>
 #include <tdebug.h>
+#include "trefcounter.h"
 
 #include "fileref.h"
 #include "asffile.h"
@@ -45,6 +42,7 @@
 #include "mp4file.h"
 #include "wavpackfile.h"
 #include "speexfile.h"
+#include "opusfile.h"
 #include "trueaudiofile.h"
 #include "aifffile.h"
 #include "wavfile.h"
@@ -217,21 +215,28 @@ File *FileRef::create(FileName fileName, bool readAudioProperties,
 
   // Ok, this is really dumb for now, but it works for testing.
 
-  String s;
-
+  String ext;
+  {
 #ifdef _WIN32
-  s = (wcslen((const wchar_t *) fileName) > 0) ? String((const wchar_t *) fileName) : String((const char *) fileName);
+
+    String s = fileName.toString();
+
 #else
-  s = fileName;
-#endif
+
+    String s = fileName;
+
+ #endif
+
+    const int pos = s.rfind(".");
+    if(pos != -1)
+      ext = s.substr(pos + 1).upper();
+  }
 
   // If this list is updated, the method defaultFileExtensions() should also be
   // updated.  However at some point that list should be created at the same time
   // that a default file type resolver is created.
 
-  int pos = s.rfind(".");
-  if(pos != -1) {
-    String ext = s.substr(pos + 1).upper();
+  if(!ext.isEmpty()) {
     if(ext == "MP3")
       return new MPEG::File(fileName, readAudioProperties, audioPropertiesStyle);
     if(ext == "OGG")
@@ -252,6 +257,8 @@ File *FileRef::create(FileName fileName, bool readAudioProperties,
       return new WavPack::File(fileName, readAudioProperties, audioPropertiesStyle);
     if(ext == "SPX")
       return new Ogg::Speex::File(fileName, readAudioProperties, audioPropertiesStyle);
+    if(ext == "OPUS")
+      return new Ogg::Opus::File(fileName, readAudioProperties, audioPropertiesStyle);
     if(ext == "TTA")
       return new TrueAudio::File(fileName, readAudioProperties, audioPropertiesStyle);
     if(ext == "M4A" || ext == "M4R" || ext == "M4B" || ext == "M4P" || ext == "MP4" || ext == "3G2")
diff --git a/3rdparty/taglib/flac/flacfile.cpp b/3rdparty/taglib/flac/flacfile.cpp
index d30825675..5340250ea 100644
--- a/3rdparty/taglib/flac/flacfile.cpp
+++ b/3rdparty/taglib/flac/flacfile.cpp
@@ -70,7 +70,8 @@ public:
 
   ~FilePrivate()
   {
-    for(uint i = 0; i < blocks.size(); i++) {
+    uint size = blocks.size();
+    for(uint i = 0; i < size; i++) {
       delete blocks[i];
     }
     delete properties;
@@ -108,7 +109,8 @@ FLAC::File::File(FileName file, bool readProperties,
   TagLib::File(file)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 FLAC::File::File(FileName file, ID3v2::FrameFactory *frameFactory,
@@ -117,7 +119,8 @@ FLAC::File::File(FileName file, ID3v2::FrameFactory *frameFactory,
 {
   d = new FilePrivate;
   d->ID3v2FrameFactory = frameFactory;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 FLAC::File::File(IOStream *stream, ID3v2::FrameFactory *frameFactory,
@@ -126,7 +129,8 @@ FLAC::File::File(IOStream *stream, ID3v2::FrameFactory *frameFactory,
 {
   d = new FilePrivate;
   d->ID3v2FrameFactory = frameFactory;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 FLAC::File::~File()
@@ -164,14 +168,7 @@ void FLAC::File::removeUnsupportedProperties(const StringList &unsupported)
 
 PropertyMap FLAC::File::setProperties(const PropertyMap &properties)
 {
-  if(d->hasXiphComment)
-    return d->tag.access<Ogg::XiphComment>(FlacXiphIndex, false)->setProperties(properties);
-  else if(d->hasID3v2)
-    return d->tag.access<ID3v2::Tag>(FlacID3v2Index, false)->setProperties(properties);
-  else if(d->hasID3v1)
-    return d->tag.access<ID3v1::Tag>(FlacID3v1Index, false)->setProperties(properties);
-  else
-    return d->tag.access<Ogg::XiphComment>(FlacXiphIndex, true)->setProperties(properties);
+  return d->tag.access<Ogg::XiphComment>(FlacXiphIndex, true)->setProperties(properties);
 }
 
 FLAC::Properties *FLAC::File::audioProperties() const
@@ -402,7 +399,7 @@ void FLAC::File::scan()
 
   char blockType = header[0] & 0x7f;
   bool isLastBlock = (header[0] & 0x80) != 0;
-  uint length = header.mid(1, 3).toUInt();
+  uint length = header.toUInt(1U, 3U);
 
   // First block should be the stream_info metadata
 
@@ -431,10 +428,10 @@ void FLAC::File::scan()
     header = readBlock(4);
     blockType = header[0] & 0x7f;
     isLastBlock = (header[0] & 0x80) != 0;
-    length = header.mid(1, 3).toUInt();
+    length = header.toUInt(1U, 3U);
 
     ByteVector data = readBlock(length);
-    if(data.size() != length) {
+    if(data.size() != length || length == 0) {
       debug("FLAC::File::scan() -- FLAC stream corrupted");
       setValid(false);
       return;
@@ -564,3 +561,17 @@ void FLAC::File::removePictures()
   d->blocks = newBlocks;
 }
 
+bool FLAC::File::hasXiphComment() const
+{
+  return d->hasXiphComment;
+}
+
+bool FLAC::File::hasID3v1Tag() const
+{
+  return d->hasID3v1;
+}
+
+bool FLAC::File::hasID3v2Tag() const
+{
+  return d->hasID3v2;
+}
diff --git a/3rdparty/taglib/flac/flacfile.h b/3rdparty/taglib/flac/flacfile.h
index 716d4478c..0963f4af6 100644
--- a/3rdparty/taglib/flac/flacfile.h
+++ b/3rdparty/taglib/flac/flacfile.h
@@ -67,9 +67,10 @@ namespace TagLib {
     {
     public:
       /*!
-       * Contructs a FLAC file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs a FLAC file from \a file.  If \a readProperties is true the
+       * file's audio properties will also be read.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        *
        * \deprecated This constructor will be dropped in favor of the one below
        * in a future version.
@@ -78,12 +79,13 @@ namespace TagLib {
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs a FLAC file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an APE file from \a file.  If \a readProperties is true the
+       * file's audio properties will also be read.
        *
        * If this file contains and ID3v2 tag the frames will be created using
        * \a frameFactory.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       // BIC: merge with the above constructor
       File(FileName file, ID3v2::FrameFactory *frameFactory,
@@ -91,15 +93,16 @@ namespace TagLib {
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs a FLAC file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs a FLAC file from \a stream.  If \a readProperties is true the
+       * file's audio properties will also be read.
+       *
+       * \note TagLib will *not* take ownership of the stream, the caller is
+       * responsible for deleting it after the File object.
        *
        * If this file contains and ID3v2 tag the frames will be created using
        * \a frameFactory.
        *
-       * \note TagLib will *not* take ownership of the stream, the caller is
-       * responsible for deleting it after the File object.
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       // BIC: merge with the above constructor
       File(IOStream *stream, ID3v2::FrameFactory *frameFactory,
@@ -133,8 +136,10 @@ namespace TagLib {
 
       /*!
        * Implements the unified property interface -- import function.
-       * As with the export, only one tag is taken into account. If the file
-       * has no tag at all, a XiphComment will be created.
+       * This always creates a Xiph comment, if none exists. The return value
+       * relates to the Xiph comment only.
+       * Ignores any changes to ID3v1 or ID3v2 comments since they are not allowed
+       * in the FLAC specification.
        */
       PropertyMap setProperties(const PropertyMap &);
 
@@ -156,39 +161,57 @@ namespace TagLib {
       /*!
        * Returns a pointer to the ID3v2 tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
+       * If \a create is false (the default) this returns a null pointer
        * if there is no valid ID3v2 tag.  If \a create is true it will create
-       * an ID3v2 tag if one does not exist.
+       * an ID3v2 tag if one does not exist and returns a valid pointer.
        *
-       * \note The Tag <b>is still</b> owned by the FLAC::File and should not be
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an ID3v2 tag.  Use hasID3v2Tag() to check if the file 
+       * on disk actually has an ID3v2 tag.
+       *
+       * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasID3v2Tag()
        */
       ID3v2::Tag *ID3v2Tag(bool create = false);
 
       /*!
        * Returns a pointer to the ID3v1 tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
-       * if there is no valid ID3v1 tag.  If \a create is true it will create
-       * an ID3v1 tag if one does not exist.
+       * If \a create is false (the default) this returns a null pointer
+       * if there is no valid APE tag.  If \a create is true it will create
+       * an APE tag if one does not exist and returns a valid pointer.
        *
-       * \note The Tag <b>is still</b> owned by the FLAC::File and should not be
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an ID3v1 tag.  Use hasID3v1Tag() to check if the file 
+       * on disk actually has an ID3v1 tag.
+       *
+       * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasID3v1Tag()
        */
       ID3v1::Tag *ID3v1Tag(bool create = false);
 
       /*!
        * Returns a pointer to the XiphComment for the file.
        *
-       * If \a create is false (the default) this will return a null pointer
+       * If \a create is false (the default) this returns a null pointer
        * if there is no valid XiphComment.  If \a create is true it will create
-       * a XiphComment if one does not exist.
+       * a XiphComment if one does not exist and returns a valid pointer.
        *
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has a XiphComment.  Use hasXiphComment() to check if the 
+       * file on disk actually has a XiphComment.
+       * 
        * \note The Tag <b>is still</b> owned by the FLAC::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasXiphComment()
        */
       Ogg::XiphComment *xiphComment(bool create = false);
 
@@ -241,6 +264,27 @@ namespace TagLib {
        */
       void addPicture(Picture *picture);
 
+      /*!
+       * Returns whether or not the file on disk actually has a XiphComment.
+       *
+       * \see xiphComment()
+       */
+      bool hasXiphComment() const;
+
+      /*!
+       * Returns whether or not the file on disk actually has an ID3v1 tag.
+       *
+       * \see ID3v1Tag()
+       */
+      bool hasID3v1Tag() const;
+
+      /*!
+       * Returns whether or not the file on disk actually has an ID3v2 tag.
+       *
+       * \see ID3v2Tag()
+       */
+      bool hasID3v2Tag() const;
+
     private:
       File(const File &);
       File &operator=(const File &);
diff --git a/3rdparty/taglib/flac/flacmetadatablock.cpp b/3rdparty/taglib/flac/flacmetadatablock.cpp
index 7d161c27d..17ab05f37 100644
--- a/3rdparty/taglib/flac/flacmetadatablock.cpp
+++ b/3rdparty/taglib/flac/flacmetadatablock.cpp
@@ -23,10 +23,6 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <taglib.h>
 #include <tdebug.h>
 #include "flacmetadatablock.h"
diff --git a/3rdparty/taglib/flac/flacpicture.cpp b/3rdparty/taglib/flac/flacpicture.cpp
index 36019248a..a2a9000b6 100644
--- a/3rdparty/taglib/flac/flacpicture.cpp
+++ b/3rdparty/taglib/flac/flacpicture.cpp
@@ -23,10 +23,6 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <taglib.h>
 #include <tdebug.h>
 #include "flacpicture.h"
@@ -82,10 +78,10 @@ bool FLAC::Picture::parse(const ByteVector &data)
     return false;
   }
 
-  int pos = 0;
-  d->type = FLAC::Picture::Type(data.mid(pos, 4).toUInt());
+  uint pos = 0;
+  d->type = FLAC::Picture::Type(data.toUInt(pos));
   pos += 4;
-  uint mimeTypeLength = data.mid(pos, 4).toUInt();
+  uint mimeTypeLength = data.toUInt(pos);
   pos += 4;
   if(pos + mimeTypeLength + 24 > data.size()) {
     debug("Invalid picture block.");
@@ -93,7 +89,7 @@ bool FLAC::Picture::parse(const ByteVector &data)
   }
   d->mimeType = String(data.mid(pos, mimeTypeLength), String::UTF8);
   pos += mimeTypeLength;
-  uint descriptionLength = data.mid(pos, 4).toUInt();
+  uint descriptionLength = data.toUInt(pos);
   pos += 4;
   if(pos + descriptionLength + 20 > data.size()) {
     debug("Invalid picture block.");
@@ -101,15 +97,15 @@ bool FLAC::Picture::parse(const ByteVector &data)
   }
   d->description = String(data.mid(pos, descriptionLength), String::UTF8);
   pos += descriptionLength;
-  d->width = data.mid(pos, 4).toUInt();
+  d->width = data.toUInt(pos);
   pos += 4;
-  d->height = data.mid(pos, 4).toUInt();
+  d->height = data.toUInt(pos);
   pos += 4;
-  d->colorDepth = data.mid(pos, 4).toUInt();
+  d->colorDepth = data.toUInt(pos);
   pos += 4;
-  d->numColors = data.mid(pos, 4).toUInt();
+  d->numColors = data.toUInt(pos);
   pos += 4;
-  uint dataLength = data.mid(pos, 4).toUInt();
+  uint dataLength = data.toUInt(pos);
   pos += 4;
   if(pos + dataLength > data.size()) {
     debug("Invalid picture block.");
diff --git a/3rdparty/taglib/flac/flacproperties.cpp b/3rdparty/taglib/flac/flacproperties.cpp
index 8bdc5d8dc..e591193ef 100644
--- a/3rdparty/taglib/flac/flacproperties.cpp
+++ b/3rdparty/taglib/flac/flacproperties.cpp
@@ -124,7 +124,7 @@ void FLAC::Properties::read()
     return;
   }
 
-  int pos = 0;
+  uint pos = 0;
 
   // Minimum block size (in samples)
   pos += 2;
@@ -138,7 +138,7 @@ void FLAC::Properties::read()
   // Maximum frame size (in bytes)
   pos += 3;
 
-  uint flags = d->data.mid(pos, 4).toUInt(true);
+  uint flags = d->data.toUInt(pos, true);
   pos += 4;
 
   d->sampleRate = flags >> 12;
@@ -149,7 +149,7 @@ void FLAC::Properties::read()
   // stream length in samples. (Audio files measured in days)
 
   unsigned long long hi = flags & 0xf;
-  unsigned long long lo = d->data.mid(pos, 4).toUInt(true);
+  unsigned long long lo = d->data.toUInt(pos, true);
   pos += 4;
 
   d->sampleFrames = (hi << 32) | lo;
diff --git a/3rdparty/taglib/flac/flacunknownmetadatablock.cpp b/3rdparty/taglib/flac/flacunknownmetadatablock.cpp
index 1265affbb..dcd5d6515 100644
--- a/3rdparty/taglib/flac/flacunknownmetadatablock.cpp
+++ b/3rdparty/taglib/flac/flacunknownmetadatablock.cpp
@@ -23,10 +23,6 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <taglib.h>
 #include <tdebug.h>
 #include <tstring.h>
diff --git a/3rdparty/taglib/it/itfile.cpp b/3rdparty/taglib/it/itfile.cpp
index 4e049518e..ad5cf0b85 100644
--- a/3rdparty/taglib/it/itfile.cpp
+++ b/3rdparty/taglib/it/itfile.cpp
@@ -45,7 +45,8 @@ IT::File::File(FileName file, bool readProperties,
   Mod::FileBase(file),
   d(new FilePrivate(propertiesStyle))
 {
-  read(readProperties);
+  if(isOpen())
+    read(readProperties);
 }
 
 IT::File::File(IOStream *stream, bool readProperties,
@@ -53,7 +54,8 @@ IT::File::File(IOStream *stream, bool readProperties,
   Mod::FileBase(stream),
   d(new FilePrivate(propertiesStyle))
 {
-  read(readProperties);
+  if(isOpen())
+    read(readProperties);
 }
 
 IT::File::~File()
diff --git a/3rdparty/taglib/it/itfile.h b/3rdparty/taglib/it/itfile.h
index 5584b7cd9..19327dc65 100644
--- a/3rdparty/taglib/it/itfile.h
+++ b/3rdparty/taglib/it/itfile.h
@@ -36,23 +36,27 @@ namespace TagLib {
     class TAGLIB_EXPORT File : public Mod::FileBase {
       public:
         /*!
-         * Contructs a Impulse Tracker file from \a file. If \a readProperties
-         * is true the file's audio properties will also be read using
-         * \a propertiesStyle. If false, \a propertiesStyle is ignored.
+         * Constructs a Impulse Tracker file from \a file.
+         *
+         * \note In the current implementation, both \a readProperties and
+         * \a propertiesStyle are ignored.  The audio properties are always
+         * read.
          */
         File(FileName file, bool readProperties = true,
              AudioProperties::ReadStyle propertiesStyle =
              AudioProperties::Average);
 
         /*!
-         * Contructs a Impulse Tracker file from \a stream. If \a readProperties
-         * is true the file's audio properties will also be read using
-         * \a propertiesStyle. If false, \a propertiesStyle is ignored.
+         * Constructs a Impulse Tracker file from \a stream.
+         *
+         * \note In the current implementation, both \a readProperties and
+         * \a propertiesStyle are ignored.  The audio properties are always
+         * read.
          *
          * \note TagLib will *not* take ownership of the stream, the caller is
          * responsible for deleting it after the File object.
          */
-        File(IOStream *stram, bool readProperties = true,
+        File(IOStream *stream, bool readProperties = true,
              AudioProperties::ReadStyle propertiesStyle =
              AudioProperties::Average);
 
diff --git a/3rdparty/taglib/mod/modfile.cpp b/3rdparty/taglib/mod/modfile.cpp
index 25fc8715a..ce974c161 100644
--- a/3rdparty/taglib/mod/modfile.cpp
+++ b/3rdparty/taglib/mod/modfile.cpp
@@ -45,7 +45,8 @@ Mod::File::File(FileName file, bool readProperties,
   Mod::FileBase(file),
   d(new FilePrivate(propertiesStyle))
 {
-  read(readProperties);
+  if(isOpen())
+    read(readProperties);
 }
 
 Mod::File::File(IOStream *stream, bool readProperties,
@@ -53,7 +54,8 @@ Mod::File::File(IOStream *stream, bool readProperties,
   Mod::FileBase(stream),
   d(new FilePrivate(propertiesStyle))
 {
-  read(readProperties);
+  if(isOpen())
+    read(readProperties);
 }
 
 Mod::File::~File()
diff --git a/3rdparty/taglib/mod/modfile.h b/3rdparty/taglib/mod/modfile.h
index ad1e43b88..c45ede242 100644
--- a/3rdparty/taglib/mod/modfile.h
+++ b/3rdparty/taglib/mod/modfile.h
@@ -37,18 +37,22 @@ namespace TagLib {
     {
     public:
       /*!
-       * Contructs a Protracker file from \a file. If \a readProperties
-       * is true the file's audio properties will also be read using
-       * \a propertiesStyle. If false, \a propertiesStyle is ignored.
+       * Constructs a Protracker file from \a file.
+       *
+       * \note In the current implementation, both \a readProperties and
+       * \a propertiesStyle are ignored.  The audio properties are always
+       * read.
        */
       File(FileName file, bool readProperties = true,
            AudioProperties::ReadStyle propertiesStyle =
            AudioProperties::Average);
 
       /*!
-       * Contructs a Protracker file from \a stream. If \a readProperties
-       * is true the file's audio properties will also be read using
-       * \a propertiesStyle. If false, \a propertiesStyle is ignored.
+       * Constructs a Protracker file from \a stream.
+       *
+       * \note In the current implementation, both \a readProperties and
+       * \a propertiesStyle are ignored.  The audio properties are always
+       * read.
        *
        * \note TagLib will *not* take ownership of the stream, the caller is
        * responsible for deleting it after the File object.
diff --git a/3rdparty/taglib/mp4/mp4atom.cpp b/3rdparty/taglib/mp4/mp4atom.cpp
index 1681ec805..7b87a4796 100644
--- a/3rdparty/taglib/mp4/mp4atom.cpp
+++ b/3rdparty/taglib/mp4/mp4atom.cpp
@@ -23,10 +23,6 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <tdebug.h>
 #include <tstring.h>
 #include "mp4atom.h"
@@ -52,10 +48,10 @@ MP4::Atom::Atom(File *file)
     return;
   }
 
-  length = header.mid(0, 4).toUInt();
+  length = header.toUInt();
 
   if (length == 1) {
-    long long longLength = file->readBlock(8).toLongLong();
+    const long long longLength = file->readBlock(8).toLongLong();
     if (longLength >= 8 && longLength <= 0xFFFFFFFF) {
         // The atom has a 64-bit length, but it's actually a 32-bit value
         length = (long)longLength;
diff --git a/3rdparty/taglib/mp4/mp4coverart.cpp b/3rdparty/taglib/mp4/mp4coverart.cpp
index 928e3c4aa..2746469d3 100644
--- a/3rdparty/taglib/mp4/mp4coverart.cpp
+++ b/3rdparty/taglib/mp4/mp4coverart.cpp
@@ -23,12 +23,9 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <taglib.h>
 #include <tdebug.h>
+#include "trefcounter.h"
 #include "mp4coverart.h"
 
 using namespace TagLib;
diff --git a/3rdparty/taglib/mp4/mp4coverart.h b/3rdparty/taglib/mp4/mp4coverart.h
index 804ab7852..64115b458 100644
--- a/3rdparty/taglib/mp4/mp4coverart.h
+++ b/3rdparty/taglib/mp4/mp4coverart.h
@@ -42,10 +42,11 @@ namespace TagLib {
        * This describes the image type.
        */
       enum Format {
-        JPEG = TypeJPEG,
-        PNG  = TypePNG,
-        BMP  = TypeBMP,
-        GIF  = TypeGIF
+        JPEG    = TypeJPEG,
+        PNG     = TypePNG,
+        BMP     = TypeBMP,
+        GIF     = TypeGIF,
+        Unknown = TypeImplicit,
       };
 
       CoverArt(Format format, const ByteVector &data);
diff --git a/3rdparty/taglib/mp4/mp4file.cpp b/3rdparty/taglib/mp4/mp4file.cpp
index 02185575d..aab1a2bea 100644
--- a/3rdparty/taglib/mp4/mp4file.cpp
+++ b/3rdparty/taglib/mp4/mp4file.cpp
@@ -23,12 +23,9 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <tdebug.h>
 #include <tstring.h>
+#include <tpropertymap.h>
 #include "mp4atom.h"
 #include "mp4tag.h"
 #include "mp4file.h"
@@ -67,14 +64,16 @@ MP4::File::File(FileName file, bool readProperties, AudioProperties::ReadStyle a
     : TagLib::File(file)
 {
   d = new FilePrivate;
-  read(readProperties, audioPropertiesStyle);
+  if(isOpen())
+    read(readProperties, audioPropertiesStyle);
 }
 
 MP4::File::File(IOStream *stream, bool readProperties, AudioProperties::ReadStyle audioPropertiesStyle)
     : TagLib::File(stream)
 {
   d = new FilePrivate;
-  read(readProperties, audioPropertiesStyle);
+  if(isOpen())
+    read(readProperties, audioPropertiesStyle);
 }
 
 MP4::File::~File()
@@ -88,6 +87,21 @@ MP4::File::tag() const
   return d->tag;
 }
 
+PropertyMap MP4::File::properties() const
+{
+  return d->tag->properties();
+}
+
+void MP4::File::removeUnsupportedProperties(const StringList &properties)
+{
+  d->tag->removeUnsupportedProperties(properties);
+}
+
+PropertyMap MP4::File::setProperties(const PropertyMap &properties)
+{
+  return d->tag->setProperties(properties);
+}
+
 MP4::Properties *
 MP4::File::audioProperties() const
 {
diff --git a/3rdparty/taglib/mp4/mp4file.h b/3rdparty/taglib/mp4/mp4file.h
index 2ed3bea58..39693d369 100644
--- a/3rdparty/taglib/mp4/mp4file.h
+++ b/3rdparty/taglib/mp4/mp4file.h
@@ -49,27 +49,25 @@ namespace TagLib {
     {
     public:
       /*!
-       * Contructs a MP4 file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an MP4 file from \a file.  If \a readProperties is true the
+       * file's audio properties will also be read.
        *
-       * \note In the current implementation, both \a readProperties and
-       * \a propertiesStyle are ignored.
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
-      File(FileName file, bool readProperties = true, Properties::ReadStyle audioPropertiesStyle = Properties::Average);
+      File(FileName file, bool readProperties = true, 
+           Properties::ReadStyle audioPropertiesStyle = Properties::Average);
 
       /*!
-       * Contructs a MP4 file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
-       *
-       * \note In the current implementation, both \a readProperties and
-       * \a propertiesStyle are ignored.
+       * Constructs an MP4 file from \a stream.  If \a readProperties is true the
+       * file's audio properties will also be read.
        *
        * \note TagLib will *not* take ownership of the stream, the caller is
        * responsible for deleting it after the File object.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
-      File(IOStream *stream, bool readProperties = true, Properties::ReadStyle audioPropertiesStyle = Properties::Average);
+      File(IOStream *stream, bool readProperties = true, 
+           Properties::ReadStyle audioPropertiesStyle = Properties::Average);
 
       /*!
        * Destroys this instance of the File.
@@ -88,6 +86,22 @@ namespace TagLib {
        */
       Tag *tag() const;
 
+      /*!
+       * Implements the unified property interface -- export function.
+       */
+      PropertyMap properties() const;
+
+      /*!
+       * Removes unsupported properties. Forwards to the actual Tag's
+       * removeUnsupportedProperties() function.
+       */
+      void removeUnsupportedProperties(const StringList &properties);
+
+      /*!
+       * Implements the unified property interface -- import function.
+       */
+      PropertyMap setProperties(const PropertyMap &);
+
       /*!
        * Returns the MP4 audio properties for this file.
        */
diff --git a/3rdparty/taglib/mp4/mp4item.cpp b/3rdparty/taglib/mp4/mp4item.cpp
index af2cc65c8..671f26b4d 100644
--- a/3rdparty/taglib/mp4/mp4item.cpp
+++ b/3rdparty/taglib/mp4/mp4item.cpp
@@ -23,12 +23,9 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <taglib.h>
 #include <tdebug.h>
+#include "trefcounter.h"
 #include "mp4item.h"
 
 using namespace TagLib;
diff --git a/3rdparty/taglib/mp4/mp4properties.cpp b/3rdparty/taglib/mp4/mp4properties.cpp
index d2d04167c..058cc61cf 100644
--- a/3rdparty/taglib/mp4/mp4properties.cpp
+++ b/3rdparty/taglib/mp4/mp4properties.cpp
@@ -23,10 +23,6 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <tdebug.h>
 #include <tstring.h>
 #include "mp4file.h"
@@ -96,8 +92,8 @@ MP4::Properties::Properties(File *file, MP4::Atoms *atoms, ReadStyle style)
       debug("MP4: Atom 'trak.mdia.mdhd' is smaller than expected");
       return;
     }
-    long long unit = data.mid(28, 8).toLongLong();
-    long long length = data.mid(36, 8).toLongLong();
+    const long long unit   = data.toLongLong(28U);
+    const long long length = data.toLongLong(36U);
     d->length = unit ? int(length / unit) : 0;
   }
   else {
@@ -105,8 +101,8 @@ MP4::Properties::Properties(File *file, MP4::Atoms *atoms, ReadStyle style)
       debug("MP4: Atom 'trak.mdia.mdhd' is smaller than expected");
       return;
     }
-    unsigned int unit = data.mid(20, 4).toUInt();
-    unsigned int length = data.mid(24, 4).toUInt();
+    const unsigned int unit   = data.toUInt(20U);
+    const unsigned int length = data.toUInt(24U);
     d->length = unit ? length / unit : 0;
   }
 
@@ -118,11 +114,11 @@ MP4::Properties::Properties(File *file, MP4::Atoms *atoms, ReadStyle style)
   file->seek(atom->offset);
   data = file->readBlock(atom->length);
   if(data.mid(20, 4) == "mp4a") {
-    d->channels = data.mid(40, 2).toShort();
-    d->bitsPerSample = data.mid(42, 2).toShort();
-    d->sampleRate = data.mid(46, 4).toUInt();
+    d->channels      = data.toShort(40U);
+    d->bitsPerSample = data.toShort(42U);
+    d->sampleRate    = data.toUInt(46U);
     if(data.mid(56, 4) == "esds" && data[64] == 0x03) {
-      long pos = 65;
+      uint pos = 65;
       if(data.mid(pos, 3) == "\x80\x80\x80") {
         pos += 3;
       }
@@ -133,16 +129,16 @@ MP4::Properties::Properties(File *file, MP4::Atoms *atoms, ReadStyle style)
           pos += 3;
         }
         pos += 10;
-        d->bitrate = (data.mid(pos, 4).toUInt() + 500) / 1000;
+        d->bitrate = (data.toUInt(pos) + 500) / 1000;
       }
     }
   }
   else if (data.mid(20, 4) == "alac") {
     if (atom->length == 88 && data.mid(56, 4) == "alac") {
       d->bitsPerSample = data.at(69);
-      d->channels = data.at(73);
-      d->bitrate = data.mid(80, 4).toUInt() / 1000;
-      d->sampleRate = data.mid(84, 4).toUInt();
+      d->channels   = data.at(73);
+      d->bitrate    = data.toUInt(80U) / 1000;
+      d->sampleRate = data.toUInt(84U);
     }
   }
 
diff --git a/3rdparty/taglib/mp4/mp4tag.cpp b/3rdparty/taglib/mp4/mp4tag.cpp
index fc75fb456..bbe3301d9 100644
--- a/3rdparty/taglib/mp4/mp4tag.cpp
+++ b/3rdparty/taglib/mp4/mp4tag.cpp
@@ -23,12 +23,9 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifdef HAVE_CONFIG_H
-#include <config.h>
-#endif
-
 #include <tdebug.h>
 #include <tstring.h>
+#include <tpropertymap.h>
 #include "mp4atom.h"
 #include "mp4tag.h"
 #include "id3v1genres.h"
@@ -113,9 +110,9 @@ MP4::Tag::parseData2(MP4::Atom *atom, TagLib::File *file, int expectedFlags, boo
   int i = 0;
   unsigned int pos = 0;
   while(pos < data.size()) {
-    int length = data.mid(pos, 4).toUInt();
+    const int length = static_cast<int>(data.toUInt(pos));
     ByteVector name = data.mid(pos + 4, 4);
-    int flags = data.mid(pos + 8, 4).toUInt();
+    const int flags = static_cast<int>(data.toUInt(pos + 8));
     if(freeForm && i < 2) {
       if(i == 0 && name != "mean") {
         debug("MP4: Unexpected atom \"" + name + "\", expecting \"mean\"");
@@ -158,7 +155,7 @@ MP4::Tag::parseInt(MP4::Atom *atom, TagLib::File *file)
 {
   ByteVectorList data = parseData(atom, file);
   if(data.size()) {
-    d->items.insert(atom->name, (int)data[0].toShort());
+    addItem(atom->name, (int)data[0].toShort());
   }
 }
 
@@ -167,7 +164,7 @@ MP4::Tag::parseUInt(MP4::Atom *atom, TagLib::File *file)
 {
   ByteVectorList data = parseData(atom, file);
   if(data.size()) {
-    d->items.insert(atom->name, data[0].toUInt());
+    addItem(atom->name, data[0].toUInt());
   }
 }
 
@@ -176,7 +173,7 @@ MP4::Tag::parseLongLong(MP4::Atom *atom, TagLib::File *file)
 {
   ByteVectorList data = parseData(atom, file);
   if(data.size()) {
-    d->items.insert(atom->name, data[0].toLongLong());
+    addItem(atom->name, data[0].toLongLong());
   }
 }
 
@@ -185,7 +182,7 @@ MP4::Tag::parseByte(MP4::Atom *atom, TagLib::File *file)
 {
   ByteVectorList data = parseData(atom, file);
   if(data.size()) {
-    d->items.insert(atom->name, (uchar)data[0].at(0));
+    addItem(atom->name, (uchar)data[0].at(0));
   }
 }
 
@@ -195,8 +192,8 @@ MP4::Tag::parseGnre(MP4::Atom *atom, TagLib::File *file)
   ByteVectorList data = parseData(atom, file);
   if(data.size()) {
     int idx = (int)data[0].toShort();
-    if(!d->items.contains("\251gen") && idx > 0) {
-      d->items.insert("\251gen", StringList(ID3v1::genre(idx - 1)));
+    if(idx > 0) {
+      addItem("\251gen", StringList(ID3v1::genre(idx - 1)));
     }
   }
 }
@@ -206,9 +203,9 @@ MP4::Tag::parseIntPair(MP4::Atom *atom, TagLib::File *file)
 {
   ByteVectorList data = parseData(atom, file);
   if(data.size()) {
-    int a = data[0].mid(2, 2).toShort();
-    int b = data[0].mid(4, 2).toShort();
-    d->items.insert(atom->name, MP4::Item(a, b));
+    const int a = data[0].toShort(2U);
+    const int b = data[0].toShort(4U);
+    addItem(atom->name, MP4::Item(a, b));
   }
 }
 
@@ -218,7 +215,7 @@ MP4::Tag::parseBool(MP4::Atom *atom, TagLib::File *file)
   ByteVectorList data = parseData(atom, file);
   if(data.size()) {
     bool value = data[0].size() ? data[0][0] != '\0' : false;
-    d->items.insert(atom->name, value);
+    addItem(atom->name, value);
   }
 }
 
@@ -231,7 +228,7 @@ MP4::Tag::parseText(MP4::Atom *atom, TagLib::File *file, int expectedFlags)
     for(unsigned int i = 0; i < data.size(); i++) {
       value.append(String(data[i], String::UTF8));
     }
-    d->items.insert(atom->name, value);
+    addItem(atom->name, value);
   }
 }
 
@@ -255,7 +252,7 @@ MP4::Tag::parseFreeForm(MP4::Atom *atom, TagLib::File *file)
       }
       Item item(value);
       item.setAtomDataType(type);
-      d->items.insert(name, item);
+      addItem(name, item);
     }
     else {
       ByteVectorList value;
@@ -264,7 +261,7 @@ MP4::Tag::parseFreeForm(MP4::Atom *atom, TagLib::File *file)
       }
       Item item(value);
       item.setAtomDataType(type);
-      d->items.insert(name, item);
+      addItem(name, item);
     }
   }
 }
@@ -276,21 +273,24 @@ MP4::Tag::parseCovr(MP4::Atom *atom, TagLib::File *file)
   ByteVector data = file->readBlock(atom->length - 8);
   unsigned int pos = 0;
   while(pos < data.size()) {
-    int length = data.mid(pos, 4).toUInt();
+    const int length = static_cast<int>(data.toUInt(pos));
     ByteVector name = data.mid(pos + 4, 4);
-    int flags = data.mid(pos + 8, 4).toUInt();
+    const int flags = static_cast<int>(data.toUInt(pos + 8));
     if(name != "data") {
       debug("MP4: Unexpected atom \"" + name + "\", expecting \"data\"");
       break;
     }
-    if(flags == TypeJPEG || flags == TypePNG || flags == TypeBMP || flags == TypeGIF) {
+    if(flags == TypeJPEG || flags == TypePNG || flags == TypeBMP || flags == TypeGIF || flags == TypeImplicit) {
       value.append(MP4::CoverArt(MP4::CoverArt::Format(flags),
                                  data.mid(pos + 16, length - 16)));
     }
+    else {
+      debug("MP4: Unknown covr format " + String::number(flags));
+    }
     pos += length;
   }
   if(value.size() > 0)
-    d->items.insert(atom->name, value);
+    addItem(atom->name, value);
 }
 
 ByteVector
@@ -526,11 +526,11 @@ MP4::Tag::updateOffsets(long delta, long offset)
       }
       d->file->seek(atom->offset + 12);
       ByteVector data = d->file->readBlock(atom->length - 12);
-      unsigned int count = data.mid(0, 4).toUInt();
+      unsigned int count = data.toUInt();
       d->file->seek(atom->offset + 16);
-      int pos = 4;
+      uint pos = 4;
       while(count--) {
-        long o = data.mid(pos, 4).toUInt();
+        long o = static_cast<long>(data.toUInt(pos));
         if(o > offset) {
           o += delta;
         }
@@ -547,11 +547,11 @@ MP4::Tag::updateOffsets(long delta, long offset)
       }
       d->file->seek(atom->offset + 12);
       ByteVector data = d->file->readBlock(atom->length - 12);
-      unsigned int count = data.mid(0, 4).toUInt();
+      unsigned int count = data.toUInt();
       d->file->seek(atom->offset + 16);
-      int pos = 4;
+      uint pos = 4;
       while(count--) {
-        long long o = data.mid(pos, 8).toLongLong();
+        long long o = data.toLongLong(pos);
         if(o > offset) {
           o += delta;
         }
@@ -570,10 +570,10 @@ MP4::Tag::updateOffsets(long delta, long offset)
         atom->offset += delta;
       }
       d->file->seek(atom->offset + 9);
-      ByteVector data = d->file->readBlock(atom->offset - 9);
-      unsigned int flags = (ByteVector(1, '\0') + data.mid(0, 3)).toUInt();
+      ByteVector data = d->file->readBlock(atom->length - 9);
+      const unsigned int flags = data.toUInt(0, 3, true);
       if(flags & 1) {
-        long long o = data.mid(7, 8).toLongLong();
+        long long o = data.toLongLong(7U);
         if(o > offset) {
           o += delta;
         }
@@ -756,3 +756,162 @@ MP4::Tag::itemListMap()
   return d->items;
 }
 
+static const char *keyTranslation[][2] = {
+  { "\251nam", "TITLE" },
+  { "\251ART", "ARTIST" },
+  { "\251alb", "ALBUM" },
+  { "\251cmt", "COMMENT" },
+  { "\251gen", "GENRE" },
+  { "\251day", "DATE" },
+  { "\251wrt", "COMPOSER" },
+  { "\251grp", "GROUPING" },
+  { "trkn", "TRACKNUMBER" },
+  { "disk", "DISCNUMBER" },
+  { "cpil", "COMPILATION" },
+  { "tmpo", "BPM" },
+  { "cprt", "COPYRIGHT" },
+  { "\251lyr", "LYRICS" },
+  { "\251too", "ENCODEDBY" },
+  { "soal", "ALBUMSORT" },
+  { "soaa", "ALBUMARTISTSORT" },
+  { "soar", "ARTISTSORT" },
+  { "sonm", "TITLESORT" },
+  { "soco", "COMPOSERSORT" },
+  { "sosn", "SHOWSORT" },
+  { "----:com.apple.iTunes:MusicBrainz Track Id", "MUSICBRAINZ_TRACKID" },
+  { "----:com.apple.iTunes:MusicBrainz Artist Id", "MUSICBRAINZ_ARTISTID" },
+  { "----:com.apple.iTunes:MusicBrainz Album Id", "MUSICBRAINZ_ALBUMID" },
+  { "----:com.apple.iTunes:MusicBrainz Album Artist Id", "MUSICBRAINZ_ALBUMARTISTID" },
+  { "----:com.apple.iTunes:MusicBrainz Release Group Id", "MUSICBRAINZ_RELEASEGROUPID" },
+  { "----:com.apple.iTunes:MusicBrainz Work Id", "MUSICBRAINZ_WORKID" },
+  { "----:com.apple.iTunes:ASIN", "ASIN" },
+  { "----:com.apple.iTunes:LABEL", "LABEL" },
+  { "----:com.apple.iTunes:LYRICIST", "LYRICIST" },
+  { "----:com.apple.iTunes:CONDUCTOR", "CONDUCTOR" },
+  { "----:com.apple.iTunes:REMIXER", "REMIXER" },
+  { "----:com.apple.iTunes:ENGINEER", "ENGINEER" },
+  { "----:com.apple.iTunes:PRODUCER", "PRODUCER" },
+  { "----:com.apple.iTunes:DJMIXER", "DJMIXER" },
+  { "----:com.apple.iTunes:MIXER", "MIXER" },
+  { "----:com.apple.iTunes:SUBTITLE", "SUBTITLE" },
+  { "----:com.apple.iTunes:DISCSUBTITLE", "DISCSUBTITLE" },
+  { "----:com.apple.iTunes:MOOD", "MOOD" },
+  { "----:com.apple.iTunes:ISRC", "ISRC" },
+  { "----:com.apple.iTunes:CATALOGNUMBER", "CATALOGNUMBER" },
+  { "----:com.apple.iTunes:BARCODE", "BARCODE" },
+  { "----:com.apple.iTunes:SCRIPT", "SCRIPT" },
+  { "----:com.apple.iTunes:LANGUAGE", "LANGUAGE" },
+  { "----:com.apple.iTunes:LICENSE", "LICENSE" },
+  { "----:com.apple.iTunes:MEDIA", "MEDIA" },
+};
+
+PropertyMap MP4::Tag::properties() const
+{
+  static Map<String, String> keyMap;
+  if(keyMap.isEmpty()) {
+    int numKeys = sizeof(keyTranslation) / sizeof(keyTranslation[0]);
+    for(int i = 0; i < numKeys; i++) {
+      keyMap[keyTranslation[i][0]] = keyTranslation[i][1];
+    }
+  }
+
+  PropertyMap props;
+  MP4::ItemListMap::ConstIterator it = d->items.begin();
+  for(; it != d->items.end(); ++it) {
+    if(keyMap.contains(it->first)) {
+      String key = keyMap[it->first];
+      if(key == "TRACKNUMBER" || key == "DISCNUMBER") {
+        MP4::Item::IntPair ip = it->second.toIntPair();
+        String value = String::number(ip.first);
+        if(ip.second) {
+          value += "/" + String::number(ip.second);
+        }
+        props[key] = value;
+      }
+      else if(key == "BPM") {
+        props[key] = String::number(it->second.toInt());
+      }
+      else if(key == "COMPILATION") {
+        props[key] = String::number(it->second.toBool());
+      }
+      else {
+        props[key] = it->second.toStringList();
+      }
+    }
+    else {
+      props.unsupportedData().append(it->first);
+    }
+  }
+  return props;
+}
+
+void MP4::Tag::removeUnsupportedProperties(const StringList &props)
+{
+  StringList::ConstIterator it = props.begin();
+  for(; it != props.end(); ++it)
+    d->items.erase(*it);
+}
+
+PropertyMap MP4::Tag::setProperties(const PropertyMap &props)
+{
+  static Map<String, String> reverseKeyMap;
+  if(reverseKeyMap.isEmpty()) {
+    int numKeys = sizeof(keyTranslation) / sizeof(keyTranslation[0]);
+    for(int i = 0; i < numKeys; i++) {
+      reverseKeyMap[keyTranslation[i][1]] = keyTranslation[i][0];
+    }
+  }
+
+  PropertyMap origProps = properties();
+  PropertyMap::ConstIterator it = origProps.begin();
+  for(; it != origProps.end(); ++it) {
+    if(!props.contains(it->first) || props[it->first].isEmpty()) {
+      d->items.erase(reverseKeyMap[it->first]);
+    }
+  }
+
+  PropertyMap ignoredProps;
+  it = props.begin();
+  for(; it != props.end(); ++it) {
+    if(reverseKeyMap.contains(it->first)) {
+      String name = reverseKeyMap[it->first];
+      if(it->first == "TRACKNUMBER" || it->first == "DISCNUMBER") {
+        int first = 0, second = 0;
+        StringList parts = StringList::split(it->second.front(), "/");
+        if(parts.size() > 0) {
+          first = parts[0].toInt();
+          if(parts.size() > 1) {
+            second = parts[1].toInt();
+          }
+          d->items[name] = MP4::Item(first, second);
+        }
+      }
+      else if(it->first == "BPM") {
+        int value = it->second.front().toInt();
+        d->items[name] = MP4::Item(value);
+      }
+      else if(it->first == "COMPILATION") {
+        bool value = (it->second.front().toInt() != 0);
+        d->items[name] = MP4::Item(value);
+      }
+      else {
+        d->items[name] = it->second;
+      }
+    }
+    else {
+      ignoredProps.insert(it->first, it->second);
+    }
+  }
+
+  return ignoredProps;
+}
+
+void MP4::Tag::addItem(const String &name, const Item &value)
+{
+  if(!d->items.contains(name)) {
+    d->items.insert(name, value);
+  }
+  else {
+    debug("MP4: Ignoring duplicate atom \"" + name + "\"");
+  }
+}
diff --git a/3rdparty/taglib/mp4/mp4tag.h b/3rdparty/taglib/mp4/mp4tag.h
index b5ea6ebbc..48d71fcb9 100644
--- a/3rdparty/taglib/mp4/mp4tag.h
+++ b/3rdparty/taglib/mp4/mp4tag.h
@@ -67,6 +67,10 @@ namespace TagLib {
 
         ItemListMap &itemListMap();
 
+        PropertyMap properties() const;
+        void removeUnsupportedProperties(const StringList& properties);
+        PropertyMap setProperties(const PropertyMap &properties);
+
     private:
         AtomDataList parseData2(Atom *atom, TagLib::File *file, int expectedFlags = -1, bool freeForm = false);
         TagLib::ByteVectorList parseData(Atom *atom, TagLib::File *file, int expectedFlags = -1, bool freeForm = false);
@@ -101,6 +105,8 @@ namespace TagLib {
         void saveNew(TagLib::ByteVector &data);
         void saveExisting(TagLib::ByteVector &data, AtomList &path);
 
+        void addItem(const String &name, const Item &value);
+
         class TagPrivate;
         TagPrivate *d;
     };
diff --git a/3rdparty/taglib/mpc/mpcfile.cpp b/3rdparty/taglib/mpc/mpcfile.cpp
index 519a04675..979c035e9 100644
--- a/3rdparty/taglib/mpc/mpcfile.cpp
+++ b/3rdparty/taglib/mpc/mpcfile.cpp
@@ -39,7 +39,7 @@ using namespace TagLib;
 
 namespace
 {
-  enum { MPCAPEIndex, MPCID3v1Index };
+  enum { MPCAPEIndex = 0, MPCID3v1Index = 1 };
 }
 
 class MPC::File::FilePrivate
@@ -94,14 +94,16 @@ MPC::File::File(FileName file, bool readProperties,
                 Properties::ReadStyle propertiesStyle) : TagLib::File(file)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 MPC::File::File(IOStream *stream, bool readProperties,
                 Properties::ReadStyle propertiesStyle) : TagLib::File(stream)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 MPC::File::~File()
@@ -133,15 +135,11 @@ void MPC::File::removeUnsupportedProperties(const StringList &properties)
 
 PropertyMap MPC::File::setProperties(const PropertyMap &properties)
 {
-  if(d->hasAPE)
-    return d->tag.access<APE::Tag>(MPCAPEIndex, false)->setProperties(properties);
-  else if(d->hasID3v1)
-    return d->tag.access<ID3v1::Tag>(MPCID3v1Index, false)->setProperties(properties);
-  else
-    return d->tag.access<APE::Tag>(APE, true)->setProperties(properties);
+  if(d->hasID3v1)
+    d->tag.access<APE::Tag>(MPCID3v1Index, false)->setProperties(properties);
+  return d->tag.access<APE::Tag>(MPCAPEIndex, true)->setProperties(properties);
 }
 
-
 MPC::Properties *MPC::File::audioProperties() const
 {
   return d->properties;
@@ -258,6 +256,15 @@ void MPC::File::remove(int tags)
   strip(tags);
 }
 
+bool MPC::File::hasID3v1Tag() const
+{
+  return d->hasID3v1;
+}
+
+bool MPC::File::hasAPETag() const
+{
+  return d->hasAPE;
+}
 
 ////////////////////////////////////////////////////////////////////////////////
 // private members
diff --git a/3rdparty/taglib/mpc/mpcfile.h b/3rdparty/taglib/mpc/mpcfile.h
index 167b768e1..1eef81032 100644
--- a/3rdparty/taglib/mpc/mpcfile.h
+++ b/3rdparty/taglib/mpc/mpcfile.h
@@ -84,20 +84,22 @@ namespace TagLib {
       };
 
       /*!
-       * Contructs an MPC file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an MPC file from \a file.  If \a readProperties is true the
+       * file's audio properties will also be read.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(FileName file, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs an MPC file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an MPC file from \a stream.  If \a readProperties is true the
+       * file's audio properties will also be read.
        *
        * \note TagLib will *not* take ownership of the stream, the caller is
        * responsible for deleting it after the File object.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(IOStream *stream, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
@@ -124,8 +126,8 @@ namespace TagLib {
 
       /*!
        * Implements the unified property interface -- import function.
-       * As with the export, only one tag is taken into account. If the file
-       * has no tag at all, an APE tag will be created.
+       * Affects only the APEv2 tag which will be created if necessary.
+       * If an ID3v1 tag exists, it will be updated as well.
        */
       PropertyMap setProperties(const PropertyMap &);
 
@@ -143,28 +145,39 @@ namespace TagLib {
       /*!
        * Returns a pointer to the ID3v1 tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
-       * if there is no valid ID3v1 tag.  If \a create is true it will create
-       * an ID3v1 tag if one does not exist. If there is already an APE tag, the
-       * new ID3v1 tag will be placed after it.
+       * If \a create is false (the default) this returns a null pointer
+       * if there is no valid APE tag.  If \a create is true it will create
+       * an APE tag if one does not exist and returns a valid pointer.
        *
-       * \note The Tag <b>is still</b> owned by the APE::File and should not be
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an ID3v1 tag.  Use hasID3v1Tag() to check if the file 
+       * on disk actually has an ID3v1 tag.
+       *
+       * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasID3v1Tag()
        */
       ID3v1::Tag *ID3v1Tag(bool create = false);
 
       /*!
        * Returns a pointer to the APE tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
+       * If \a create is false (the default) this may return a null pointer
        * if there is no valid APE tag.  If \a create is true it will create
-       * a APE tag if one does not exist. If there is already an ID3v1 tag, thes
-       * new APE tag will be placed before it.
+       * an APE tag if one does not exist and returns a valid pointer.  If 
+       * there already be an ID3v1 tag, the new APE tag will be placed before it.
        *
-       * \note The Tag <b>is still</b> owned by the APE::File and should not be
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an APE tag.  Use hasAPETag() to check if the file 
+       * on disk actually has an APE tag.
+       *
+       * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasAPETag()
        */
       APE::Tag *APETag(bool create = false);
 
@@ -185,6 +198,20 @@ namespace TagLib {
        */
       void remove(int tags = AllTags);
 
+      /*!
+       * Returns whether or not the file on disk actually has an ID3v1 tag.
+       *
+       * \see ID3v1Tag()
+       */
+      bool hasID3v1Tag() const;
+
+      /*!
+       * Returns whether or not the file on disk actually has an APE tag.
+       *
+       * \see APETag()
+       */
+      bool hasAPETag() const;
+
     private:
       File(const File &);
       File &operator=(const File &);
diff --git a/3rdparty/taglib/mpc/mpcproperties.cpp b/3rdparty/taglib/mpc/mpcproperties.cpp
index 23d419872..d406f8d73 100644
--- a/3rdparty/taglib/mpc/mpcproperties.cpp
+++ b/3rdparty/taglib/mpc/mpcproperties.cpp
@@ -207,7 +207,7 @@ void MPC::Properties::readSV8(File *file)
       d->sampleFrames = readSize(data.mid(pos), pos);
       ulong begSilence = readSize(data.mid(pos), pos);
 
-      std::bitset<16> flags(TAGLIB_CONSTRUCT_BITSET(data.mid(pos, 2).toUShort(true)));
+      std::bitset<16> flags(TAGLIB_CONSTRUCT_BITSET(data.toUShort(pos, true)));
       pos += 2;
 
       d->sampleRate = sftable[flags[15] * 4 + flags[14] * 2 + flags[13]];
@@ -228,10 +228,10 @@ void MPC::Properties::readSV8(File *file)
 
       int replayGainVersion = data[0];
       if(replayGainVersion == 1) {
-        d->trackGain = data.mid(1, 2).toUInt(true);
-        d->trackPeak = data.mid(3, 2).toUInt(true);
-        d->albumGain = data.mid(5, 2).toUInt(true);
-        d->albumPeak = data.mid(7, 2).toUInt(true);
+        d->trackGain = data.toShort(1, true);
+        d->trackPeak = data.toShort(3, true);
+        d->albumGain = data.toShort(5, true);
+        d->albumPeak = data.toShort(7, true);
       }
     }
 
@@ -252,18 +252,18 @@ void MPC::Properties::readSV7(const ByteVector &data)
     if(d->version < 7)
       return;
 
-    d->totalFrames = data.mid(4, 4).toUInt(false);
+    d->totalFrames = data.toUInt(4, false);
 
-    std::bitset<32> flags(TAGLIB_CONSTRUCT_BITSET(data.mid(8, 4).toUInt(false)));
+    std::bitset<32> flags(TAGLIB_CONSTRUCT_BITSET(data.toUInt(8, false)));
     d->sampleRate = sftable[flags[17] * 2 + flags[16]];
     d->channels = 2;
 
-    uint gapless = data.mid(5, 4).toUInt(false);
+    uint gapless = data.toUInt(5, false);
 
-    d->trackGain = data.mid(14,2).toShort(false);
-    d->trackPeak = data.mid(12,2).toUInt(false);
-    d->albumGain = data.mid(18,2).toShort(false);
-    d->albumPeak = data.mid(16,2).toUInt(false);
+    d->trackGain = data.toShort(14, false);
+    d->trackPeak = data.toShort(12, false);
+    d->albumGain = data.toShort(18, false);
+    d->albumPeak = data.toShort(16, false);
 
     // convert gain info
     if(d->trackGain != 0) {
@@ -293,7 +293,7 @@ void MPC::Properties::readSV7(const ByteVector &data)
       d->sampleFrames = d->totalFrames * 1152 - 576;
   }
   else {
-    uint headerData = data.mid(0, 4).toUInt(false);
+    uint headerData = data.toUInt(0, false);
 
     d->bitrate = (headerData >> 23) & 0x01ff;
     d->version = (headerData >> 11) & 0x03ff;
@@ -301,9 +301,9 @@ void MPC::Properties::readSV7(const ByteVector &data)
     d->channels = 2;
 
     if(d->version >= 5)
-      d->totalFrames = data.mid(4, 4).toUInt(false);
+      d->totalFrames = data.toUInt(4, false);
     else
-      d->totalFrames = data.mid(6, 2).toUInt(false);
+      d->totalFrames = data.toUShort(6, false);
 
     d->sampleFrames = d->totalFrames * 1152 - 576;
   }
diff --git a/3rdparty/taglib/mpeg/id3v1/id3v1tag.cpp b/3rdparty/taglib/mpeg/id3v1/id3v1tag.cpp
index e9eb47bf1..9fc8cfd7d 100644
--- a/3rdparty/taglib/mpeg/id3v1/id3v1tag.cpp
+++ b/3rdparty/taglib/mpeg/id3v1/id3v1tag.cpp
@@ -182,16 +182,26 @@ void ID3v1::Tag::setGenre(const String &s)
   d->genre = ID3v1::genreIndex(s);
 }
 
-void ID3v1::Tag::setYear(uint i)
+void ID3v1::Tag::setYear(TagLib::uint i)
 {
   d->year = i > 0 ? String::number(i) : String::null;
 }
 
-void ID3v1::Tag::setTrack(uint i)
+void ID3v1::Tag::setTrack(TagLib::uint i)
 {
   d->track = i < 256 ? i : 0;
 }
 
+TagLib::uint ID3v1::Tag::genreNumber() const
+{
+  return d->genre;
+}
+
+void ID3v1::Tag::setGenreNumber(TagLib::uint i)
+{
+  d->genre = i < 256 ? i : 255;
+}
+
 void ID3v1::Tag::setStringHandler(const StringHandler *handler)
 {
   if (handler)
diff --git a/3rdparty/taglib/mpeg/id3v1/id3v1tag.h b/3rdparty/taglib/mpeg/id3v1/id3v1tag.h
index 9332d4dce..7b26d023e 100644
--- a/3rdparty/taglib/mpeg/id3v1/id3v1tag.h
+++ b/3rdparty/taglib/mpeg/id3v1/id3v1tag.h
@@ -140,16 +140,31 @@ namespace TagLib {
       virtual String album() const;
       virtual String comment() const;
       virtual String genre() const;
-      virtual uint year() const;
-      virtual uint track() const;
+      virtual TagLib::uint year() const;
+      virtual TagLib::uint track() const;
 
       virtual void setTitle(const String &s);
       virtual void setArtist(const String &s);
       virtual void setAlbum(const String &s);
       virtual void setComment(const String &s);
       virtual void setGenre(const String &s);
-      virtual void setYear(uint i);
-      virtual void setTrack(uint i);
+      virtual void setYear(TagLib::uint i);
+      virtual void setTrack(TagLib::uint i);
+
+      /*!
+       * Returns the genre in number.
+       *
+       * /note Normally 255 indicates that this tag contains no genre.
+       */
+      TagLib::uint genreNumber() const;
+
+      /*!
+       * Sets the genre in number to \a i.
+       *
+       * /note Valid value is from 0 up to 255. Normally 255 indicates that
+       * this tag contains no genre.
+       */
+      void setGenreNumber(TagLib::uint i);
 
       /*!
        * Sets the string handler that decides how the ID3v1 data will be
diff --git a/3rdparty/taglib/mpeg/id3v2/frames/popularimeterframe.cpp b/3rdparty/taglib/mpeg/id3v2/frames/popularimeterframe.cpp
index cfe8c9f44..3d4429f7d 100644
--- a/3rdparty/taglib/mpeg/id3v2/frames/popularimeterframe.cpp
+++ b/3rdparty/taglib/mpeg/id3v2/frames/popularimeterframe.cpp
@@ -109,7 +109,7 @@ void PopularimeterFrame::parseFields(const ByteVector &data)
   if(pos < size) {
     d->rating = (unsigned char)(data[pos++]);
     if(pos < size) {
-      d->counter = data.mid(pos, 4).toUInt();
+      d->counter = data.toUInt(static_cast<uint>(pos));
     }
   }
 }
diff --git a/3rdparty/taglib/mpeg/id3v2/frames/relativevolumeframe.cpp b/3rdparty/taglib/mpeg/id3v2/frames/relativevolumeframe.cpp
index 955b3ad03..e3efbc380 100644
--- a/3rdparty/taglib/mpeg/id3v2/frames/relativevolumeframe.cpp
+++ b/3rdparty/taglib/mpeg/id3v2/frames/relativevolumeframe.cpp
@@ -191,7 +191,7 @@ void RelativeVolumeFrame::parseFields(const ByteVector &data)
 
     ChannelData &channel = d->channels[type];
 
-    channel.volumeAdjustment = data.mid(pos, 2).toShort();
+    channel.volumeAdjustment = data.toShort(static_cast<uint>(pos));
     pos += 2;
 
     channel.peakVolume.bitsRepresentingPeak = data[pos];
diff --git a/3rdparty/taglib/mpeg/id3v2/frames/textidentificationframe.cpp b/3rdparty/taglib/mpeg/id3v2/frames/textidentificationframe.cpp
index 3287063ca..70ea50f8f 100644
--- a/3rdparty/taglib/mpeg/id3v2/frames/textidentificationframe.cpp
+++ b/3rdparty/taglib/mpeg/id3v2/frames/textidentificationframe.cpp
@@ -381,18 +381,12 @@ void UserTextIdentificationFrame::setDescription(const String &s)
 
 PropertyMap UserTextIdentificationFrame::asProperties() const
 {
-  String tagName = description();
-
   PropertyMap map;
-  String key = tagName.upper();
-  if(key.isNull()) // this frame's description is not a valid PropertyMap key -> add to unsupported list
-    map.unsupportedData().append(L"TXXX/" + description());
-  else {
-    StringList v = fieldList();
-    for(StringList::ConstIterator it = v.begin(); it != v.end(); ++it)
-      if(*it != description())
-        map.insert(key, *it);
-  }
+  String tagName = txxxToKey(description());
+  StringList v = fieldList();
+  for(StringList::ConstIterator it = v.begin(); it != v.end(); ++it)
+    if(it != v.begin())
+      map.insert(tagName, *it);
   return map;
 }
 
diff --git a/3rdparty/taglib/mpeg/id3v2/frames/uniquefileidentifierframe.cpp b/3rdparty/taglib/mpeg/id3v2/frames/uniquefileidentifierframe.cpp
index e12583ade..a0e842e00 100644
--- a/3rdparty/taglib/mpeg/id3v2/frames/uniquefileidentifierframe.cpp
+++ b/3rdparty/taglib/mpeg/id3v2/frames/uniquefileidentifierframe.cpp
@@ -24,8 +24,10 @@
  ***************************************************************************/
 
 #include <tbytevectorlist.h>
+#include <tpropertymap.h>
 #include <tdebug.h>
 
+#include "id3v2tag.h"
 #include "uniquefileidentifierframe.h"
 
 using namespace TagLib;
@@ -87,6 +89,34 @@ String UniqueFileIdentifierFrame::toString() const
   return String::null;
 }
 
+PropertyMap UniqueFileIdentifierFrame::asProperties() const
+{
+  PropertyMap map;
+  if(d->owner == "http://musicbrainz.org") {
+    map.insert("MUSICBRAINZ_TRACKID", String(d->identifier));
+  }
+  else {
+    map.unsupportedData().append(frameID() + String("/") + d->owner);
+  }
+  return map;
+}
+
+UniqueFileIdentifierFrame *UniqueFileIdentifierFrame::findByOwner(const ID3v2::Tag *tag, const String &o) // static
+{
+  ID3v2::FrameList comments = tag->frameList("UFID");
+
+  for(ID3v2::FrameList::ConstIterator it = comments.begin();
+      it != comments.end();
+      ++it)
+  {
+    UniqueFileIdentifierFrame *frame = dynamic_cast<UniqueFileIdentifierFrame *>(*it);
+    if(frame && frame->owner() == o)
+      return frame;
+  }
+
+  return 0;
+}
+
 void UniqueFileIdentifierFrame::parseFields(const ByteVector &data)
 {
   if(data.size() < 1) {
diff --git a/3rdparty/taglib/mpeg/id3v2/frames/uniquefileidentifierframe.h b/3rdparty/taglib/mpeg/id3v2/frames/uniquefileidentifierframe.h
index 0cf4b8f6e..add5a2e0c 100644
--- a/3rdparty/taglib/mpeg/id3v2/frames/uniquefileidentifierframe.h
+++ b/3rdparty/taglib/mpeg/id3v2/frames/uniquefileidentifierframe.h
@@ -94,6 +94,16 @@ namespace TagLib {
 
       virtual String toString() const;
 
+      PropertyMap asProperties() const;
+
+      /*!
+       * UFID frames each have a unique owner. This searches for a UFID
+       * frame with the owner \a o and returns a pointer to it.
+       *
+       * \see owner()
+       */
+      static UniqueFileIdentifierFrame *findByOwner(const Tag *tag, const String &o);
+
     protected:
       virtual void parseFields(const ByteVector &data);
       virtual ByteVector renderFields() const;
diff --git a/3rdparty/taglib/mpeg/id3v2/id3v2frame.cpp b/3rdparty/taglib/mpeg/id3v2/id3v2frame.cpp
index e15adc8c0..c5c5585d4 100644
--- a/3rdparty/taglib/mpeg/id3v2/id3v2frame.cpp
+++ b/3rdparty/taglib/mpeg/id3v2/id3v2frame.cpp
@@ -44,6 +44,7 @@
 #include "frames/urllinkframe.h"
 #include "frames/unsynchronizedlyricsframe.h"
 #include "frames/commentsframe.h"
+#include "frames/uniquefileidentifierframe.h"
 #include "frames/unknownframe.h"
 
 using namespace TagLib;
@@ -120,16 +121,20 @@ Frame *Frame::createTextualFrame(const String &key, const StringList &values) //
       TextIdentificationFrame *frame = new TextIdentificationFrame(frameID, String::UTF8);
       frame->setText(values);
       return frame;
-    } else if(values.size() == 1){  // URL frame (not WXXX); support only one value
+    } else if((frameID[0] == 'W') && (values.size() == 1)){  // URL frame (not WXXX); support only one value
         UrlLinkFrame* frame = new UrlLinkFrame(frameID);
         frame->setUrl(values.front());
         return frame;
     }
   }
+  if(key == "MUSICBRAINZ_TRACKID" && values.size() == 1) {
+    UniqueFileIdentifierFrame *frame = new UniqueFileIdentifierFrame("http://musicbrainz.org", values.front().data(String::UTF8));
+    return frame;
+  }
   // now we check if it's one of the "special" cases:
   // -LYRICS: depending on the number of values, use USLT or TXXX (with description=LYRICS)
   if((key == "LYRICS" || key.startsWith(lyricsPrefix)) && values.size() == 1){
-    UnsynchronizedLyricsFrame *frame = new UnsynchronizedLyricsFrame();
+    UnsynchronizedLyricsFrame *frame = new UnsynchronizedLyricsFrame(String::UTF8);
     frame->setDescription(key == "LYRICS" ? key : key.substr(lyricsPrefix.size()));
     frame->setText(values.front());
     return frame;
@@ -144,12 +149,14 @@ Frame *Frame::createTextualFrame(const String &key, const StringList &values) //
   // -COMMENT: depending on the number of values, use COMM or TXXX (with description=COMMENT)
   if((key == "COMMENT" || key.startsWith(commentPrefix)) && values.size() == 1){
     CommentsFrame *frame = new CommentsFrame(String::UTF8);
-    frame->setDescription(key == "COMMENT" ? key : key.substr(commentPrefix.size()));
+    if (key != "COMMENT"){
+      frame->setDescription(key.substr(commentPrefix.size()));
+    }
     frame->setText(values.front());
     return frame;
   }
   // if non of the above cases apply, we use a TXXX frame with the key as description
-  return new UserTextIdentificationFrame(key, values, String::UTF8);
+  return new UserTextIdentificationFrame(keyToTXXX(key), values, String::UTF8);
 }
 
 Frame::~Frame()
@@ -348,7 +355,7 @@ static const char *frameTranslation[][2] = {
   { "TLAN", "LANGUAGE" },
   { "TLEN", "LENGTH" },
   //{ "TMCL", "MUSICIANCREDITS" }, handled separately
-  { "TMED", "MEDIATYPE" },
+  { "TMED", "MEDIA" },
   { "TMOO", "MOOD" },
   { "TOAL", "ORIGINALALBUM" },
   { "TOFN", "ORIGINALFILENAME" },
@@ -361,7 +368,7 @@ static const char *frameTranslation[][2] = {
   { "TPE4", "REMIXER" }, // could also be ARRANGER
   { "TPOS", "DISCNUMBER" },
   { "TPRO", "PRODUCEDNOTICE" },
-  { "TPUB", "PUBLISHER" },
+  { "TPUB", "LABEL" },
   { "TRCK", "TRACKNUMBER" },
   { "TRSN", "RADIOSTATION" },
   { "TRSO", "RADIOSTATIONOWNER" },
@@ -385,6 +392,18 @@ static const char *frameTranslation[][2] = {
   //{ "USLT", "LYRICS" }, handled specially
 };
 
+static const TagLib::uint txxxFrameTranslationSize = 7;
+static const char *txxxFrameTranslation[][2] = {
+  { "MusicBrainz Album Id", "MUSICBRAINZ_ALBUMID" },
+  { "MusicBrainz Artist Id", "MUSICBRAINZ_ARTISTID" },
+  { "MusicBrainz Album Artist Id", "MUSICBRAINZ_ALBUMARTISTID" },
+  { "MusicBrainz Release Group Id", "MUSICBRAINZ_RELEASEGROUPID" },
+  { "MusicBrainz Work Id", "MUSICBRAINZ_WORKID" },
+  { "Acoustid Id", "ACOUSTID_ID" },
+  { "Acoustid Fingerprint", "ACOUSTID_FINGERPRINT" },
+  { "MusicIP PUID", "MUSICIP_PUID" },
+};
+
 Map<ByteVector, String> &idMap()
 {
   static Map<ByteVector, String> m;
@@ -394,6 +413,18 @@ Map<ByteVector, String> &idMap()
   return m;
 }
 
+Map<String, String> &txxxMap()
+{
+  static Map<String, String> m;
+  if(m.isEmpty()) {
+    for(size_t i = 0; i < txxxFrameTranslationSize; ++i) {
+      String key = String(txxxFrameTranslation[i][0]).upper();
+      m[key] = txxxFrameTranslation[i][1];
+    }
+  }
+  return m;
+}
+
 // list of deprecated frames and their successors
 static const TagLib::uint deprecatedFramesSize = 4;
 static const char *deprecatedFrames[][2] = {
@@ -433,6 +464,26 @@ ByteVector Frame::keyToFrameID(const String &s)
   return ByteVector::null;
 }
 
+String Frame::txxxToKey(const String &description)
+{
+  Map<String, String> &m = txxxMap();
+  String d = description.upper();
+  if(m.contains(d))
+    return m[d];
+  return d;
+}
+
+String Frame::keyToTXXX(const String &s)
+{
+  static Map<String, String> m;
+  if(m.isEmpty())
+    for(size_t i = 0; i < txxxFrameTranslationSize; ++i)
+      m[txxxFrameTranslation[i][1]] = txxxFrameTranslation[i][0];
+  if(m.contains(s.upper()))
+    return m[s];
+  return s;
+}
+
 PropertyMap Frame::asProperties() const
 {
   if(dynamic_cast< const UnknownFrame *>(this)) {
@@ -454,6 +505,8 @@ PropertyMap Frame::asProperties() const
     return dynamic_cast< const CommentsFrame* >(this)->asProperties();
   else if(id == "USLT")
     return dynamic_cast< const UnsynchronizedLyricsFrame* >(this)->asProperties();
+  else if(id == "UFID")
+    return dynamic_cast< const UniqueFileIdentifierFrame* >(this)->asProperties();
   PropertyMap m;
   m.unsupportedData().append(id);
   return m;
@@ -589,7 +642,7 @@ void Frame::Header::setData(const ByteVector &data, uint version)
       return;
     }
 
-    d->frameSize = data.mid(3, 3).toUInt();
+    d->frameSize = data.toUInt(3, 3, true);
 
     break;
   }
@@ -617,7 +670,7 @@ void Frame::Header::setData(const ByteVector &data, uint version)
     // Set the size -- the frame size is the four bytes starting at byte four in
     // the frame header (structure 4)
 
-    d->frameSize = data.mid(4, 4).toUInt();
+    d->frameSize = data.toUInt(4U);
 
     { // read the first byte of flags
       std::bitset<8> flags(data[8]);
@@ -664,7 +717,7 @@ void Frame::Header::setData(const ByteVector &data, uint version)
     // iTunes writes v2.4 tags with v2.3-like frame sizes
     if(d->frameSize > 127) {
       if(!isValidFrameID(data.mid(d->frameSize + 10, 4))) {
-        unsigned int uintSize = data.mid(4, 4).toUInt();
+        unsigned int uintSize = data.toUInt(4U);
         if(isValidFrameID(data.mid(uintSize + 10, 4))) {
           d->frameSize = uintSize;
         }
diff --git a/3rdparty/taglib/mpeg/id3v2/id3v2frame.h b/3rdparty/taglib/mpeg/id3v2/id3v2frame.h
index 95c4070b4..3e257d4f5 100644
--- a/3rdparty/taglib/mpeg/id3v2/id3v2frame.h
+++ b/3rdparty/taglib/mpeg/id3v2/id3v2frame.h
@@ -274,6 +274,15 @@ namespace TagLib {
        */
       static String frameIDToKey(const ByteVector &);
 
+      /*!
+       * Returns an appropriate TXXX frame description for the given free-form tag key.
+       */
+      static String keyToTXXX(const String &);
+
+      /*!
+       * Returns a free-form tag name for the given ID3 frame description.
+       */
+      static String txxxToKey(const String &);
 
       /*!
        * This helper function splits the PropertyMap \a original into three ProperytMaps
diff --git a/3rdparty/taglib/mpeg/id3v2/id3v2framefactory.cpp b/3rdparty/taglib/mpeg/id3v2/id3v2framefactory.cpp
index 7e2e98a8a..3371ca7dd 100644
--- a/3rdparty/taglib/mpeg/id3v2/id3v2framefactory.cpp
+++ b/3rdparty/taglib/mpeg/id3v2/id3v2framefactory.cpp
@@ -99,7 +99,7 @@ Frame *FrameFactory::createFrame(const ByteVector &origData, Header *tagHeader)
   // A quick sanity check -- make sure that the frameID is 4 uppercase Latin1
   // characters.  Also make sure that there is data in the frame.
 
-  if(!frameID.size() == (version < 3 ? 3 : 4) ||
+  if(frameID.size() != (version < 3 ? 3 : 4) ||
      header->frameSize() <= uint(header->dataLengthIndicator() ? 4 : 0) ||
      header->frameSize() > data.size())
   {
@@ -107,6 +107,17 @@ Frame *FrameFactory::createFrame(const ByteVector &origData, Header *tagHeader)
     return 0;
   }
 
+#ifndef NO_ITUNES_HACKS
+  if(version == 3 && frameID.size() == 4 && frameID[3] == '\0') {
+    // iTunes v2.3 tags store v2.2 frames - convert now
+    frameID = frameID.mid(0, 3);
+    header->setFrameID(frameID);
+    header->setVersion(2);
+    updateFrame(header);
+    header->setVersion(3);
+  }
+#endif
+
   for(ByteVector::ConstIterator it = frameID.begin(); it != frameID.end(); it++) {
     if( (*it < 'A' || *it > 'Z') && (*it < '0' || *it > '9') ) {
       delete header;
@@ -125,7 +136,7 @@ Frame *FrameFactory::createFrame(const ByteVector &origData, Header *tagHeader)
   // TagLib doesn't mess with encrypted frames, so just treat them
   // as unknown frames.
 
-#if HAVE_ZLIB == 0
+#if !defined(HAVE_ZLIB) || HAVE_ZLIB == 0
   if(header->compression()) {
     debug("Compressed frames are currently not supported.");
     return new UnknownFrame(data, header);
diff --git a/3rdparty/taglib/mpeg/id3v2/id3v2synchdata.cpp b/3rdparty/taglib/mpeg/id3v2/id3v2synchdata.cpp
index d4bab6795..4acfd9147 100644
--- a/3rdparty/taglib/mpeg/id3v2/id3v2synchdata.cpp
+++ b/3rdparty/taglib/mpeg/id3v2/id3v2synchdata.cpp
@@ -49,7 +49,14 @@ TagLib::uint SynchData::toUInt(const ByteVector &data)
     // Invalid data; assume this was created by some buggy software that just
     // put normal integers here rather than syncsafe ones, and try it that
     // way.
-    sum = (data.size() > 4) ? data.mid(0, 4).toUInt() : data.toUInt();
+    if(data.size() >= 4) {
+      sum = data.toUInt(0, true);
+    }
+    else {
+      ByteVector tmp(data);
+      tmp.resize(4);
+      sum = tmp.toUInt(0, true);
+    }
   }
 
   return sum;
diff --git a/3rdparty/taglib/mpeg/id3v2/id3v2tag.cpp b/3rdparty/taglib/mpeg/id3v2/id3v2tag.cpp
index 54e639203..57637a36e 100644
--- a/3rdparty/taglib/mpeg/id3v2/id3v2tag.cpp
+++ b/3rdparty/taglib/mpeg/id3v2/id3v2tag.cpp
@@ -23,6 +23,10 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif
+
 #include <tfile.h>
 
 #include "id3v2tag.h"
@@ -379,10 +383,12 @@ void ID3v2::Tag::removeUnsupportedProperties(const StringList &properties)
       for(FrameList::ConstIterator fit = l.begin(); fit != l.end(); fit++)
         if (dynamic_cast<const UnknownFrame *>(*fit) != 0)
           removeFrame(*fit);
-    } else if(it->size() == 4){
+    }
+    else if(it->size() == 4){
       ByteVector id = it->data(String::Latin1);
       removeFrames(id);
-    } else {
+    }
+    else {
       ByteVector id = it->substr(0,4).data(String::Latin1);
       if(it->size() <= 5)
         continue; // invalid specification
@@ -396,6 +402,8 @@ void ID3v2::Tag::removeUnsupportedProperties(const StringList &properties)
         frame = CommentsFrame::findByDescription(this, description);
       else if(id == "USLT")
         frame = UnsynchronizedLyricsFrame::findByDescription(this, description);
+      else if(id == "UFID")
+        frame = UniqueFileIdentifierFrame::findByOwner(this, description);
       if(frame)
         removeFrame(frame);
     }
diff --git a/3rdparty/taglib/mpeg/mpegfile.cpp b/3rdparty/taglib/mpeg/mpegfile.cpp
index 6ebff897a..f765befb0 100644
--- a/3rdparty/taglib/mpeg/mpegfile.cpp
+++ b/3rdparty/taglib/mpeg/mpegfile.cpp
@@ -156,16 +156,13 @@ void MPEG::File::removeUnsupportedProperties(const StringList &properties)
   else if(d->hasID3v1)
     d->tag.access<ID3v1::Tag>(ID3v1Index, false)->removeUnsupportedProperties(properties);
 }
+
 PropertyMap MPEG::File::setProperties(const PropertyMap &properties)
 {
-  if(d->hasID3v2)
-    return d->tag.access<ID3v2::Tag>(ID3v2Index, false)->setProperties(properties);
-  else if(d->hasAPE)
-    return d->tag.access<APE::Tag>(APEIndex, false)->setProperties(properties);
-  else if(d->hasID3v1)
-    return d->tag.access<ID3v1::Tag>(ID3v1Index, false)->setProperties(properties);
-  else
-    return d->tag.access<ID3v2::Tag>(ID3v2Index, true)->setProperties(properties);
+  if(d->hasID3v1)
+    // update ID3v1 tag if it exists, but ignore the return value
+    d->tag.access<ID3v1::Tag>(ID3v1Index, false)->setProperties(properties);
+  return d->tag.access<ID3v2::Tag>(ID3v2Index, true)->setProperties(properties);
 }
 
 MPEG::Properties *MPEG::File::audioProperties() const
@@ -189,6 +186,11 @@ bool MPEG::File::save(int tags, bool stripOthers)
 }
 
 bool MPEG::File::save(int tags, bool stripOthers, int id3v2Version)
+{
+  return save(tags, stripOthers, id3v2Version, true);
+}
+
+bool MPEG::File::save(int tags, bool stripOthers, int id3v2Version, bool duplicateTags)
 {
   if(tags == NoTags && stripOthers)
     return strip(AllTags);
@@ -206,14 +208,19 @@ bool MPEG::File::save(int tags, bool stripOthers, int id3v2Version)
     return false;
   }
 
-  // Create the tags if we've been asked to.  Copy the values from the tag that
-  // does exist into the new tag, except if the existing tag is to be stripped.
+  // Create the tags if we've been asked to.
 
-  if((tags & ID3v2) && ID3v1Tag() && !(stripOthers && !(tags & ID3v1)))
-    Tag::duplicate(ID3v1Tag(), ID3v2Tag(true), false);
+  if (duplicateTags) {
 
-  if((tags & ID3v1) && d->tag[ID3v2Index] && !(stripOthers && !(tags & ID3v2)))
-    Tag::duplicate(ID3v2Tag(), ID3v1Tag(true), false);
+    // Copy the values from the tag that does exist into the new tag,
+    // except if the existing tag is to be stripped.
+
+    if((tags & ID3v2) && ID3v1Tag() && !(stripOthers && !(tags & ID3v1)))
+      Tag::duplicate(ID3v1Tag(), ID3v2Tag(true), false);
+
+    if((tags & ID3v1) && d->tag[ID3v2Index] && !(stripOthers && !(tags & ID3v2)))
+      Tag::duplicate(ID3v2Tag(), ID3v1Tag(true), false);
+  }
 
   bool success = true;
 
@@ -437,6 +444,21 @@ long MPEG::File::lastFrameOffset()
   return previousFrameOffset(ID3v1Tag() ? d->ID3v1Location - 1 : length());
 }
 
+bool MPEG::File::hasID3v1Tag() const
+{
+  return d->hasID3v1;
+}
+
+bool MPEG::File::hasID3v2Tag() const
+{
+  return d->hasID3v2;
+}
+
+bool MPEG::File::hasAPETag() const
+{
+  return d->hasAPE;
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // private members
 ////////////////////////////////////////////////////////////////////////////////
@@ -634,9 +656,6 @@ void MPEG::File::findAPE()
 
 bool MPEG::File::secondSynchByte(char byte)
 {
-  if(uchar(byte) == 0xff)
-    return false;
-
   std::bitset<8> b(byte);
 
   // check to see if the byte matches 111xxxxx
diff --git a/3rdparty/taglib/mpeg/mpegfile.h b/3rdparty/taglib/mpeg/mpegfile.h
index 9967a9912..3fc01e680 100644
--- a/3rdparty/taglib/mpeg/mpegfile.h
+++ b/3rdparty/taglib/mpeg/mpegfile.h
@@ -71,9 +71,10 @@ namespace TagLib {
       };
 
       /*!
-       * Contructs an MPEG file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an MPEG file from \a file.  If \a readProperties is true the
+       * file's audio properties will also be read.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        *
        * \deprecated This constructor will be dropped in favor of the one below
        * in a future version.
@@ -82,28 +83,31 @@ namespace TagLib {
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs an MPEG file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.  The frames will be created using
+       * Constructs an MPEG file from \a file.  If \a readProperties is true the
+       * file's audio properties will also be read.
+       *
+       * If this file contains and ID3v2 tag the frames will be created using
        * \a frameFactory.
        *
-       * \deprecated This constructor will be dropped in favor of the one below
-       * in a future version.
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
+      // BIC: merge with the above constructor
       File(FileName file, ID3v2::FrameFactory *frameFactory,
            bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs an MPEG file from \a stream.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.  The frames will be created using
-       * \a frameFactory.
+       * Constructs an MPEG file from \a stream.  If \a readProperties is true the
+       * file's audio properties will also be read.
        *
        * \note TagLib will *not* take ownership of the stream, the caller is
        * responsible for deleting it after the File object.
+       *
+       * If this file contains and ID3v2 tag the frames will be created using
+       * \a frameFactory.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
-      // BIC: merge with the above constructor
       File(IOStream *stream, ID3v2::FrameFactory *frameFactory,
            bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
@@ -133,7 +137,7 @@ namespace TagLib {
       virtual Tag *tag() const;
 
       /*!
-       * Implements the unified property interface -- export function.
+       * Implements the reading part of the unified property interface.
        * If the file contains more than one tag, only the
        * first one (in the order ID3v2, APE, ID3v1) will be converted to the
        * PropertyMap.
@@ -143,9 +147,12 @@ namespace TagLib {
       void removeUnsupportedProperties(const StringList &properties);
 
       /*!
-       * Implements the unified tag dictionary interface -- import function.
-       * As with the export, only one tag is taken into account. If the file
-       * has no tag at all, ID3v2 will be created.
+       * Implements the writing part of the unified tag dictionary interface.
+       * In order to avoid problems with deprecated tag formats, this method
+       * always creates an ID3v2 tag if necessary.
+       * If an ID3v1 tag  exists, it will be updated as well, within the
+       * limitations of that format.
+       * The returned PropertyMap refers to the ID3v2 tag only.
        */
       PropertyMap setProperties(const PropertyMap &);
 
@@ -210,42 +217,78 @@ namespace TagLib {
       // BIC: combine with the above method
       bool save(int tags, bool stripOthers, int id3v2Version);
 
+      /*!
+       * Save the file.  This will attempt to save all of the tag types that are
+       * specified by OR-ing together TagTypes values.  The save() method above
+       * uses AllTags.  This returns true if saving was successful.
+       *
+       * If \a stripOthers is true this strips all tags not included in the mask,
+       * but does not modify them in memory, so later calls to save() which make
+       * use of these tags will remain valid.  This also strips empty tags.
+       *
+       * The \a id3v2Version parameter specifies the version of the saved
+       * ID3v2 tag. It can be either 4 or 3.
+       *
+       * If \a duplicateTags is true and at least one tag -- ID3v1 or ID3v2 --
+       * exists this will duplicate its content into the other tag.
+       */
+      // BIC: combine with the above method
+      bool save(int tags, bool stripOthers, int id3v2Version, bool duplicateTags);
+
       /*!
        * Returns a pointer to the ID3v2 tag of the file.
        *
-       * A tag will always be returned, regardless of whether there is a
-       * tag in the file or not. Use ID3v2::Tag::isEmpty() to check if
-       * the tag contains no data.
+       * If \a create is false (the default) this may return a null pointer
+       * if there is no valid ID3v2 tag.  If \a create is true it will create
+       * an ID3v2 tag if one does not exist and returns a valid pointer.
+       *
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an ID3v2 tag.  Use hasID3v2Tag() to check if the file 
+       * on disk actually has an ID3v2 tag.
        *
        * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasID3v2Tag()
        */
       ID3v2::Tag *ID3v2Tag(bool create = false);
 
       /*!
        * Returns a pointer to the ID3v1 tag of the file.
        *
-       * A tag will always be returned, regardless of whether there is a
-       * tag in the file or not. Use Tag::isEmpty() to check if
-       * the tag contains no data.
+       * If \a create is false (the default) this may return a null pointer
+       * if there is no valid ID3v1 tag.  If \a create is true it will create
+       * an ID3v1 tag if one does not exist and returns a valid pointer.
+       *
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an ID3v1 tag.  Use hasID3v1Tag() to check if the file 
+       * on disk actually has an ID3v1 tag.
        *
        * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasID3v1Tag()
        */
       ID3v1::Tag *ID3v1Tag(bool create = false);
 
       /*!
        * Returns a pointer to the APE tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
+       * If \a create is false (the default) this may return a null pointer
        * if there is no valid APE tag.  If \a create is true it will create
-       * an APE tag if one does not exist.
+       * an APE tag if one does not exist and returns a valid pointer.
+       *
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an APE tag.  Use hasAPETag() to check if the file 
+       * on disk actually has an APE tag.
        *
        * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasAPETag()
        */
       APE::Tag *APETag(bool create = false);
 
@@ -301,6 +344,27 @@ namespace TagLib {
        */
       long lastFrameOffset();
 
+      /*!
+       * Returns whether or not the file on disk actually has an ID3v1 tag.
+       *
+       * \see ID3v1Tag()
+       */
+      bool hasID3v1Tag() const;
+
+      /*!
+       * Returns whether or not the file on disk actually has an ID3v2 tag.
+       *
+       * \see ID3v2Tag()
+       */
+      bool hasID3v2Tag() const;
+
+      /*!
+       * Returns whether or not the file on disk actually has an APE tag.
+       *
+       * \see APETag()
+       */
+      bool hasAPETag() const;
+
     private:
       File(const File &);
       File &operator=(const File &);
diff --git a/3rdparty/taglib/mpeg/mpegheader.cpp b/3rdparty/taglib/mpeg/mpegheader.cpp
index c715dbc14..a582f7589 100644
--- a/3rdparty/taglib/mpeg/mpegheader.cpp
+++ b/3rdparty/taglib/mpeg/mpegheader.cpp
@@ -28,6 +28,7 @@
 #include <tbytevector.h>
 #include <tstring.h>
 #include <tdebug.h>
+#include "trefcounter.h"
 
 #include "mpegheader.h"
 
diff --git a/3rdparty/taglib/mpeg/xingheader.cpp b/3rdparty/taglib/mpeg/xingheader.cpp
index 1ba932de3..9e20127ee 100644
--- a/3rdparty/taglib/mpeg/xingheader.cpp
+++ b/3rdparty/taglib/mpeg/xingheader.cpp
@@ -108,8 +108,8 @@ void MPEG::XingHeader::parse(const ByteVector &data)
     return;
   }
 
-  d->frames = data.mid(8, 4).toUInt();
-  d->size = data.mid(12, 4).toUInt();
+  d->frames = data.toUInt(8U);
+  d->size   = data.toUInt(12U);
 
   d->valid = true;
 }
diff --git a/3rdparty/taglib/ogg/flac/oggflacfile.cpp b/3rdparty/taglib/ogg/flac/oggflacfile.cpp
index 9d9c303d7..bdf824595 100644
--- a/3rdparty/taglib/ogg/flac/oggflacfile.cpp
+++ b/3rdparty/taglib/ogg/flac/oggflacfile.cpp
@@ -26,6 +26,7 @@
 #include <tbytevector.h>
 #include <tstring.h>
 #include <tdebug.h>
+#include <tpropertymap.h>
 
 #include <xiphcomment.h>
 #include "oggflacfile.h"
@@ -72,14 +73,16 @@ Ogg::FLAC::File::File(FileName file, bool readProperties,
                       Properties::ReadStyle propertiesStyle) : Ogg::File(file)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 Ogg::FLAC::File::File(IOStream *stream, bool readProperties,
                       Properties::ReadStyle propertiesStyle) : Ogg::File(stream)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 Ogg::FLAC::File::~File()
@@ -92,6 +95,16 @@ Ogg::XiphComment *Ogg::FLAC::File::tag() const
   return d->comment;
 }
 
+PropertyMap Ogg::FLAC::File::properties() const
+{
+  return d->comment->properties();
+}
+
+PropertyMap Ogg::FLAC::File::setProperties(const PropertyMap &properties)
+{
+  return d->comment->setProperties(properties);
+}  
+
 Properties *Ogg::FLAC::File::audioProperties() const
 {
   return d->properties;
@@ -124,6 +137,11 @@ bool Ogg::FLAC::File::save()
   return Ogg::File::save();
 }
 
+bool Ogg::FLAC::File::hasXiphComment() const
+{
+  return d->hasXiphComment;
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // private members
 ////////////////////////////////////////////////////////////////////////////////
@@ -228,7 +246,7 @@ void Ogg::FLAC::File::scan()
 
   char blockType = header[0] & 0x7f;
   bool lastBlock = (header[0] & 0x80) != 0;
-  uint length = header.mid(1, 3).toUInt();
+  uint length = header.toUInt(1, 3, true);
   overhead += length;
 
   // Sanity: First block should be the stream_info metadata
@@ -238,7 +256,7 @@ void Ogg::FLAC::File::scan()
     return;
   }
 
-  d->streamInfoData = metadataHeader.mid(4,length);
+  d->streamInfoData = metadataHeader.mid(4, length);
 
   // Search through the remaining metadata
 
@@ -251,7 +269,7 @@ void Ogg::FLAC::File::scan()
     header = metadataHeader.mid(0, 4);
     blockType = header[0] & 0x7f;
     lastBlock = (header[0] & 0x80) != 0;
-    length = header.mid(1, 3).toUInt();
+    length = header.toUInt(1, 3, true);
     overhead += length;
 
     if(blockType == 1) {
diff --git a/3rdparty/taglib/ogg/flac/oggflacfile.h b/3rdparty/taglib/ogg/flac/oggflacfile.h
index 8558cfdfa..8d87e486b 100644
--- a/3rdparty/taglib/ogg/flac/oggflacfile.h
+++ b/3rdparty/taglib/ogg/flac/oggflacfile.h
@@ -64,20 +64,22 @@ namespace TagLib {
     {
     public:
       /*!
-       * Contructs an Ogg/FLAC file from \a file.  If \a readProperties is true
-       * the file's audio properties will also be read using \a propertiesStyle.
-       * If false, \a propertiesStyle is ignored.
+       * Constructs an Ogg/FLAC file from \a file.  If \a readProperties is true 
+       * the file's audio properties will also be read.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(FileName file, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs an Ogg/FLAC file from \a file.  If \a readProperties is true
-       * the file's audio properties will also be read using \a propertiesStyle.
-       * If false, \a propertiesStyle is ignored.
+       * Constructs an Ogg/FLAC file from \a stream.  If \a readProperties is true 
+       * the file's audio properties will also be read.
        *
        * \note TagLib will *not* take ownership of the stream, the caller is
        * responsible for deleting it after the File object.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(IOStream *stream, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
@@ -89,6 +91,16 @@ namespace TagLib {
 
       /*!
        * Returns the Tag for this file.  This will always be a XiphComment.
+       *
+       * \note This always returns a valid pointer regardless of whether or not 
+       * the file on disk has a XiphComment.  Use hasXiphComment() to check if 
+       * the file on disk actually has a XiphComment.
+       * 
+       * \note The Tag <b>is still</b> owned by the FLAC::File and should not be
+       * deleted by the user.  It will be deleted when the file (object) is
+       * destroyed.
+       *
+       * \see hasXiphComment()
        */
       virtual XiphComment *tag() const;
 
@@ -98,6 +110,20 @@ namespace TagLib {
        */
       virtual Properties *audioProperties() const;
 
+
+      /*! 
+       * Implements the unified property interface -- export function.
+       * This forwards directly to XiphComment::properties().
+       */
+      PropertyMap properties() const;
+
+      /*! 
+       * Implements the unified tag dictionary interface -- import function.
+       * Like properties(), this is a forwarder to the file's XiphComment.
+       */
+      PropertyMap setProperties(const PropertyMap &); 
+
+
       /*!
        * Save the file.  This will primarily save and update the XiphComment.
        * Returns true if the save is successful.
@@ -110,6 +136,13 @@ namespace TagLib {
        */
       long streamLength();
 
+      /*!
+       * Returns whether or not the file on disk actually has a XiphComment.
+       *
+       * \see tag()
+       */
+      bool hasXiphComment() const;
+
     private:
       File(const File &);
       File &operator=(const File &);
diff --git a/3rdparty/taglib/ogg/oggfile.h b/3rdparty/taglib/ogg/oggfile.h
index b36daecb1..8fed4ba09 100644
--- a/3rdparty/taglib/ogg/oggfile.h
+++ b/3rdparty/taglib/ogg/oggfile.h
@@ -82,9 +82,7 @@ namespace TagLib {
 
     protected:
       /*!
-       * Contructs an Ogg file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an Ogg file from \a file.
        *
        * \note This constructor is protected since Ogg::File shouldn't be
        * instantiated directly but rather should be used through the codec
@@ -93,9 +91,7 @@ namespace TagLib {
       File(FileName file);
 
       /*!
-       * Contructs an Ogg file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs an Ogg file from \a stream.
        *
        * \note This constructor is protected since Ogg::File shouldn't be
        * instantiated directly but rather should be used through the codec
diff --git a/3rdparty/taglib/ogg/oggpageheader.cpp b/3rdparty/taglib/ogg/oggpageheader.cpp
index f9608ab7f..b93331354 100644
--- a/3rdparty/taglib/ogg/oggpageheader.cpp
+++ b/3rdparty/taglib/ogg/oggpageheader.cpp
@@ -255,9 +255,9 @@ void Ogg::PageHeader::read()
   d->firstPageOfStream = flags.test(1);
   d->lastPageOfStream = flags.test(2);
 
-  d->absoluteGranularPosition = data.mid(6, 8).toLongLong(false);
-  d->streamSerialNumber = data.mid(14, 4).toUInt(false);
-  d->pageSequenceNumber = data.mid(18, 4).toUInt(false);
+  d->absoluteGranularPosition = data.toLongLong(6, false);
+  d->streamSerialNumber = data.toUInt(14, false);
+  d->pageSequenceNumber = data.toUInt(18, false);
 
   // Byte number 27 is the number of page segments, which is the only variable
   // length portion of the page header.  After reading the number of page
diff --git a/3rdparty/taglib/ogg/opus/opusfile.cpp b/3rdparty/taglib/ogg/opus/opusfile.cpp
new file mode 100644
index 000000000..cb81a32b0
--- /dev/null
+++ b/3rdparty/taglib/ogg/opus/opusfile.cpp
@@ -0,0 +1,139 @@
+/***************************************************************************
+    copyright            : (C) 2012 by Lukáš Lalinský
+    email                : lalinsky@gmail.com
+
+    copyright            : (C) 2002 - 2008 by Scott Wheeler
+    email                : wheeler@kde.org
+                           (original Vorbis implementation)
+ ***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#include <bitset>
+
+#include <tstring.h>
+#include <tdebug.h>
+#include <tpropertymap.h>
+
+#include "opusfile.h"
+
+using namespace TagLib;
+using namespace TagLib::Ogg;
+
+class Opus::File::FilePrivate
+{
+public:
+  FilePrivate() :
+    comment(0),
+    properties(0) {}
+
+  ~FilePrivate()
+  {
+    delete comment;
+    delete properties;
+  }
+
+  Ogg::XiphComment *comment;
+  Properties *properties;
+};
+
+////////////////////////////////////////////////////////////////////////////////
+// public members
+////////////////////////////////////////////////////////////////////////////////
+
+Opus::File::File(FileName file, bool readProperties, Properties::ReadStyle propertiesStyle) :
+  Ogg::File(file),
+  d(new FilePrivate())
+{
+  if(isOpen())
+    read(readProperties, propertiesStyle);
+}
+
+Opus::File::File(IOStream *stream, bool readProperties, Properties::ReadStyle propertiesStyle) :
+  Ogg::File(stream),
+  d(new FilePrivate())
+{
+  if(isOpen())
+    read(readProperties, propertiesStyle);
+}
+
+Opus::File::~File()
+{
+  delete d;
+}
+
+Ogg::XiphComment *Opus::File::tag() const
+{
+  return d->comment;
+}
+
+PropertyMap Opus::File::properties() const
+{
+  return d->comment->properties();
+}
+
+PropertyMap Opus::File::setProperties(const PropertyMap &properties)
+{
+  return d->comment->setProperties(properties);
+}
+
+Opus::Properties *Opus::File::audioProperties() const
+{
+  return d->properties;
+}
+
+bool Opus::File::save()
+{
+  if(!d->comment)
+    d->comment = new Ogg::XiphComment;
+
+  setPacket(1, ByteVector("OpusTags", 8) + d->comment->render(false));
+
+  return Ogg::File::save();
+}
+
+////////////////////////////////////////////////////////////////////////////////
+// private members
+////////////////////////////////////////////////////////////////////////////////
+
+void Opus::File::read(bool readProperties, Properties::ReadStyle propertiesStyle)
+{
+  ByteVector opusHeaderData = packet(0);
+
+  if(!opusHeaderData.startsWith("OpusHead")) {
+    setValid(false);
+    debug("Opus::File::read() -- invalid Opus identification header");
+    return;
+  }
+
+  ByteVector commentHeaderData = packet(1);
+
+  if(!commentHeaderData.startsWith("OpusTags")) {
+    setValid(false);
+    debug("Opus::File::read() -- invalid Opus tags header");
+    return;
+  }
+
+  d->comment = new Ogg::XiphComment(commentHeaderData.mid(8));
+
+  if(readProperties)
+    d->properties = new Properties(this, propertiesStyle);
+}
diff --git a/3rdparty/taglib/ogg/opus/opusfile.h b/3rdparty/taglib/ogg/opus/opusfile.h
new file mode 100644
index 000000000..275167e40
--- /dev/null
+++ b/3rdparty/taglib/ogg/opus/opusfile.h
@@ -0,0 +1,124 @@
+/***************************************************************************
+    copyright            : (C) 2012 by Lukáš Lalinský
+    email                : lalinsky@gmail.com
+
+    copyright            : (C) 2002 - 2008 by Scott Wheeler
+    email                : wheeler@kde.org
+                           (original Vorbis implementation)
+***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#ifndef TAGLIB_OPUSFILE_H
+#define TAGLIB_OPUSFILE_H
+
+#include "oggfile.h"
+#include "xiphcomment.h"
+
+#include "opusproperties.h"
+
+namespace TagLib {
+
+  namespace Ogg {
+
+    //! A namespace containing classes for Opus metadata
+
+    namespace Opus {
+
+      //! An implementation of Ogg::File with Opus specific methods
+
+      /*!
+       * This is the central class in the Ogg Opus metadata processing collection
+       * of classes.  It's built upon Ogg::File which handles processing of the Ogg
+       * logical bitstream and breaking it down into pages which are handled by
+       * the codec implementations, in this case Opus specifically.
+       */
+
+      class TAGLIB_EXPORT File : public Ogg::File
+      {
+      public:
+        /*!
+         * Constructs an Opus file from \a file.  If \a readProperties is true the
+         * file's audio properties will also be read.
+         *
+         * \note In the current implementation, \a propertiesStyle is ignored.
+         */
+        File(FileName file, bool readProperties = true,
+             Properties::ReadStyle propertiesStyle = Properties::Average);
+
+        /*!
+         * Constructs an Opus file from \a stream.  If \a readProperties is true the
+         * file's audio properties will also be read.
+         *
+         * \note TagLib will *not* take ownership of the stream, the caller is
+         * responsible for deleting it after the File object.
+         *
+         * \note In the current implementation, \a propertiesStyle is ignored.
+         */
+        File(IOStream *stream, bool readProperties = true,
+             Properties::ReadStyle propertiesStyle = Properties::Average);
+
+        /*!
+         * Destroys this instance of the File.
+         */
+        virtual ~File();
+
+        /*!
+         * Returns the XiphComment for this file.  XiphComment implements the tag
+         * interface, so this serves as the reimplementation of
+         * TagLib::File::tag().
+         */
+        virtual Ogg::XiphComment *tag() const;
+
+        /*!
+         * Implements the unified property interface -- export function.
+         * This forwards directly to XiphComment::properties().
+         */
+        PropertyMap properties() const;
+
+        /*!
+         * Implements the unified tag dictionary interface -- import function.
+         * Like properties(), this is a forwarder to the file's XiphComment.
+         */
+        PropertyMap setProperties(const PropertyMap &);
+
+        /*!
+         * Returns the Opus::Properties for this file.  If no audio properties
+         * were read then this will return a null pointer.
+         */
+        virtual Properties *audioProperties() const;
+
+        virtual bool save();
+
+      private:
+        File(const File &);
+        File &operator=(const File &);
+
+        void read(bool readProperties, Properties::ReadStyle propertiesStyle);
+
+        class FilePrivate;
+        FilePrivate *d;
+      };
+    }
+  }
+}
+
+#endif
diff --git a/3rdparty/taglib/ogg/opus/opusproperties.cpp b/3rdparty/taglib/ogg/opus/opusproperties.cpp
new file mode 100644
index 000000000..7bdcd39df
--- /dev/null
+++ b/3rdparty/taglib/ogg/opus/opusproperties.cpp
@@ -0,0 +1,161 @@
+/***************************************************************************
+    copyright            : (C) 2012 by Lukáš Lalinský
+    email                : lalinsky@gmail.com
+
+    copyright            : (C) 2002 - 2008 by Scott Wheeler
+    email                : wheeler@kde.org
+                           (original Vorbis implementation)
+ ***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#include <tstring.h>
+#include <tdebug.h>
+
+#include <oggpageheader.h>
+
+#include "opusproperties.h"
+#include "opusfile.h"
+
+using namespace TagLib;
+using namespace TagLib::Ogg;
+
+class Opus::Properties::PropertiesPrivate
+{
+public:
+  PropertiesPrivate(File *f, ReadStyle s) :
+    file(f),
+    style(s),
+    length(0),
+    inputSampleRate(0),
+    channels(0),
+    opusVersion(0) {}
+
+  File *file;
+  ReadStyle style;
+  int length;
+  int inputSampleRate;
+  int channels;
+  int opusVersion;
+};
+
+////////////////////////////////////////////////////////////////////////////////
+// public members
+////////////////////////////////////////////////////////////////////////////////
+
+Opus::Properties::Properties(File *file, ReadStyle style) : AudioProperties(style)
+{
+  d = new PropertiesPrivate(file, style);
+  read();
+}
+
+Opus::Properties::~Properties()
+{
+  delete d;
+}
+
+int Opus::Properties::length() const
+{
+  return d->length;
+}
+
+int Opus::Properties::bitrate() const
+{
+  return 0;
+}
+
+int Opus::Properties::sampleRate() const
+{
+  // Opus can decode any stream at a sample rate of 8, 12, 16, 24, or 48 kHz,
+  // so there is no single sample rate. Let's assume it's the highest
+  // possible.
+  return 48000;
+}
+
+int Opus::Properties::channels() const
+{
+  return d->channels;
+}
+
+int Opus::Properties::inputSampleRate() const
+{
+  return d->inputSampleRate;
+}
+
+int Opus::Properties::opusVersion() const
+{
+  return d->opusVersion;
+}
+
+////////////////////////////////////////////////////////////////////////////////
+// private members
+////////////////////////////////////////////////////////////////////////////////
+
+void Opus::Properties::read()
+{
+  // Get the identification header from the Ogg implementation.
+
+  // http://tools.ietf.org/html/draft-terriberry-oggopus-01#section-5.1
+
+  ByteVector data = d->file->packet(0);
+
+  // *Magic Signature*
+  uint pos = 8;
+
+  // *Version* (8 bits, unsigned)
+  d->opusVersion = uchar(data.at(pos));
+  pos += 1;
+
+  // *Output Channel Count* 'C' (8 bits, unsigned)
+  d->channels = uchar(data.at(pos));
+  pos += 1;
+
+  // *Pre-skip* (16 bits, unsigned, little endian)
+  const ushort preSkip = data.toUShort(pos, false);
+  pos += 2;
+
+  // *Input Sample Rate* (32 bits, unsigned, little endian)
+  d->inputSampleRate = data.toUInt(pos, false);
+  pos += 4;
+
+  // *Output Gain* (16 bits, signed, little endian)
+  pos += 2;
+
+  // *Channel Mapping Family* (8 bits, unsigned)
+  pos += 1;
+
+  const Ogg::PageHeader *first = d->file->firstPageHeader();
+  const Ogg::PageHeader *last = d->file->lastPageHeader();
+
+  if(first && last) {
+    long long start = first->absoluteGranularPosition();
+    long long end = last->absoluteGranularPosition();
+
+    if(start >= 0 && end >= 0)
+      d->length = (int) ((end - start - preSkip) / 48000);
+    else {
+      debug("Opus::Properties::read() -- The PCM values for the start or "
+            "end of this file was incorrect.");
+    }
+  }
+  else
+    debug("Opus::Properties::read() -- Could not find valid first and last Ogg pages.");
+}
diff --git a/3rdparty/taglib/ogg/opus/opusproperties.h b/3rdparty/taglib/ogg/opus/opusproperties.h
new file mode 100644
index 000000000..946f16757
--- /dev/null
+++ b/3rdparty/taglib/ogg/opus/opusproperties.h
@@ -0,0 +1,96 @@
+/***************************************************************************
+    copyright            : (C) 2012 by Lukáš Lalinský
+    email                : lalinsky@gmail.com
+
+    copyright            : (C) 2002 - 2008 by Scott Wheeler
+    email                : wheeler@kde.org
+                           (original Vorbis implementation)
+***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#ifndef TAGLIB_OPUSPROPERTIES_H
+#define TAGLIB_OPUSPROPERTIES_H
+
+#include "audioproperties.h"
+
+namespace TagLib {
+
+  namespace Ogg {
+
+    namespace Opus {
+
+      class File;
+
+      //! An implementation of audio property reading for Ogg Opus
+
+      /*!
+       * This reads the data from an Ogg Opus stream found in the AudioProperties
+       * API.
+       */
+
+      class TAGLIB_EXPORT Properties : public AudioProperties
+      {
+      public:
+        /*!
+         * Create an instance of Opus::Properties with the data read from the
+         * Opus::File \a file.
+         */
+        Properties(File *file, ReadStyle style = Average);
+
+        /*!
+         * Destroys this Opus::Properties instance.
+         */
+        virtual ~Properties();
+
+        // Reimplementations.
+
+        virtual int length() const;
+        virtual int bitrate() const;
+        virtual int sampleRate() const;
+        virtual int channels() const;
+
+        /*!
+         * The Opus codec supports decoding at multiple sample rates, there is no
+         * single sample rate of the encoded stream. This returns the sample rate
+         * of the original audio stream.
+         */
+        int inputSampleRate() const;
+
+        /*!
+         * Returns the Opus version, currently "0" (as specified by the spec).
+         */
+        int opusVersion() const;
+
+      private:
+        Properties(const Properties &);
+        Properties &operator=(const Properties &);
+
+        void read();
+
+        class PropertiesPrivate;
+        PropertiesPrivate *d;
+      };
+    }
+  }
+}
+
+#endif
diff --git a/3rdparty/taglib/ogg/speex/speexfile.cpp b/3rdparty/taglib/ogg/speex/speexfile.cpp
index 3a4940a26..e83f0ad9b 100644
--- a/3rdparty/taglib/ogg/speex/speexfile.cpp
+++ b/3rdparty/taglib/ogg/speex/speexfile.cpp
@@ -31,6 +31,7 @@
 
 #include <tstring.h>
 #include <tdebug.h>
+#include <tpropertymap.h>
 
 #include "speexfile.h"
 
@@ -62,14 +63,16 @@ Speex::File::File(FileName file, bool readProperties,
                    Properties::ReadStyle propertiesStyle) : Ogg::File(file)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 Speex::File::File(IOStream *stream, bool readProperties,
                    Properties::ReadStyle propertiesStyle) : Ogg::File(stream)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 Speex::File::~File()
@@ -82,6 +85,16 @@ Ogg::XiphComment *Speex::File::tag() const
   return d->comment;
 }
 
+PropertyMap Speex::File::properties() const
+{
+  return d->comment->properties();
+}
+
+PropertyMap Speex::File::setProperties(const PropertyMap &properties)
+{
+  return d->comment->setProperties(properties);
+}
+
 Speex::Properties *Speex::File::audioProperties() const
 {
   return d->properties;
diff --git a/3rdparty/taglib/ogg/speex/speexfile.h b/3rdparty/taglib/ogg/speex/speexfile.h
index dfe51ec4d..880e68904 100644
--- a/3rdparty/taglib/ogg/speex/speexfile.h
+++ b/3rdparty/taglib/ogg/speex/speexfile.h
@@ -56,20 +56,22 @@ namespace TagLib {
       {
       public:
         /*!
-         * Contructs a Speex file from \a file.  If \a readProperties is true the
-         * file's audio properties will also be read using \a propertiesStyle.  If
-         * false, \a propertiesStyle is ignored.
+         * Constructs a Speex file from \a file.  If \a readProperties is true the
+         * file's audio properties will also be read.
+         *
+         * \note In the current implementation, \a propertiesStyle is ignored.
          */
         File(FileName file, bool readProperties = true,
              Properties::ReadStyle propertiesStyle = Properties::Average);
 
         /*!
-         * Contructs a Speex file from \a file.  If \a readProperties is true the
-         * file's audio properties will also be read using \a propertiesStyle.  If
-         * false, \a propertiesStyle is ignored.
+         * Constructs a Speex file from \a stream.  If \a readProperties is true the
+         * file's audio properties will also be read.
          *
          * \note TagLib will *not* take ownership of the stream, the caller is
          * responsible for deleting it after the File object.
+         *
+         * \note In the current implementation, \a propertiesStyle is ignored.
          */
         File(IOStream *stream, bool readProperties = true,
              Properties::ReadStyle propertiesStyle = Properties::Average);
@@ -86,12 +88,26 @@ namespace TagLib {
          */
         virtual Ogg::XiphComment *tag() const;
 
+        /*!
+         * Implements the unified property interface -- export function.
+         * This forwards directly to XiphComment::properties().
+         */
+        PropertyMap properties() const;
+
+        /*!
+         * Implements the unified tag dictionary interface -- import function.
+         * Like properties(), this is a forwarder to the file's XiphComment.
+         */
+        PropertyMap setProperties(const PropertyMap &);
+
         /*!
          * Returns the Speex::Properties for this file.  If no audio properties
          * were read then this will return a null pointer.
          */
         virtual Properties *audioProperties() const;
 
+
+
         virtual bool save();
 
       private:
diff --git a/3rdparty/taglib/ogg/speex/speexproperties.cpp b/3rdparty/taglib/ogg/speex/speexproperties.cpp
index 980f12dfd..5aaa91530 100644
--- a/3rdparty/taglib/ogg/speex/speexproperties.cpp
+++ b/3rdparty/taglib/ogg/speex/speexproperties.cpp
@@ -113,32 +113,32 @@ void Speex::Properties::read()
 
   ByteVector data = d->file->packet(0);
 
-  int pos = 28;
+  uint pos = 28;
 
   // speex_version_id;       /**< Version for Speex (for checking compatibility) */
-  d->speexVersion = data.mid(pos, 4).toUInt(false);
+  d->speexVersion = data.toUInt(pos, false);
   pos += 4;
 
   // header_size;            /**< Total size of the header ( sizeof(SpeexHeader) ) */
   pos += 4;
 
   // rate;                   /**< Sampling rate used */
-  d->sampleRate = data.mid(pos, 4).toUInt(false);
+  d->sampleRate = data.toUInt(pos, false);
   pos += 4;
 
   // mode;                   /**< Mode used (0 for narrowband, 1 for wideband) */
-  d->mode = data.mid(pos, 4).toUInt(false);
+  d->mode = data.toUInt(pos, false);
   pos += 4;
 
   // mode_bitstream_version; /**< Version ID of the bit-stream */
   pos += 4;
 
   // nb_channels;            /**< Number of channels encoded */
-  d->channels = data.mid(pos, 4).toUInt(false);
+  d->channels = data.toUInt(pos, false);
   pos += 4;
 
   // bitrate;                /**< Bit-rate used */
-  d->bitrate = data.mid(pos, 4).toUInt(false);
+  d->bitrate = data.toUInt(pos, false);
   pos += 4;
 
   // frame_size;             /**< Size of frames */
@@ -146,7 +146,7 @@ void Speex::Properties::read()
   pos += 4;
 
   // vbr;                    /**< 1 for a VBR encoding, 0 otherwise */
-  d->vbr = data.mid(pos, 4).toUInt(false) == 1;
+  d->vbr = data.toUInt(pos, false) == 1;
   pos += 4;
 
   // frames_per_packet;      /**< Number of frames stored per Ogg packet */
diff --git a/3rdparty/taglib/ogg/vorbis/vorbisfile.cpp b/3rdparty/taglib/ogg/vorbis/vorbisfile.cpp
index e2eed9e25..82984536b 100644
--- a/3rdparty/taglib/ogg/vorbis/vorbisfile.cpp
+++ b/3rdparty/taglib/ogg/vorbis/vorbisfile.cpp
@@ -67,14 +67,16 @@ Vorbis::File::File(FileName file, bool readProperties,
                    Properties::ReadStyle propertiesStyle) : Ogg::File(file)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 Vorbis::File::File(IOStream *stream, bool readProperties,
                    Properties::ReadStyle propertiesStyle) : Ogg::File(stream)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 Vorbis::File::~File()
diff --git a/3rdparty/taglib/ogg/vorbis/vorbisfile.h b/3rdparty/taglib/ogg/vorbis/vorbisfile.h
index 6e4d4fc42..7735a11be 100644
--- a/3rdparty/taglib/ogg/vorbis/vorbisfile.h
+++ b/3rdparty/taglib/ogg/vorbis/vorbisfile.h
@@ -63,20 +63,22 @@ namespace TagLib {
     {
     public:
       /*!
-       * Contructs a Vorbis file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs a Vorbis file from \a file.  If \a readProperties is true the
+       * file's audio properties will also be read.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(FileName file, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs a Vorbis file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs a Vorbis file from \a stream.  If \a readProperties is true the
+       * file's audio properties will also be read.
        *
        * \note TagLib will *not* take ownership of the stream, the caller is
        * responsible for deleting it after the File object.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(IOStream *stream, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
diff --git a/3rdparty/taglib/ogg/vorbis/vorbisproperties.cpp b/3rdparty/taglib/ogg/vorbis/vorbisproperties.cpp
index c67e16773..b5e88bfd0 100644
--- a/3rdparty/taglib/ogg/vorbis/vorbisproperties.cpp
+++ b/3rdparty/taglib/ogg/vorbis/vorbisproperties.cpp
@@ -133,7 +133,7 @@ void Vorbis::Properties::read()
 
   ByteVector data = d->file->packet(0);
 
-  int pos = 0;
+  uint pos = 0;
 
   if(data.mid(pos, 7) != vorbisSetupHeaderID) {
     debug("Vorbis::Properties::read() -- invalid Vorbis identification header");
@@ -142,22 +142,22 @@ void Vorbis::Properties::read()
 
   pos += 7;
 
-  d->vorbisVersion = data.mid(pos, 4).toUInt(false);
+  d->vorbisVersion = data.toUInt(pos, false);
   pos += 4;
 
   d->channels = uchar(data[pos]);
   pos += 1;
 
-  d->sampleRate = data.mid(pos, 4).toUInt(false);
+  d->sampleRate = data.toUInt(pos, false);
   pos += 4;
 
-  d->bitrateMaximum = data.mid(pos, 4).toUInt(false);
+  d->bitrateMaximum = data.toUInt(pos, false);
   pos += 4;
 
-  d->bitrateNominal = data.mid(pos, 4).toUInt(false);
+  d->bitrateNominal = data.toUInt(pos, false);
   pos += 4;
 
-  d->bitrateMinimum = data.mid(pos, 4).toUInt(false);
+  d->bitrateMinimum = data.toUInt(pos, false);
 
   // TODO: Later this should be only the "fast" mode.
   d->bitrate = d->bitrateNominal;
diff --git a/3rdparty/taglib/ogg/xiphcomment.cpp b/3rdparty/taglib/ogg/xiphcomment.cpp
index 675614b3f..51c2f9f04 100644
--- a/3rdparty/taglib/ogg/xiphcomment.cpp
+++ b/3rdparty/taglib/ogg/xiphcomment.cpp
@@ -340,7 +340,7 @@ void Ogg::XiphComment::parse(const ByteVector &data)
 
   uint pos = 0;
 
-  uint vendorLength = data.mid(0, 4).toUInt(false);
+  const uint vendorLength = data.toUInt(0, false);
   pos += 4;
 
   d->vendorID = String(data.mid(pos, vendorLength), String::UTF8);
@@ -348,7 +348,7 @@ void Ogg::XiphComment::parse(const ByteVector &data)
 
   // Next the number of fields in the comment vector.
 
-  uint commentFields = data.mid(pos, 4).toUInt(false);
+  const uint commentFields = data.toUInt(pos, false);
   pos += 4;
 
   if(commentFields > (data.size() - 8) / 4) {
@@ -360,7 +360,7 @@ void Ogg::XiphComment::parse(const ByteVector &data)
     // Each comment field is in the format "KEY=value" in a UTF8 string and has
     // 4 bytes before the text starts that gives the length.
 
-    uint commentLength = data.mid(pos, 4).toUInt(false);
+    const uint commentLength = data.toUInt(pos, false);
     pos += 4;
 
     String comment = String(data.mid(pos, commentLength), String::UTF8);
diff --git a/3rdparty/taglib/riff/aiff/aifffile.cpp b/3rdparty/taglib/riff/aiff/aifffile.cpp
index ece84f058..d20c148be 100644
--- a/3rdparty/taglib/riff/aiff/aifffile.cpp
+++ b/3rdparty/taglib/riff/aiff/aifffile.cpp
@@ -90,6 +90,11 @@ PropertyMap RIFF::AIFF::File::properties() const
   return d->tag->properties();
 }
 
+void RIFF::AIFF::File::removeUnsupportedProperties(const StringList &unsupported)
+{
+  d->tag->removeUnsupportedProperties(unsupported);
+}
+
 PropertyMap RIFF::AIFF::File::setProperties(const PropertyMap &properties)
 {
   return d->tag->setProperties(properties);
@@ -118,6 +123,7 @@ bool RIFF::AIFF::File::save()
   return true;
 }
 
+
 ////////////////////////////////////////////////////////////////////////////////
 // private members
 ////////////////////////////////////////////////////////////////////////////////
diff --git a/3rdparty/taglib/riff/aiff/aifffile.h b/3rdparty/taglib/riff/aiff/aifffile.h
index e1284db08..c75b94b4c 100644
--- a/3rdparty/taglib/riff/aiff/aifffile.h
+++ b/3rdparty/taglib/riff/aiff/aifffile.h
@@ -58,20 +58,22 @@ namespace TagLib {
       {
       public:
         /*!
-         * Contructs an AIFF file from \a file.  If \a readProperties is true the
-         * file's audio properties will also be read using \a propertiesStyle.  If
-         * false, \a propertiesStyle is ignored.
+         * Constructs an AIFF file from \a file.  If \a readProperties is true the
+         * file's audio properties will also be read.
+         *
+         * \note In the current implementation, \a propertiesStyle is ignored.
          */
         File(FileName file, bool readProperties = true,
              Properties::ReadStyle propertiesStyle = Properties::Average);
 
         /*!
-         * Contructs an AIFF file from \a file.  If \a readProperties is true the
-         * file's audio properties will also be read using \a propertiesStyle.  If
-         * false, \a propertiesStyle is ignored.
+         * Constructs an AIFF file from \a stream.  If \a readProperties is true the
+         * file's audio properties will also be read.
          *
          * \note TagLib will *not* take ownership of the stream, the caller is
          * responsible for deleting it after the File object.
+         *
+         * \note In the current implementation, \a propertiesStyle is ignored.
          */
         File(IOStream *stream, bool readProperties = true,
              Properties::ReadStyle propertiesStyle = Properties::Average);
@@ -83,6 +85,12 @@ namespace TagLib {
 
         /*!
          * Returns the Tag for this file.
+         *
+         * \note This always returns a valid pointer regardless of whether or not 
+         * the file on disk has an ID3v2 tag.  Use hasID3v2Tag() to check if the file 
+         * on disk actually has an ID3v2 tag.
+         *
+         * \see hasID3v2Tag()
          */
         virtual ID3v2::Tag *tag() const;
 
@@ -92,6 +100,8 @@ namespace TagLib {
          */
         PropertyMap properties() const;
 
+        void removeUnsupportedProperties(const StringList &properties);
+
         /*!
          * Implements the unified property interface -- import function.
          * This method forwards to ID3v2::Tag::setProperties().
@@ -109,6 +119,13 @@ namespace TagLib {
          */
         virtual bool save();
 
+        /*!
+         * Returns whether or not the file on disk actually has an ID3v2 tag.
+         *
+         * \see ID3v2Tag()
+         */
+        bool hasID3v2Tag() const;
+
       private:
         File(const File &);
         File &operator=(const File &);
diff --git a/3rdparty/taglib/riff/aiff/aiffproperties.cpp b/3rdparty/taglib/riff/aiff/aiffproperties.cpp
index 3ea275825..18b973eb0 100644
--- a/3rdparty/taglib/riff/aiff/aiffproperties.cpp
+++ b/3rdparty/taglib/riff/aiff/aiffproperties.cpp
@@ -39,7 +39,7 @@
 
 #define UnsignedToFloat(u) (((double)((long)(u - 2147483647L - 1))) + 2147483648.0)
 
-static double ConvertFromIeeeExtended(unsigned char *bytes)
+static double ConvertFromIeeeExtended(const TagLib::uchar *bytes)
 {
   double f;
   int expon;
@@ -150,10 +150,10 @@ TagLib::uint RIFF::AIFF::Properties::sampleFrames() const
 
 void RIFF::AIFF::Properties::read(const ByteVector &data)
 {
-  d->channels       = data.mid(0, 2).toShort();
-  d->sampleFrames   = data.mid(2, 4).toUInt();
-  d->sampleWidth    = data.mid(6, 2).toShort();
-  double sampleRate = ConvertFromIeeeExtended(reinterpret_cast<unsigned char *>(data.mid(8, 10).data()));
+  d->channels       = data.toShort(0U);
+  d->sampleFrames   = data.toUInt(2U);
+  d->sampleWidth    = data.toShort(6U);
+  double sampleRate = ConvertFromIeeeExtended(reinterpret_cast<const uchar *>(data.data() + 8));
   d->sampleRate     = (int)sampleRate;
   d->bitrate        = (int)((sampleRate * d->sampleWidth * d->channels) / 1000.0);
   d->length         = d->sampleRate > 0 ? d->sampleFrames / d->sampleRate : 0;
diff --git a/3rdparty/taglib/riff/rifffile.cpp b/3rdparty/taglib/riff/rifffile.cpp
index df3b36635..20eec5ffe 100644
--- a/3rdparty/taglib/riff/rifffile.cpp
+++ b/3rdparty/taglib/riff/rifffile.cpp
@@ -138,34 +138,49 @@ ByteVector RIFF::File::chunkData(uint i)
   return readBlock(d->chunks[i].size);
 }
 
+void RIFF::File::setChunkData(uint i, const ByteVector &data)
+{
+  // First we update the global size
+
+  d->size += ((data.size() + 1) & ~1) - (d->chunks[i].size + d->chunks[i].padding);
+  insert(ByteVector::fromUInt(d->size, d->endianness == BigEndian), 4, 4);
+
+  // Now update the specific chunk
+
+  writeChunk(chunkName(i), data, d->chunks[i].offset - 8, d->chunks[i].size + d->chunks[i].padding + 8);
+
+  d->chunks[i].size = data.size();
+  d->chunks[i].padding = (data.size() & 0x01) ? 1 : 0;
+
+  // Now update the internal offsets
+
+  for(i++; i < d->chunks.size(); i++)
+    d->chunks[i].offset = d->chunks[i-1].offset + 8 + d->chunks[i-1].size + d->chunks[i-1].padding;
+}
+
 void RIFF::File::setChunkData(const ByteVector &name, const ByteVector &data)
+{
+  setChunkData(name, data, false);
+}
+
+void RIFF::File::setChunkData(const ByteVector &name, const ByteVector &data, bool alwaysCreate)
 {
   if(d->chunks.size() == 0) {
     debug("RIFF::File::setChunkData - No valid chunks found.");
     return;
   }
 
-  for(uint i = 0; i < d->chunks.size(); i++) {
-    if(d->chunks[i].name == name) {
+  if(alwaysCreate && name != "LIST") {
+    debug("RIFF::File::setChunkData - alwaysCreate should be used for only \"LIST\" chunks.");
+    return;
+  }
 
-      // First we update the global size
-
-      d->size += ((data.size() + 1) & ~1) - (d->chunks[i].size + d->chunks[i].padding);
-      insert(ByteVector::fromUInt(d->size, d->endianness == BigEndian), 4, 4);
-
-      // Now update the specific chunk
-
-      writeChunk(name, data, d->chunks[i].offset - 8, d->chunks[i].size + d->chunks[i].padding + 8);
-
-      d->chunks[i].size = data.size();
-      d->chunks[i].padding = (data.size() & 0x01) ? 1 : 0;
-
-      // Now update the internal offsets
-
-      for(i++; i < d->chunks.size(); i++)
-        d->chunks[i].offset = d->chunks[i-1].offset + 8 + d->chunks[i-1].size + d->chunks[i-1].padding;
-
-      return;
+  if(!alwaysCreate) {
+    for(uint i = 0; i < d->chunks.size(); i++) {
+      if(d->chunks[i].name == name) {
+        setChunkData(i, data);
+        return;
+      }
     }
   }
 
@@ -181,7 +196,7 @@ void RIFF::File::setChunkData(const ByteVector &name, const ByteVector &data)
 
   // Now add the chunk to the file
 
-  writeChunk(name, data, offset, std::max(ulong(0), length() - offset), (offset & 1) ? 1 : 0);
+  writeChunk(name, data, offset, std::max<long>(0, length() - offset), (offset & 1) ? 1 : 0);
 
   // And update our internal structure
 
@@ -199,6 +214,28 @@ void RIFF::File::setChunkData(const ByteVector &name, const ByteVector &data)
   d->chunks.push_back(chunk);
 }
 
+void RIFF::File::removeChunk(uint i)
+{
+  if(i >= d->chunks.size())
+    return;
+  
+  removeBlock(d->chunks[i].offset - 8, d->chunks[i].size + 8);
+  d->chunks.erase(d->chunks.begin() + i);
+}
+
+void RIFF::File::removeChunk(const ByteVector &name)
+{
+  std::vector<Chunk> newChunks;
+  for(size_t i = 0; i < d->chunks.size(); ++i) {
+    if(d->chunks[i].name == name)
+      removeBlock(d->chunks[i].offset - 8, d->chunks[i].size + 8);
+    else
+      newChunks.push_back(d->chunks[i]);
+  }
+
+  d->chunks.swap(newChunks);
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // private members
 ////////////////////////////////////////////////////////////////////////////////
diff --git a/3rdparty/taglib/riff/rifffile.h b/3rdparty/taglib/riff/rifffile.h
index e418dbb67..6b8fe197f 100644
--- a/3rdparty/taglib/riff/rifffile.h
+++ b/3rdparty/taglib/riff/rifffile.h
@@ -95,6 +95,13 @@ namespace TagLib {
        */
       ByteVector chunkData(uint i);
 
+      /*!
+       * Sets the data for the the specified chunk to \a data. 
+       *
+       * \warning This will update the file immediately.
+       */
+      void setChunkData(uint i, const ByteVector &data);
+
       /*!
        * Sets the data for the chunk \a name to \a data.  If a chunk with the
        * given name already exists it will be overwritten, otherwise it will be
@@ -104,6 +111,34 @@ namespace TagLib {
        */
       void setChunkData(const ByteVector &name, const ByteVector &data);
 
+      /*!
+       * Sets the data for the chunk \a name to \a data.  If a chunk with the
+       * given name already exists it will be overwritten, otherwise it will be
+       * created after the existing chunks.
+       *
+       * \note If \a alwaysCreate is true, a new chunk is created regardless of 
+       * whether or not the chunk \a name exists. It should only be used for 
+       * "LIST" chunks. 
+       *
+       * \warning This will update the file immediately.
+       */
+      void setChunkData(const ByteVector &name, const ByteVector &data, bool alwaysCreate);
+
+      /*!
+       * Removes the specified chunk.
+       *
+       * \warning This will update the file immediately.
+       */
+      void removeChunk(uint i);
+
+      /*!
+       * Removes the chunk \a name.
+       *
+       * \warning This will update the file immediately.
+       * \warning This removes all the chunks with the given name.
+       */
+      void removeChunk(const ByteVector &name);
+
     private:
       File(const File &);
       File &operator=(const File &);
diff --git a/3rdparty/taglib/riff/wav/infotag.cpp b/3rdparty/taglib/riff/wav/infotag.cpp
new file mode 100644
index 000000000..658ede886
--- /dev/null
+++ b/3rdparty/taglib/riff/wav/infotag.cpp
@@ -0,0 +1,261 @@
+/***************************************************************************
+    copyright            : (C) 2012 by Tsuda Kageyu
+    email                : wheeler@kde.org
+ ***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#include <tdebug.h>
+#include <tfile.h>
+
+#include "infotag.h"
+
+using namespace TagLib;
+using namespace RIFF::Info;
+
+namespace {
+  static bool isValidChunkID(const ByteVector &name)
+  {
+    if(name.size() != 4)
+      return false;
+
+    for(int i = 0; i < 4; i++) {
+      if(name[i] < 32 || name[i] > 127)
+        return false;
+    }
+
+    return true;
+  }
+}
+
+class RIFF::Info::Tag::TagPrivate
+{
+public:
+  TagPrivate() 
+  {}
+
+  FieldListMap fieldListMap;
+
+  static const StringHandler *stringHandler;
+};
+
+////////////////////////////////////////////////////////////////////////////////
+// StringHandler implementation
+////////////////////////////////////////////////////////////////////////////////
+
+StringHandler::StringHandler()
+{
+}
+
+StringHandler::~StringHandler()
+{
+}
+
+String RIFF::Info::StringHandler::parse(const ByteVector &data) const
+{
+  return String(data, String::UTF8);
+}
+
+ByteVector RIFF::Info::StringHandler::render(const String &s) const
+{
+  return s.data(String::UTF8);
+}
+
+////////////////////////////////////////////////////////////////////////////////
+// public members
+////////////////////////////////////////////////////////////////////////////////
+
+static const StringHandler defaultStringHandler;
+const RIFF::Info::StringHandler *RIFF::Info::Tag::TagPrivate::stringHandler = &defaultStringHandler;
+
+RIFF::Info::Tag::Tag(const ByteVector &data) 
+  : TagLib::Tag()
+  , d(new TagPrivate())
+{
+  parse(data);
+}
+
+RIFF::Info::Tag::Tag() 
+  : TagLib::Tag()
+  , d(new TagPrivate())
+{
+}
+
+RIFF::Info::Tag::~Tag()
+{
+  delete d;
+}
+
+String RIFF::Info::Tag::title() const
+{
+  return fieldText("INAM");
+}
+
+String RIFF::Info::Tag::artist() const
+{
+  return fieldText("IART");
+}
+
+String RIFF::Info::Tag::album() const
+{
+  return fieldText("IPRD");
+}
+
+String RIFF::Info::Tag::comment() const
+{
+  return fieldText("ICMT");
+}
+
+String RIFF::Info::Tag::genre() const
+{
+  return fieldText("IGNR");
+}
+
+TagLib::uint RIFF::Info::Tag::year() const
+{
+  return fieldText("ICRD").substr(0, 4).toInt();
+}
+
+TagLib::uint RIFF::Info::Tag::track() const
+{
+  return fieldText("IPRT").toInt();
+}
+
+void RIFF::Info::Tag::setTitle(const String &s)
+{
+  setFieldText("INAM", s);
+}
+
+void RIFF::Info::Tag::setArtist(const String &s)
+{
+  setFieldText("IART", s);
+}
+
+void RIFF::Info::Tag::setAlbum(const String &s)
+{
+  setFieldText("IPRD", s);
+}
+
+void RIFF::Info::Tag::setComment(const String &s)
+{
+  setFieldText("ICMT", s);
+}
+
+void RIFF::Info::Tag::setGenre(const String &s)
+{
+  setFieldText("IGNR", s);
+}
+
+void RIFF::Info::Tag::setYear(uint i)
+{
+  if(i != 0)
+    setFieldText("ICRD", String::number(i));
+  else
+    d->fieldListMap.erase("ICRD");
+}
+
+void RIFF::Info::Tag::setTrack(uint i)
+{
+  if(i != 0)
+    setFieldText("IPRT", String::number(i));
+  else
+    d->fieldListMap.erase("IPRT");
+}
+
+bool RIFF::Info::Tag::isEmpty() const
+{
+  return d->fieldListMap.isEmpty();
+}
+
+String RIFF::Info::Tag::fieldText(const ByteVector &id) const
+{
+  if(d->fieldListMap.contains(id))
+    return String(d->fieldListMap[id]);
+  else
+    return String();
+}
+
+void RIFF::Info::Tag::setFieldText(const ByteVector &id, const String &s)
+{
+  // id must be four-byte long pure ascii string.
+  if(!isValidChunkID(id))
+    return;
+
+  if(!s.isEmpty())
+    d->fieldListMap[id] = s;
+  else
+    removeField(id);
+}
+
+void RIFF::Info::Tag::removeField(const ByteVector &id)
+{
+  if(d->fieldListMap.contains(id))
+    d->fieldListMap.erase(id);
+}
+
+ByteVector RIFF::Info::Tag::render() const
+{
+  ByteVector data("INFO");
+
+  FieldListMap::ConstIterator it = d->fieldListMap.begin();
+  for(; it != d->fieldListMap.end(); ++it) {
+    ByteVector text = TagPrivate::stringHandler->render(it->second);
+    if(text.isEmpty())
+      continue;
+
+    data.append(it->first);
+    data.append(ByteVector::fromUInt(text.size() + 1, false));
+    data.append(text);
+    
+    do {
+      data.append('\0');
+    } while(data.size() & 1);
+  }
+
+  if(data.size() == 4)
+    return ByteVector();
+  else
+    return data;
+}
+
+void RIFF::Info::Tag::setStringHandler(const StringHandler *handler)
+{
+  if(handler)
+    TagPrivate::stringHandler = handler;
+  else
+    TagPrivate::stringHandler = &defaultStringHandler;
+}
+
+////////////////////////////////////////////////////////////////////////////////
+// protected members
+////////////////////////////////////////////////////////////////////////////////
+
+void RIFF::Info::Tag::parse(const ByteVector &data)
+{
+  uint p = 4;
+  while(p < data.size()) {
+    const uint size = data.toUInt(p + 4, false);
+    d->fieldListMap[data.mid(p, 4)] = TagPrivate::stringHandler->parse(data.mid(p + 8, size));
+
+    p += ((size + 1) & ~1) + 8;
+  }
+}
+
diff --git a/3rdparty/taglib/riff/wav/infotag.h b/3rdparty/taglib/riff/wav/infotag.h
new file mode 100644
index 000000000..9a26d4cb4
--- /dev/null
+++ b/3rdparty/taglib/riff/wav/infotag.h
@@ -0,0 +1,180 @@
+/***************************************************************************
+    copyright            : (C) 2012 by Tsuda Kageyu
+    email                : tsuda.kageyu@gmail.com
+ ***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#ifndef TAGLIB_INFOTAG_H
+#define TAGLIB_INFOTAG_H
+
+#include "tag.h"
+#include "tmap.h"
+#include "tstring.h"
+#include "tstringlist.h"
+#include "tbytevector.h"
+#include "taglib_export.h"
+
+namespace TagLib {
+
+  class File;
+
+  //! A RIFF Info tag implementation. 
+  namespace RIFF {
+  namespace Info {
+
+    typedef Map<ByteVector, String> FieldListMap;
+
+    //! A abstraction for the string to data encoding in Info tags.
+
+    /*!
+     * RIFF Info tag has no clear definitions about character encodings.
+     * In practice, local encoding of each system is largely used and UTF-8 is
+     * popular too.
+     *
+     * Here is an option to read and write tags in your preferrd encoding 
+     * by subclassing this class, reimplementing parse() and render() and setting 
+     * your reimplementation as the default with Info::Tag::setStringHandler().
+     *
+     * \see ID3v1::Tag::setStringHandler()
+     */
+
+    class TAGLIB_EXPORT StringHandler
+    {
+    public:
+      StringHandler();
+      ~StringHandler();
+
+      /*!
+       * Decode a string from \a data.  The default implementation assumes that
+       * \a data is an UTF-8 character array.
+       */
+      virtual String parse(const ByteVector &data) const;
+
+      /*!
+       * Encode a ByteVector with the data from \a s.  The default implementation
+       * assumes that \a s is an UTF-8 string. 
+       */
+      virtual ByteVector render(const String &s) const;
+    };
+
+    //! The main class in the ID3v2 implementation
+
+    /*!
+     * This is the main class in the INFO tag implementation.  RIFF INFO tag is a 
+     * metadata format found in WAV audio and AVI video files.  Though it is a part 
+     * of Microsoft/IBM's RIFF specification, the author could not find the official 
+     * documents about it.  So, this implementation is referring to unofficial documents 
+     * online and some applications' behaviors especially Windows Explorer.
+     */
+    class TAGLIB_EXPORT Tag : public TagLib::Tag
+    {
+    public:
+      /*!
+       * Constructs an empty Info tag.
+       */
+      Tag();
+
+      /*!
+       * Constructs an Info tag read from \a data which is contents of "LIST" chunk.
+       */
+      Tag(const ByteVector &data);
+
+      virtual ~Tag();
+
+      // Reimplementations
+
+      virtual String title() const;
+      virtual String artist() const;
+      virtual String album() const;
+      virtual String comment() const;
+      virtual String genre() const;
+      virtual uint year() const;
+      virtual uint track() const;
+
+      virtual void setTitle(const String &s);
+      virtual void setArtist(const String &s);
+      virtual void setAlbum(const String &s);
+      virtual void setComment(const String &s);
+      virtual void setGenre(const String &s);
+      virtual void setYear(uint i);
+      virtual void setTrack(uint i);
+
+      virtual bool isEmpty() const;
+
+      /*
+       * Gets the value of the field with the ID \a id.
+       */
+      String fieldText(const ByteVector &id) const;
+        
+      /*
+        * Sets the value of the field with the ID \a id to \a s.
+        * If the field does not exist, it is created.
+        * If \s is empty, the field is removed.
+        *
+        * \note fieldId must be four-byte long pure ASCII string.  This function 
+        * performs nothing if fieldId is invalid.
+        */
+      void setFieldText(const ByteVector &id, const String &s);
+
+      /*
+       * Removes the field with the ID \a id.
+       */
+      void removeField(const ByteVector &id);
+
+      /*!
+       * Render the tag back to binary data, suitable to be written to disk.
+       *
+       * \note Returns empty ByteVector is the tag contains no fields. 
+       */
+      ByteVector render() const;
+
+      /*!
+       * Sets the string handler that decides how the text data will be
+       * converted to and from binary data.
+       * If the parameter \a handler is null, the previous handler is
+       * released and default UTF-8 handler is restored.
+       *
+       * \note The caller is responsible for deleting the previous handler
+       * as needed after it is released.
+       *
+       * \see StringHandler
+       */
+      static void setStringHandler(const StringHandler *handler);
+    
+    protected:
+      /*!
+       * Pareses the body of the tag in \a data.
+       */
+      void parse(const ByteVector &data);
+
+
+    private:
+      Tag(const Tag &);
+      Tag &operator=(const Tag &);
+
+      class TagPrivate;
+      TagPrivate *d;
+    };
+  }}
+}
+
+#endif
diff --git a/3rdparty/taglib/riff/wav/wavfile.cpp b/3rdparty/taglib/riff/wav/wavfile.cpp
index 613db4ef9..abecea351 100644
--- a/3rdparty/taglib/riff/wav/wavfile.cpp
+++ b/3rdparty/taglib/riff/wav/wavfile.cpp
@@ -23,36 +23,47 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#include <tbytevector.h>
-#include <tdebug.h>
-#include <id3v2tag.h>
-#include <tstringlist.h>
-#include <tpropertymap.h>
+#include "tbytevector.h"
+#include "tdebug.h"
+#include "tstringlist.h"
+#include "tpropertymap.h"
 
 #include "wavfile.h"
+#include "id3v2tag.h"
+#include "infotag.h"
+#include "tagunion.h"
 
 using namespace TagLib;
 
+namespace
+{
+  enum { ID3v2Index = 0, InfoIndex = 1 };
+}
+
 class RIFF::WAV::File::FilePrivate
 {
 public:
   FilePrivate() :
     properties(0),
-    tag(0),
-    tagChunkID("ID3 ")
+    tagChunkID("ID3 "),
+    hasID3v2(false),
+    hasInfo(false)
   {
-
   }
 
   ~FilePrivate()
   {
     delete properties;
-    delete tag;
   }
 
   Properties *properties;
-  ID3v2::Tag *tag;
+  
   ByteVector tagChunkID;
+
+  TagUnion tag;
+
+  bool hasID3v2;
+  bool hasInfo;
 };
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -82,26 +93,45 @@ RIFF::WAV::File::~File()
 
 ID3v2::Tag *RIFF::WAV::File::tag() const
 {
-  return d->tag;
+  return ID3v2Tag();
+}
+
+ID3v2::Tag *RIFF::WAV::File::ID3v2Tag() const
+{
+  return d->tag.access<ID3v2::Tag>(ID3v2Index, false);
+}
+
+RIFF::Info::Tag *RIFF::WAV::File::InfoTag() const
+{
+  return d->tag.access<RIFF::Info::Tag>(InfoIndex, false);
 }
 
 PropertyMap RIFF::WAV::File::properties() const
 {
-  return d->tag->properties();
+  return tag()->properties();
+}
+
+void RIFF::WAV::File::removeUnsupportedProperties(const StringList &unsupported)
+{
+  tag()->removeUnsupportedProperties(unsupported);
 }
 
 PropertyMap RIFF::WAV::File::setProperties(const PropertyMap &properties)
 {
-  return d->tag->setProperties(properties);
+  return tag()->setProperties(properties);
 }
 
-
 RIFF::WAV::Properties *RIFF::WAV::File::audioProperties() const
 {
   return d->properties;
 }
 
 bool RIFF::WAV::File::save()
+{
+  return RIFF::WAV::File::save(AllTags);
+}
+
+bool RIFF::WAV::File::save(TagTypes tags, bool stripOthers, int id3v2Version)
 {
   if(readOnly()) {
     debug("RIFF::WAV::File::save() -- File is read only.");
@@ -113,11 +143,43 @@ bool RIFF::WAV::File::save()
     return false;
   }
 
-  setChunkData(d->tagChunkID, d->tag->render());
+  if(stripOthers)
+    strip(static_cast<TagTypes>(AllTags & ~tags));
+
+  ID3v2::Tag *id3v2tag = d->tag.access<ID3v2::Tag>(ID3v2Index, false);
+  if(!id3v2tag->isEmpty()) {
+    if(tags & ID3v2) {
+      setChunkData(d->tagChunkID, id3v2tag->render(id3v2Version));
+      d->hasID3v2 = true;
+    }
+  }
+
+  Info::Tag *infotag = d->tag.access<Info::Tag>(InfoIndex, false);
+  if(!infotag->isEmpty()) {
+    if(tags & Info) {
+      int chunkId = findInfoTagChunk();
+      if(chunkId != -1)
+        setChunkData(chunkId, infotag->render());
+      else
+        setChunkData("LIST", infotag->render(), true);
+
+      d->hasInfo = true;
+    }
+  }
 
   return true;
 }
 
+bool RIFF::WAV::File::hasID3v2Tag() const
+{
+  return d->hasID3v2;
+}
+
+bool RIFF::WAV::File::hasInfoTag() const
+{
+  return d->hasInfo;
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // private members
 ////////////////////////////////////////////////////////////////////////////////
@@ -127,19 +189,56 @@ void RIFF::WAV::File::read(bool readProperties, Properties::ReadStyle properties
   ByteVector formatData;
   uint streamLength = 0;
   for(uint i = 0; i < chunkCount(); i++) {
-    if(chunkName(i) == "ID3 " || chunkName(i) == "id3 ") {
+    String name = chunkName(i);
+    if(name == "ID3 " || name == "id3 ") {
       d->tagChunkID = chunkName(i);
-      d->tag = new ID3v2::Tag(this, chunkOffset(i));
+      d->tag.set(ID3v2Index, new ID3v2::Tag(this, chunkOffset(i)));
+      d->hasID3v2 = true;
     }
-    else if(chunkName(i) == "fmt " && readProperties)
+    else if(name == "fmt " && readProperties)
       formatData = chunkData(i);
-    else if(chunkName(i) == "data" && readProperties)
+    else if(name == "data" && readProperties)
       streamLength = chunkDataSize(i);
+    else if(name == "LIST") {
+      ByteVector data = chunkData(i);
+      ByteVector type = data.mid(0, 4);
+
+      if(type == "INFO") {
+        d->tag.set(InfoIndex, new RIFF::Info::Tag(data));
+        d->hasInfo = true;
+      }
+    }
   }
 
+  if (!d->tag[ID3v2Index])
+    d->tag.set(ID3v2Index, new ID3v2::Tag);
+
+  if (!d->tag[InfoIndex])
+    d->tag.set(InfoIndex, new RIFF::Info::Tag);
+
   if(!formatData.isEmpty())
     d->properties = new Properties(formatData, streamLength, propertiesStyle);
-
-  if(!d->tag)
-    d->tag = new ID3v2::Tag;
+}
+
+void RIFF::WAV::File::strip(TagTypes tags)
+{
+  if(tags & ID3v2)
+    removeChunk(d->tagChunkID);
+
+  if(tags & Info){
+    TagLib::uint chunkId = findInfoTagChunk();
+    if(chunkId != TagLib::uint(-1))
+      removeChunk(chunkId);
+  }
+}
+
+TagLib::uint RIFF::WAV::File::findInfoTagChunk()
+{
+  for(uint i = 0; i < chunkCount(); ++i) {
+    if(chunkName(i) == "LIST" && chunkData(i).mid(0, 4) == "INFO") {
+      return i;
+    }
+  }
+  
+  return TagLib::uint(-1);
 }
diff --git a/3rdparty/taglib/riff/wav/wavfile.h b/3rdparty/taglib/riff/wav/wavfile.h
index 1c4708707..2f83332a2 100644
--- a/3rdparty/taglib/riff/wav/wavfile.h
+++ b/3rdparty/taglib/riff/wav/wavfile.h
@@ -28,6 +28,7 @@
 
 #include "rifffile.h"
 #include "id3v2tag.h"
+#include "infotag.h"
 #include "wavproperties.h"
 
 namespace TagLib {
@@ -57,21 +58,34 @@ namespace TagLib {
       class TAGLIB_EXPORT File : public TagLib::RIFF::File
       {
       public:
+        enum TagTypes {
+          //! Empty set.  Matches no tag types.
+          NoTags  = 0x0000,
+          //! Matches ID3v2 tags.
+          ID3v2   = 0x0001,
+          //! Matches Info tags.
+          Info    = 0x0002,
+          //! Matches all tag types.
+          AllTags = 0xffff
+        };
+
         /*!
-         * Contructs an WAV file from \a file.  If \a readProperties is true the
-         * file's audio properties will also be read using \a propertiesStyle.  If
-         * false, \a propertiesStyle is ignored.
+         * Constructs a WAV file from \a file.  If \a readProperties is true the
+         * file's audio properties will also be read.
+         *
+         * \note In the current implementation, \a propertiesStyle is ignored.
          */
         File(FileName file, bool readProperties = true,
              Properties::ReadStyle propertiesStyle = Properties::Average);
 
         /*!
-         * Contructs an WAV file from \a file.  If \a readProperties is true the
-         * file's audio properties will also be read using \a propertiesStyle.  If
-         * false, \a propertiesStyle is ignored.
+         * Constructs a WAV file from \a stream.  If \a readProperties is true the
+         * file's audio properties will also be read.
          *
          * \note TagLib will *not* take ownership of the stream, the caller is
          * responsible for deleting it after the File object.
+         *
+         * \note In the current implementation, \a propertiesStyle is ignored.
          */
         File(IOStream *stream, bool readProperties = true,
              Properties::ReadStyle propertiesStyle = Properties::Average);
@@ -82,9 +96,34 @@ namespace TagLib {
         virtual ~File();
 
         /*!
-         * Returns the Tag for this file.
+         * Returns the ID3v2 Tag for this file.
+         * 
+         * \note This method does not return all the tags for this file for 
+         * backward compatibility.  Will be fixed in TagLib 2.0.
          */
-        virtual ID3v2::Tag *tag() const;
+        ID3v2::Tag *tag() const;
+
+        /*!
+         * Returns the ID3v2 Tag for this file.
+         *
+         * \note This always returns a valid pointer regardless of whether or not 
+         * the file on disk has an ID3v2 tag.  Use hasID3v2Tag() to check if the 
+         * file on disk actually has an ID3v2 tag.
+         *
+         * \see hasID3v2Tag()
+         */
+        ID3v2::Tag *ID3v2Tag() const;
+
+        /*!
+         * Returns the RIFF INFO Tag for this file.
+         *
+         * \note This always returns a valid pointer regardless of whether or not 
+         * the file on disk has a RIFF INFO tag.  Use hasInfoTag() to check if the 
+         * file on disk actually has a RIFF INFO tag.
+         *
+         * \see hasInfoTag()
+         */
+        Info::Tag *InfoTag() const;
 
         /*!
          * Implements the unified property interface -- export function.
@@ -92,6 +131,8 @@ namespace TagLib {
          */
         PropertyMap properties() const;
 
+        void removeUnsupportedProperties(const StringList &properties);
+
         /*!
          * Implements the unified property interface -- import function.
          * This method forwards to ID3v2::Tag::setProperties().
@@ -109,12 +150,35 @@ namespace TagLib {
          */
         virtual bool save();
 
+        bool save(TagTypes tags, bool stripOthers = true, int id3v2Version = 4);
+        
+        /*!
+         * Returns whether or not the file on disk actually has an ID3v2 tag.
+         *
+         * \see ID3v2Tag()
+         */
+        bool hasID3v2Tag() const;
+
+        /*!
+         * Returns whether or not the file on disk actually has a RIFF INFO tag.
+         *
+         * \see InfoTag()
+         */
+        bool hasInfoTag() const;
+
       private:
         File(const File &);
         File &operator=(const File &);
 
         void read(bool readProperties, Properties::ReadStyle propertiesStyle);
 
+        void strip(TagTypes tags);
+
+        /*!
+         * Returns the index of the chunk that its name is "LIST" and list type is "INFO".
+         */
+        uint findInfoTagChunk();
+
         class FilePrivate;
         FilePrivate *d;
       };
diff --git a/3rdparty/taglib/riff/wav/wavproperties.cpp b/3rdparty/taglib/riff/wav/wavproperties.cpp
index 9e7ea70cc..8062df5f7 100644
--- a/3rdparty/taglib/riff/wav/wavproperties.cpp
+++ b/3rdparty/taglib/riff/wav/wavproperties.cpp
@@ -115,12 +115,12 @@ TagLib::uint RIFF::WAV::Properties::sampleFrames() const
 
 void RIFF::WAV::Properties::read(const ByteVector &data)
 {
-  d->format = data.mid(0, 2).toShort(false);
-  d->channels = data.mid(2, 2).toShort(false);
-  d->sampleRate = data.mid(4, 4).toUInt(false);
-  d->sampleWidth = data.mid(14, 2).toShort(false);
+  d->format      = data.toShort(0, false);
+  d->channels    = data.toShort(2, false);
+  d->sampleRate  = data.toUInt(4, false);
+  d->sampleWidth = data.toShort(14, false);
 
-  uint byteRate = data.mid(8, 4).toUInt(false);
+  const uint byteRate = data.toUInt(8, false);
   d->bitrate = byteRate * 8 / 1000;
 
   d->length = byteRate > 0 ? d->streamLength / byteRate : 0;
diff --git a/3rdparty/taglib/s3m/s3mfile.cpp b/3rdparty/taglib/s3m/s3mfile.cpp
index 7ffdf910b..371340a59 100644
--- a/3rdparty/taglib/s3m/s3mfile.cpp
+++ b/3rdparty/taglib/s3m/s3mfile.cpp
@@ -47,7 +47,8 @@ S3M::File::File(FileName file, bool readProperties,
   Mod::FileBase(file),
   d(new FilePrivate(propertiesStyle))
 {
-  read(readProperties);
+  if(isOpen())
+    read(readProperties);
 }
 
 S3M::File::File(IOStream *stream, bool readProperties,
@@ -55,7 +56,8 @@ S3M::File::File(IOStream *stream, bool readProperties,
   Mod::FileBase(stream),
   d(new FilePrivate(propertiesStyle))
 {
-  read(readProperties);
+  if(isOpen())
+    read(readProperties);
 }
 
 S3M::File::~File()
diff --git a/3rdparty/taglib/s3m/s3mfile.h b/3rdparty/taglib/s3m/s3mfile.h
index c862108ee..5f4529772 100644
--- a/3rdparty/taglib/s3m/s3mfile.h
+++ b/3rdparty/taglib/s3m/s3mfile.h
@@ -36,18 +36,22 @@ namespace TagLib {
     class TAGLIB_EXPORT File : public Mod::FileBase {
       public:
         /*!
-         * Contructs a ScreamTracker III file from \a file. If \a readProperties
-         * is true the file's audio properties will also be read using
-         * \a propertiesStyle. If false, \a propertiesStyle is ignored.
+         * Constructs a ScreamTracker III from \a file.
+         *
+         * \note In the current implementation, both \a readProperties and
+         * \a propertiesStyle are ignored.  The audio properties are always
+         * read.
          */
         File(FileName file, bool readProperties = true,
              AudioProperties::ReadStyle propertiesStyle =
              AudioProperties::Average);
 
         /*!
-         * Contructs a ScreamTracker III file from \a stream. If \a readProperties
-         * is true the file's audio properties will also be read using
-         * \a propertiesStyle. If false, \a propertiesStyle is ignored.
+         * Constructs a ScreamTracker III file from \a stream.
+         *
+         * \note In the current implementation, both \a readProperties and
+         * \a propertiesStyle are ignored.  The audio properties are always
+         * read.
          *
          * \note TagLib will *not* take ownership of the stream, the caller is
          * responsible for deleting it after the File object.
diff --git a/3rdparty/taglib/taglib_export.h b/3rdparty/taglib/taglib_export.h
index 3e8685517..737ae6442 100644
--- a/3rdparty/taglib/taglib_export.h
+++ b/3rdparty/taglib/taglib_export.h
@@ -40,8 +40,4 @@
 #define TAGLIB_EXPORT
 #endif
 
-#ifndef TAGLIB_NO_CONFIG
-#include "taglib_config.h"
-#endif
-
 #endif
diff --git a/3rdparty/taglib/toolkit/taglib.h b/3rdparty/taglib/toolkit/taglib.h
index dda9c83c2..f93f280d6 100644
--- a/3rdparty/taglib/toolkit/taglib.h
+++ b/3rdparty/taglib/toolkit/taglib.h
@@ -26,8 +26,10 @@
 #ifndef TAGLIB_H
 #define TAGLIB_H
 
+#include "taglib_config.h"
+
 #define TAGLIB_MAJOR_VERSION 1
-#define TAGLIB_MINOR_VERSION 7
+#define TAGLIB_MINOR_VERSION 8
 #define TAGLIB_PATCH_VERSION 0
 
 #if defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ > 1))
@@ -44,23 +46,6 @@
 
 #include <string>
 
-#ifdef __APPLE__
-#  include <libkern/OSAtomic.h>
-#  define TAGLIB_ATOMIC_MAC
-#elif defined(WIN32) || defined(_WIN32) || defined(__WIN32__) || defined(__CYGWIN__)
-#  define NOMINMAX
-#  include <windows.h>
-#  define TAGLIB_ATOMIC_WIN
-#elif defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 401)    \
-      && (defined(__i386__) || defined(__i486__) || defined(__i586__) || \
-          defined(__i686__) || defined(__x86_64) || defined(__ia64)) \
-      && !defined(__INTEL_COMPILER)
-#  define TAGLIB_ATOMIC_GCC
-#elif defined(__ia64) && defined(__INTEL_COMPILER)
-#  include <ia64intrin.h>
-#  define TAGLIB_ATOMIC_GCC
-#endif
-
 //! A namespace for all TagLib related classes and functions
 
 /*!
@@ -75,10 +60,13 @@ namespace TagLib {
 
   class String;
 
-  typedef wchar_t wchar;
-  typedef unsigned char  uchar;
-  typedef unsigned short ushort;
-  typedef unsigned int   uint;
+  typedef wchar_t            wchar;   // Assumed to be sufficient to store a UTF-16 char.
+  typedef unsigned char      uchar;
+  typedef unsigned short     ushort;
+  typedef unsigned int       uint;
+  typedef unsigned long long ulonglong;
+
+  // long/ulong can be either 32-bit or 64-bit wide.
   typedef unsigned long  ulong;
 
   /*!
@@ -86,50 +74,6 @@ namespace TagLib {
    * so I'm providing something here that should be constant.
    */
   typedef std::basic_string<wchar> wstring;
-
-#ifndef DO_NOT_DOCUMENT // Tell Doxygen to skip this class.
-  /*!
-   * \internal
-   * This is just used as a base class for shared classes in TagLib.
-   *
-   * \warning This <b>is not</b> part of the TagLib public API!
-   */
-
-  class RefCounter
-  {
-  public:
-    RefCounter() : refCount(1) {}
-
-#ifdef TAGLIB_ATOMIC_MAC
-    void ref() { OSAtomicIncrement32Barrier(const_cast<int32_t*>(&refCount)); }
-    bool deref() { return ! OSAtomicDecrement32Barrier(const_cast<int32_t*>(&refCount)); }
-    int32_t count() { return refCount; }
-  private:
-    volatile int32_t refCount;
-#elif defined(TAGLIB_ATOMIC_WIN)
-    void ref() { InterlockedIncrement(&refCount); }
-    bool deref() { return ! InterlockedDecrement(&refCount); }
-    long count() { return refCount; }
-  private:
-    volatile long refCount;
-#elif defined(TAGLIB_ATOMIC_GCC)
-    void ref() { __sync_add_and_fetch(&refCount, 1); }
-    bool deref() { return ! __sync_sub_and_fetch(&refCount, 1); }
-    int count() { return refCount; }
-  private:
-    volatile int refCount;
-#else
-    void ref() { refCount++; }
-    bool deref() { return ! --refCount; }
-    int count() { return refCount; }
-  private:
-    uint refCount;
-#endif
-
-  };
-
-#endif // DO_NOT_DOCUMENT
-
 }
 
 /*!
@@ -143,7 +87,7 @@ namespace TagLib {
  * - A clean, high level, C++ API to handling audio meta data.
  * - Format specific APIs for advanced API users.
  * - ID3v1, ID3v2, APE, FLAC, Xiph, iTunes-style MP4 and WMA tag formats.
- * - MP3, MPC, FLAC, MP4, ASF, AIFF, WAV, TrueAudio, WavPack, Ogg FLAC, Ogg Vorbis and Speex file formats.
+ * - MP3, MPC, FLAC, MP4, ASF, AIFF, WAV, TrueAudio, WavPack, Ogg FLAC, Ogg Vorbis, Speex and Opus file formats.
  * - Basic audio file properties such as length, sample rate, etc.
  * - Long term binary and source compatibility.
  * - Extensible design, notably the ability to add other formats or extend current formats as a library user.
diff --git a/3rdparty/taglib/toolkit/tbytevector.cpp b/3rdparty/taglib/toolkit/tbytevector.cpp
index 59e3a9ed3..ddc3d4f43 100644
--- a/3rdparty/taglib/toolkit/tbytevector.cpp
+++ b/3rdparty/taglib/toolkit/tbytevector.cpp
@@ -23,230 +23,327 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif
+
 #include <iostream>
+#include <cstdio>
+#include <cstring>
 
 #include <tstring.h>
 #include <tdebug.h>
-
-#include <string.h>
+#include "trefcounter.h"
+#include "tutils.h"
 
 #include "tbytevector.h"
 
 // This is a bit ugly to keep writing over and over again.
 
 // A rather obscure feature of the C++ spec that I hadn't thought of that makes
-// working with C libs much more effecient.  There's more here:
+// working with C libs much more efficient.  There's more here:
 //
 // http://www.informit.com/isapi/product_id~{9C84DAB4-FE6E-49C5-BB0A-FB50331233EA}/content/index.asp
 
-#define DATA(x) (&(x->data[0]))
+#define DATA(x) (&(x->data->data[0]))
 
 namespace TagLib {
-  static const char hexTable[17] = "0123456789abcdef";
 
-  static const uint crcTable[256] = {
-    0x00000000, 0x04c11db7, 0x09823b6e, 0x0d4326d9, 0x130476dc, 0x17c56b6b,
-    0x1a864db2, 0x1e475005, 0x2608edb8, 0x22c9f00f, 0x2f8ad6d6, 0x2b4bcb61,
-    0x350c9b64, 0x31cd86d3, 0x3c8ea00a, 0x384fbdbd, 0x4c11db70, 0x48d0c6c7,
-    0x4593e01e, 0x4152fda9, 0x5f15adac, 0x5bd4b01b, 0x569796c2, 0x52568b75,
-    0x6a1936c8, 0x6ed82b7f, 0x639b0da6, 0x675a1011, 0x791d4014, 0x7ddc5da3,
-    0x709f7b7a, 0x745e66cd, 0x9823b6e0, 0x9ce2ab57, 0x91a18d8e, 0x95609039,
-    0x8b27c03c, 0x8fe6dd8b, 0x82a5fb52, 0x8664e6e5, 0xbe2b5b58, 0xbaea46ef,
-    0xb7a96036, 0xb3687d81, 0xad2f2d84, 0xa9ee3033, 0xa4ad16ea, 0xa06c0b5d,
-    0xd4326d90, 0xd0f37027, 0xddb056fe, 0xd9714b49, 0xc7361b4c, 0xc3f706fb,
-    0xceb42022, 0xca753d95, 0xf23a8028, 0xf6fb9d9f, 0xfbb8bb46, 0xff79a6f1,
-    0xe13ef6f4, 0xe5ffeb43, 0xe8bccd9a, 0xec7dd02d, 0x34867077, 0x30476dc0,
-    0x3d044b19, 0x39c556ae, 0x278206ab, 0x23431b1c, 0x2e003dc5, 0x2ac12072,
-    0x128e9dcf, 0x164f8078, 0x1b0ca6a1, 0x1fcdbb16, 0x018aeb13, 0x054bf6a4,
-    0x0808d07d, 0x0cc9cdca, 0x7897ab07, 0x7c56b6b0, 0x71159069, 0x75d48dde,
-    0x6b93dddb, 0x6f52c06c, 0x6211e6b5, 0x66d0fb02, 0x5e9f46bf, 0x5a5e5b08,
-    0x571d7dd1, 0x53dc6066, 0x4d9b3063, 0x495a2dd4, 0x44190b0d, 0x40d816ba,
-    0xaca5c697, 0xa864db20, 0xa527fdf9, 0xa1e6e04e, 0xbfa1b04b, 0xbb60adfc,
-    0xb6238b25, 0xb2e29692, 0x8aad2b2f, 0x8e6c3698, 0x832f1041, 0x87ee0df6,
-    0x99a95df3, 0x9d684044, 0x902b669d, 0x94ea7b2a, 0xe0b41de7, 0xe4750050,
-    0xe9362689, 0xedf73b3e, 0xf3b06b3b, 0xf771768c, 0xfa325055, 0xfef34de2,
-    0xc6bcf05f, 0xc27dede8, 0xcf3ecb31, 0xcbffd686, 0xd5b88683, 0xd1799b34,
-    0xdc3abded, 0xd8fba05a, 0x690ce0ee, 0x6dcdfd59, 0x608edb80, 0x644fc637,
-    0x7a089632, 0x7ec98b85, 0x738aad5c, 0x774bb0eb, 0x4f040d56, 0x4bc510e1,
-    0x46863638, 0x42472b8f, 0x5c007b8a, 0x58c1663d, 0x558240e4, 0x51435d53,
-    0x251d3b9e, 0x21dc2629, 0x2c9f00f0, 0x285e1d47, 0x36194d42, 0x32d850f5,
-    0x3f9b762c, 0x3b5a6b9b, 0x0315d626, 0x07d4cb91, 0x0a97ed48, 0x0e56f0ff,
-    0x1011a0fa, 0x14d0bd4d, 0x19939b94, 0x1d528623, 0xf12f560e, 0xf5ee4bb9,
-    0xf8ad6d60, 0xfc6c70d7, 0xe22b20d2, 0xe6ea3d65, 0xeba91bbc, 0xef68060b,
-    0xd727bbb6, 0xd3e6a601, 0xdea580d8, 0xda649d6f, 0xc423cd6a, 0xc0e2d0dd,
-    0xcda1f604, 0xc960ebb3, 0xbd3e8d7e, 0xb9ff90c9, 0xb4bcb610, 0xb07daba7,
-    0xae3afba2, 0xaafbe615, 0xa7b8c0cc, 0xa379dd7b, 0x9b3660c6, 0x9ff77d71,
-    0x92b45ba8, 0x9675461f, 0x8832161a, 0x8cf30bad, 0x81b02d74, 0x857130c3,
-    0x5d8a9099, 0x594b8d2e, 0x5408abf7, 0x50c9b640, 0x4e8ee645, 0x4a4ffbf2,
-    0x470cdd2b, 0x43cdc09c, 0x7b827d21, 0x7f436096, 0x7200464f, 0x76c15bf8,
-    0x68860bfd, 0x6c47164a, 0x61043093, 0x65c52d24, 0x119b4be9, 0x155a565e,
-    0x18197087, 0x1cd86d30, 0x029f3d35, 0x065e2082, 0x0b1d065b, 0x0fdc1bec,
-    0x3793a651, 0x3352bbe6, 0x3e119d3f, 0x3ad08088, 0x2497d08d, 0x2056cd3a,
-    0x2d15ebe3, 0x29d4f654, 0xc5a92679, 0xc1683bce, 0xcc2b1d17, 0xc8ea00a0,
-    0xd6ad50a5, 0xd26c4d12, 0xdf2f6bcb, 0xdbee767c, 0xe3a1cbc1, 0xe760d676,
-    0xea23f0af, 0xeee2ed18, 0xf0a5bd1d, 0xf464a0aa, 0xf9278673, 0xfde69bc4,
-    0x89b8fd09, 0x8d79e0be, 0x803ac667, 0x84fbdbd0, 0x9abc8bd5, 0x9e7d9662,
-    0x933eb0bb, 0x97ffad0c, 0xafb010b1, 0xab710d06, 0xa6322bdf, 0xa2f33668,
-    0xbcb4666d, 0xb8757bda, 0xb5365d03, 0xb1f740b4
-  };
+static const char hexTable[17] = "0123456789abcdef";
 
-  /*!
-   * A templatized KMP find that works both with a ByteVector and a ByteVectorMirror.
-   */
-
-  template <class Vector>
-  int vectorFind(const Vector &v, const Vector &pattern, uint offset, int byteAlign)
-  {
-    if(pattern.size() > v.size() || offset > v.size() - 1)
-      return -1;
-
-    // Let's go ahead and special case a pattern of size one since that's common
-    // and easy to make fast.
-
-    if(pattern.size() == 1) {
-      char p = pattern[0];
-      for(uint i = offset; i < v.size(); i++) {
-        if(v[i] == p && (i - offset) % byteAlign == 0)
-          return i;
-      }
-      return -1;
-    }
-
-    uchar lastOccurrence[256];
-
-    for(uint i = 0; i < 256; ++i)
-      lastOccurrence[i] = uchar(pattern.size());
-
-    for(uint i = 0; i < pattern.size() - 1; ++i)
-      lastOccurrence[uchar(pattern[i])] = uchar(pattern.size() - i - 1);
-
-    for(uint i = pattern.size() - 1 + offset; i < v.size(); i += lastOccurrence[uchar(v.at(i))]) {
-      int iBuffer = i;
-      int iPattern = pattern.size() - 1;
-
-      while(iPattern >= 0 && v.at(iBuffer) == pattern[iPattern]) {
-        --iBuffer;
-        --iPattern;
-      }
-
-      if(-1 == iPattern && (iBuffer + 1 - offset) % byteAlign == 0)
-        return iBuffer + 1;
-    }
+static const uint crcTable[256] = {
+  0x00000000, 0x04c11db7, 0x09823b6e, 0x0d4326d9, 0x130476dc, 0x17c56b6b,
+  0x1a864db2, 0x1e475005, 0x2608edb8, 0x22c9f00f, 0x2f8ad6d6, 0x2b4bcb61,
+  0x350c9b64, 0x31cd86d3, 0x3c8ea00a, 0x384fbdbd, 0x4c11db70, 0x48d0c6c7,
+  0x4593e01e, 0x4152fda9, 0x5f15adac, 0x5bd4b01b, 0x569796c2, 0x52568b75,
+  0x6a1936c8, 0x6ed82b7f, 0x639b0da6, 0x675a1011, 0x791d4014, 0x7ddc5da3,
+  0x709f7b7a, 0x745e66cd, 0x9823b6e0, 0x9ce2ab57, 0x91a18d8e, 0x95609039,
+  0x8b27c03c, 0x8fe6dd8b, 0x82a5fb52, 0x8664e6e5, 0xbe2b5b58, 0xbaea46ef,
+  0xb7a96036, 0xb3687d81, 0xad2f2d84, 0xa9ee3033, 0xa4ad16ea, 0xa06c0b5d,
+  0xd4326d90, 0xd0f37027, 0xddb056fe, 0xd9714b49, 0xc7361b4c, 0xc3f706fb,
+  0xceb42022, 0xca753d95, 0xf23a8028, 0xf6fb9d9f, 0xfbb8bb46, 0xff79a6f1,
+  0xe13ef6f4, 0xe5ffeb43, 0xe8bccd9a, 0xec7dd02d, 0x34867077, 0x30476dc0,
+  0x3d044b19, 0x39c556ae, 0x278206ab, 0x23431b1c, 0x2e003dc5, 0x2ac12072,
+  0x128e9dcf, 0x164f8078, 0x1b0ca6a1, 0x1fcdbb16, 0x018aeb13, 0x054bf6a4,
+  0x0808d07d, 0x0cc9cdca, 0x7897ab07, 0x7c56b6b0, 0x71159069, 0x75d48dde,
+  0x6b93dddb, 0x6f52c06c, 0x6211e6b5, 0x66d0fb02, 0x5e9f46bf, 0x5a5e5b08,
+  0x571d7dd1, 0x53dc6066, 0x4d9b3063, 0x495a2dd4, 0x44190b0d, 0x40d816ba,
+  0xaca5c697, 0xa864db20, 0xa527fdf9, 0xa1e6e04e, 0xbfa1b04b, 0xbb60adfc,
+  0xb6238b25, 0xb2e29692, 0x8aad2b2f, 0x8e6c3698, 0x832f1041, 0x87ee0df6,
+  0x99a95df3, 0x9d684044, 0x902b669d, 0x94ea7b2a, 0xe0b41de7, 0xe4750050,
+  0xe9362689, 0xedf73b3e, 0xf3b06b3b, 0xf771768c, 0xfa325055, 0xfef34de2,
+  0xc6bcf05f, 0xc27dede8, 0xcf3ecb31, 0xcbffd686, 0xd5b88683, 0xd1799b34,
+  0xdc3abded, 0xd8fba05a, 0x690ce0ee, 0x6dcdfd59, 0x608edb80, 0x644fc637,
+  0x7a089632, 0x7ec98b85, 0x738aad5c, 0x774bb0eb, 0x4f040d56, 0x4bc510e1,
+  0x46863638, 0x42472b8f, 0x5c007b8a, 0x58c1663d, 0x558240e4, 0x51435d53,
+  0x251d3b9e, 0x21dc2629, 0x2c9f00f0, 0x285e1d47, 0x36194d42, 0x32d850f5,
+  0x3f9b762c, 0x3b5a6b9b, 0x0315d626, 0x07d4cb91, 0x0a97ed48, 0x0e56f0ff,
+  0x1011a0fa, 0x14d0bd4d, 0x19939b94, 0x1d528623, 0xf12f560e, 0xf5ee4bb9,
+  0xf8ad6d60, 0xfc6c70d7, 0xe22b20d2, 0xe6ea3d65, 0xeba91bbc, 0xef68060b,
+  0xd727bbb6, 0xd3e6a601, 0xdea580d8, 0xda649d6f, 0xc423cd6a, 0xc0e2d0dd,
+  0xcda1f604, 0xc960ebb3, 0xbd3e8d7e, 0xb9ff90c9, 0xb4bcb610, 0xb07daba7,
+  0xae3afba2, 0xaafbe615, 0xa7b8c0cc, 0xa379dd7b, 0x9b3660c6, 0x9ff77d71,
+  0x92b45ba8, 0x9675461f, 0x8832161a, 0x8cf30bad, 0x81b02d74, 0x857130c3,
+  0x5d8a9099, 0x594b8d2e, 0x5408abf7, 0x50c9b640, 0x4e8ee645, 0x4a4ffbf2,
+  0x470cdd2b, 0x43cdc09c, 0x7b827d21, 0x7f436096, 0x7200464f, 0x76c15bf8,
+  0x68860bfd, 0x6c47164a, 0x61043093, 0x65c52d24, 0x119b4be9, 0x155a565e,
+  0x18197087, 0x1cd86d30, 0x029f3d35, 0x065e2082, 0x0b1d065b, 0x0fdc1bec,
+  0x3793a651, 0x3352bbe6, 0x3e119d3f, 0x3ad08088, 0x2497d08d, 0x2056cd3a,
+  0x2d15ebe3, 0x29d4f654, 0xc5a92679, 0xc1683bce, 0xcc2b1d17, 0xc8ea00a0,
+  0xd6ad50a5, 0xd26c4d12, 0xdf2f6bcb, 0xdbee767c, 0xe3a1cbc1, 0xe760d676,
+  0xea23f0af, 0xeee2ed18, 0xf0a5bd1d, 0xf464a0aa, 0xf9278673, 0xfde69bc4,
+  0x89b8fd09, 0x8d79e0be, 0x803ac667, 0x84fbdbd0, 0x9abc8bd5, 0x9e7d9662,
+  0x933eb0bb, 0x97ffad0c, 0xafb010b1, 0xab710d06, 0xa6322bdf, 0xa2f33668,
+  0xbcb4666d, 0xb8757bda, 0xb5365d03, 0xb1f740b4
+};
 
+/*!
+  * A templatized straightforward find that works with the types 
+  * std::vector<char>::iterator and std::vector<char>::reverse_iterator.
+  */
+template <class TIterator>
+int findChar(
+  const TIterator dataBegin, const TIterator dataEnd,
+  char c, uint offset, int byteAlign)
+{
+  const size_t dataSize = dataEnd - dataBegin;
+  if(dataSize == 0 || offset > dataSize - 1)
     return -1;
+
+  // n % 0 is invalid
+
+  if(byteAlign == 0)
+    return -1;
+
+  for(TIterator it = dataBegin + offset; it < dataEnd; it += byteAlign) {
+    if(*it == c)
+      return (it - dataBegin);
   }
 
-  /*!
-   * Wraps the accessors to a ByteVector to make the search algorithm access the
-   * elements in reverse.
-   *
-   * \see vectorFind()
-   * \see ByteVector::rfind()
-   */
-
-  class ByteVectorMirror
-  {
-  public:
-    ByteVectorMirror(const ByteVector &source) : v(source) {}
-
-    char operator[](int index) const
-    {
-      return v[v.size() - index - 1];
-    }
-
-    char at(int index) const
-    {
-      return v.at(v.size() - index - 1);
-    }
-
-    ByteVectorMirror mid(uint index, uint length = 0xffffffff) const
-    {
-      return length == 0xffffffff ? v.mid(0, index) : v.mid(index - length, length);
-    }
-
-    uint size() const
-    {
-      return v.size();
-    }
-
-    int find(const ByteVectorMirror &pattern, uint offset = 0, int byteAlign = 1) const
-    {
-      ByteVectorMirror v(*this);
-
-      if(offset > 0) {
-        offset = size() - offset - pattern.size();
-        if(offset >= size())
-          offset = 0;
-      }
-
-      const int pos = vectorFind<ByteVectorMirror>(v, pattern, offset, byteAlign);
-
-      // If the offset is zero then we need to adjust the location in the search
-      // to be appropriately reversed.  If not we need to account for the fact
-      // that the recursive call (called from the above line) has already ajusted
-      // for this but that the normal templatized find above will add the offset
-      // to the returned value.
-      //
-      // This is a little confusing at first if you don't first stop to think
-      // through the logic involved in the forward search.
-
-      if(pos == -1)
-        return -1;
-
-      return size() - pos - pattern.size();
-    }
-
-  private:
-    const ByteVector &v;
-  };
-
-  template <class T>
-  T toNumber(const std::vector<char> &data, bool mostSignificantByteFirst)
-  {
-    T sum = 0;
-
-    if(data.size() <= 0) {
-      debug("ByteVectorMirror::toNumber<T>() -- data is empty, returning 0");
-      return sum;
-    }
-
-    uint size = sizeof(T);
-    uint last = data.size() > size ? size - 1 : data.size() - 1;
-
-    for(uint i = 0; i <= last; i++)
-      sum |= (T) uchar(data[i]) << ((mostSignificantByteFirst ? last - i : i) * 8);
-
-    return sum;
-  }
-
-  template <class T>
-  ByteVector fromNumber(T value, bool mostSignificantByteFirst)
-  {
-    int size = sizeof(T);
-
-    ByteVector v(size, 0);
-
-    for(int i = 0; i < size; i++)
-      v[i] = uchar(value >> ((mostSignificantByteFirst ? size - 1 - i : i) * 8) & 0xff);
-
-    return v;
-  }
+  return -1;
 }
 
-using namespace TagLib;
+/*!
+  * A templatized KMP find that works with the types 
+  * std::vector<char>::iterator and std::vector<char>::reverse_iterator.
+  */
+template <class TIterator>
+int findVector(
+  const TIterator dataBegin, const TIterator dataEnd,
+  const TIterator patternBegin, const TIterator patternEnd,
+  uint offset, int byteAlign)
+{
+  const size_t dataSize    = dataEnd    - dataBegin;
+  const size_t patternSize = patternEnd - patternBegin;
+  if(patternSize > dataSize || offset > dataSize - 1)
+    return -1;
+
+  // n % 0 is invalid
+
+  if(byteAlign == 0)
+    return -1;
+
+  // Special case that pattern contains just single char.
+
+  if(patternSize == 1)
+    return findChar(dataBegin, dataEnd, *patternBegin, offset, byteAlign);
+
+  size_t lastOccurrence[256];
+
+  for(size_t i = 0; i < 256; ++i)
+    lastOccurrence[i] = patternSize;
+
+  for(size_t i = 0; i < patternSize - 1; ++i)
+    lastOccurrence[static_cast<uchar>(*(patternBegin + i))] = patternSize - i - 1;
+
+  TIterator it = dataBegin + patternSize - 1 + offset;
+  while(true)
+  {
+    TIterator itBuffer = it;
+    TIterator itPattern = patternBegin + patternSize - 1;
+
+    while(*itBuffer == *itPattern)
+    {
+      if(itPattern == patternBegin)
+      {
+        if((itBuffer - dataBegin - offset) % byteAlign == 0)
+          return (itBuffer - dataBegin);
+        else
+          break;
+      }
+
+      --itBuffer;
+      --itPattern;
+    }
+
+    const size_t step = lastOccurrence[static_cast<uchar>(*it)];
+    if(dataEnd - step <= it)
+      break;
+
+    it += step;
+  }
+
+  return -1;
+}
+
+template <class T>
+T toNumber(const ByteVector &v, size_t offset, size_t length, bool mostSignificantByteFirst)
+{
+  if(offset >= v.size()) {
+    debug("toNumber<T>() -- No data to convert. Returning 0.");
+    return 0;
+  }
+
+  length = std::min(length, v.size() - offset);
+
+  T sum = 0;
+  for(size_t i = 0; i < length; i++) {
+    const size_t shift = (mostSignificantByteFirst ? length - 1 - i : i) * 8;
+    sum |= static_cast<T>(static_cast<uchar>(v[offset + i])) << shift;
+  }
+
+  return sum;
+}
+
+template <class T>
+T toNumber(const ByteVector &v, size_t offset, bool mostSignificantByteFirst)
+{
+  if(offset + sizeof(T) > v.size()) 
+    return toNumber<T>(v, offset, v.size() - offset, mostSignificantByteFirst);
+
+  // Uses memcpy instead of reinterpret_cast to avoid an alignment exception.
+  T tmp;
+  ::memcpy(&tmp, v.data() + offset, sizeof(T));
+
+#if SYSTEM_BYTEORDER == 1
+  const bool swap = mostSignificantByteFirst;
+#else
+  const bool swap != mostSignificantByteFirst;
+#endif
+  if(swap)
+    return byteSwap(tmp);
+  else
+    return tmp;
+}
+
+template <class T>
+ByteVector fromNumber(T value, bool mostSignificantByteFirst)
+{
+  const size_t size = sizeof(T);
+
+#if SYSTEM_BYTEORDER == 1
+  const bool swap = mostSignificantByteFirst;
+#else
+  const bool swap != mostSignificantByteFirst;
+#endif
+ if(swap)
+    value = byteSwap(value);
+
+  return ByteVector(reinterpret_cast<const char *>(&value), size);
+}
+
+class DataPrivate : public RefCounter
+{
+public:
+  DataPrivate()
+  {
+  }
+
+  DataPrivate(const std::vector<char> &v, uint offset, uint length) 
+    : data(v.begin() + offset, v.begin() + offset + length) 
+  {
+  }
+
+  // A char* can be an iterator.
+  DataPrivate(const char *begin, const char *end)
+    : data(begin, end)
+  {
+  }
+
+  DataPrivate(uint len, char c) 
+    : data(len, c) 
+  {
+  }
+
+  std::vector<char> data;
+};
 
 class ByteVector::ByteVectorPrivate : public RefCounter
 {
 public:
-  ByteVectorPrivate() : RefCounter(), size(0) {}
-  ByteVectorPrivate(const std::vector<char> &v) : RefCounter(), data(v), size(v.size()) {}
-  ByteVectorPrivate(TagLib::uint len, char value) : RefCounter(), data(len, value), size(len) {}
+  ByteVectorPrivate() 
+    : RefCounter()
+    , data(new DataPrivate())
+    , offset(0)
+    , length(0) 
+  {
+  }
 
-  std::vector<char> data;
+  ByteVectorPrivate(ByteVectorPrivate *d, uint o, uint l)
+    : RefCounter()
+    , data(d->data)
+    , offset(d->offset + o)
+    , length(l)
+  {
+    data->ref();
+  }
 
-  // std::vector<T>::size() is very slow, so we'll cache the value
+  ByteVectorPrivate(const std::vector<char> &v, uint o, uint l)
+    : RefCounter()
+    , data(new DataPrivate(v, o, l))
+    , offset(0)
+    , length(l)
+  {
+  }
 
-  uint size;
+  ByteVectorPrivate(uint l, char c) 
+    : RefCounter()
+    , data(new DataPrivate(l, c))
+    , offset(0)
+    , length(l)
+  {
+  }
+
+  ByteVectorPrivate(const char *s, uint l) 
+    : RefCounter()
+    , data(new DataPrivate(s, s + l))
+    , offset(0)
+    , length(l)
+  {
+  }
+  
+  void detach()
+  {
+    if(data->count() > 1) {
+      data->deref();
+      data = new DataPrivate(data->data, offset, length);
+      offset = 0;
+    }
+  }
+
+  ~ByteVectorPrivate()
+  {
+    if(data->deref())
+      delete data;
+  }
+
+  ByteVectorPrivate &operator=(const ByteVectorPrivate &x)
+  {
+    if(&x != this)
+    {
+      if(data->deref())
+        delete data;
+
+      data = x.data;
+      data->ref();
+    }
+
+    return *this;
+  }
+
+  DataPrivate *data;
+  uint offset;
+  uint length;
 };
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -257,14 +354,10 @@ ByteVector ByteVector::null;
 
 ByteVector ByteVector::fromCString(const char *s, uint length)
 {
-  ByteVector v;
-
   if(length == 0xffffffff)
-    v.setData(s);
+    return ByteVector(s, ::strlen(s));
   else
-    v.setData(s, length);
-
-  return v;
+    return ByteVector(s, length);
 }
 
 ByteVector ByteVector::fromUInt(uint value, bool mostSignificantByteFirst)
@@ -274,12 +367,12 @@ ByteVector ByteVector::fromUInt(uint value, bool mostSignificantByteFirst)
 
 ByteVector ByteVector::fromShort(short value, bool mostSignificantByteFirst)
 {
-  return fromNumber<short>(value, mostSignificantByteFirst);
+  return fromNumber<ushort>(value, mostSignificantByteFirst);
 }
 
 ByteVector ByteVector::fromLongLong(long long value, bool mostSignificantByteFirst)
 {
-  return fromNumber<long long>(value, mostSignificantByteFirst);
+  return fromNumber<unsigned long long>(value, mostSignificantByteFirst);
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -287,37 +380,39 @@ ByteVector ByteVector::fromLongLong(long long value, bool mostSignificantByteFir
 ////////////////////////////////////////////////////////////////////////////////
 
 ByteVector::ByteVector()
+  : d(new ByteVectorPrivate())
 {
-  d = new ByteVectorPrivate;
 }
 
 ByteVector::ByteVector(uint size, char value)
+  : d(new ByteVectorPrivate(size, value))
 {
-  d = new ByteVectorPrivate(size, value);
 }
 
-ByteVector::ByteVector(const ByteVector &v) : d(v.d)
+ByteVector::ByteVector(const ByteVector &v) 
+  : d(v.d)
 {
   d->ref();
 }
 
-ByteVector::ByteVector(char c)
+ByteVector::ByteVector(const ByteVector &v, uint offset, uint length)
+  : d(new ByteVectorPrivate(v.d, offset, length))
+{
+}
+
+ByteVector::ByteVector(char c)
+  : d(new ByteVectorPrivate(1, c))
 {
-  d = new ByteVectorPrivate;
-  d->data.push_back(c);
-  d->size = 1;
 }
 
 ByteVector::ByteVector(const char *data, uint length)
+  : d(new ByteVectorPrivate(data, length))
 {
-  d = new ByteVectorPrivate;
-  setData(data, length);
 }
 
 ByteVector::ByteVector(const char *data)
+  : d(new ByteVectorPrivate(data, ::strlen(data)))
 {
-  d = new ByteVectorPrivate;
-  setData(data);
 }
 
 ByteVector::~ByteVector()
@@ -326,74 +421,68 @@ ByteVector::~ByteVector()
     delete d;
 }
 
-ByteVector &ByteVector::setData(const char *data, uint length)
+ByteVector &ByteVector::setData(const char *s, uint length)
 {
-  detach();
-
-  resize(length);
-
-  if(length > 0)
-    ::memcpy(DATA(d), data, length);
-
+  *this = ByteVector(s, length);
   return *this;
 }
 
 ByteVector &ByteVector::setData(const char *data)
 {
-  return setData(data, ::strlen(data));
+  *this = ByteVector(data);
+  return *this;
 }
 
 char *ByteVector::data()
 {
   detach();
-  return size() > 0 ? DATA(d) : 0;
+  return size() > 0 ? (DATA(d) + d->offset) : 0;
 }
 
 const char *ByteVector::data() const
 {
-  return size() > 0 ? DATA(d) : 0;
+  return size() > 0 ? (DATA(d) + d->offset) : 0;
 }
 
 ByteVector ByteVector::mid(uint index, uint length) const
 {
-  ByteVector v;
+  index  = std::min(index, size());
+  length = std::min(length, size() - index);
 
-  if(index > size())
-    return v;
-
-  ConstIterator endIt;
-
-  if(length < size() - index)
-    endIt = d->data.begin() + index + length;
-  else
-    endIt = d->data.end();
-
-  v.d->data.insert(v.d->data.begin(), ConstIterator(d->data.begin() + index), endIt);
-  v.d->size = v.d->data.size();
-
-  return v;
+  return ByteVector(*this, index, length);
 }
 
 char ByteVector::at(uint index) const
 {
-  return index < size() ? d->data[index] : 0;
+  return index < size() ? DATA(d)[d->offset + index] : 0;
 }
 
 int ByteVector::find(const ByteVector &pattern, uint offset, int byteAlign) const
 {
-  return vectorFind<ByteVector>(*this, pattern, offset, byteAlign);
+  return findVector<ConstIterator>(
+    begin(), end(), pattern.begin(), pattern.end(), offset, byteAlign);
+}
+
+int ByteVector::find(char c, uint offset, int byteAlign) const
+{
+  return findChar<ConstIterator>(begin(), end(), c, offset, byteAlign);
 }
 
 int ByteVector::rfind(const ByteVector &pattern, uint offset, int byteAlign) const
 {
-  // Ok, this is a little goofy, but pretty cool after it sinks in.  Instead of
-  // reversing the find method's Boyer-Moore search algorithm I created a "mirror"
-  // for a ByteVector to reverse the behavior of the accessors.
+  if(offset > 0) {
+    offset = size() - offset - pattern.size();
+    if(offset >= size())
+      offset = 0;
+  }
 
-  ByteVectorMirror v(*this);
-  ByteVectorMirror p(pattern);
+  const int pos = findVector<ConstReverseIterator>(
+    rbegin(), rend(), pattern.rbegin(), pattern.rend(), offset, byteAlign);
 
-  return v.find(p, offset, byteAlign);
+  if(pos == -1)
+    return -1;
+  else
+    return size() - pos - pattern.size();
 }
 
 bool ByteVector::containsAt(const ByteVector &pattern, uint offset, uint patternOffset, uint patternLength) const
@@ -402,18 +491,11 @@ bool ByteVector::containsAt(const ByteVector &pattern, uint offset, uint pattern
     patternLength = pattern.size();
 
   // do some sanity checking -- all of these things are needed for the search to be valid
-
-  if(patternLength > size() || offset >= size() || patternOffset >= pattern.size() || patternLength == 0)
+  const uint compareLength = patternLength - patternOffset;
+  if(offset + compareLength > size() || patternOffset >= pattern.size() || patternLength == 0)    
     return false;
-
-  // loop through looking for a mismatch
-
-  for(uint i = 0; i < patternLength - patternOffset; i++) {
-    if(at(i + offset) != pattern[i + patternOffset])
-      return false;
-  }
-
-  return true;
+  
+  return (::memcmp(data() + offset, pattern.data() + patternOffset, compareLength) == 0);
 }
 
 bool ByteVector::startsWith(const ByteVector &pattern) const
@@ -511,74 +593,92 @@ int ByteVector::endsWithPartialMatch(const ByteVector &pattern) const
 
 ByteVector &ByteVector::append(const ByteVector &v)
 {
-  if(v.d->size == 0)
-    return *this; // Simply return if appending nothing.
+  if(v.d->length != 0)
+  {
+    detach();
 
-  detach();
-
-  uint originalSize = d->size;
-  resize(d->size + v.d->size);
-  ::memcpy(DATA(d) + originalSize, DATA(v.d), v.size());
+    uint originalSize = size();
+    resize(originalSize + v.size());
+    ::memcpy(data() + originalSize, v.data(), v.size());
+  }
 
   return *this;
 }
 
 ByteVector &ByteVector::clear()
 {
-  detach();
-  d->data.clear();
-  d->size = 0;
-
+  *this = ByteVector();
   return *this;
 }
 
 TagLib::uint ByteVector::size() const
 {
-  return d->size;
+  return d->length;
 }
 
 ByteVector &ByteVector::resize(uint size, char padding)
 {
-  if(d->size < size) {
-    d->data.reserve(size);
-    d->data.insert(d->data.end(), size - d->size, padding);
+  if(size != d->length) {
+    detach();
+    d->data->data.resize(d->offset + size, padding);
+    d->length = size;
   }
-  else
-    d->data.erase(d->data.begin() + size, d->data.end());
-
-  d->size = size;
 
   return *this;
 }
 
 ByteVector::Iterator ByteVector::begin()
 {
-  return d->data.begin();
+  return d->data->data.begin() + d->offset;
 }
 
 ByteVector::ConstIterator ByteVector::begin() const
 {
-  return d->data.begin();
+  return d->data->data.begin() + d->offset;
 }
 
 ByteVector::Iterator ByteVector::end()
 {
-  return d->data.end();
+  return d->data->data.begin() + d->offset + d->length;
 }
 
 ByteVector::ConstIterator ByteVector::end() const
 {
-  return d->data.end();
+  return d->data->data.begin() + d->offset + d->length;
+}
+
+ByteVector::ReverseIterator ByteVector::rbegin()
+{
+  std::vector<char> &v = d->data->data;
+  return v.rbegin() + (v.size() - (d->offset + d->length));
+}
+
+ByteVector::ConstReverseIterator ByteVector::rbegin() const
+{
+  std::vector<char> &v = d->data->data;
+  return v.rbegin() + (v.size() - (d->offset + d->length));
+}
+
+ByteVector::ReverseIterator ByteVector::rend()
+{
+  std::vector<char> &v = d->data->data;
+  return v.rbegin() + (v.size() - d->offset);
+}
+
+ByteVector::ConstReverseIterator ByteVector::rend() const
+{
+  std::vector<char> &v = d->data->data;
+  return v.rbegin() + (v.size() - d->offset);
 }
 
 bool ByteVector::isNull() const
 {
-  return d == null.d;
+  return (d == null.d);
 }
 
 bool ByteVector::isEmpty() const
 {
-  return d->data.size() == 0;
+  return (d->length == 0);
 }
 
 TagLib::uint ByteVector::checksum() const
@@ -591,42 +691,66 @@ TagLib::uint ByteVector::checksum() const
 
 TagLib::uint ByteVector::toUInt(bool mostSignificantByteFirst) const
 {
-  return toNumber<uint>(d->data, mostSignificantByteFirst);
+  return toNumber<uint>(*this, 0, mostSignificantByteFirst);
+}
+
+TagLib::uint ByteVector::toUInt(uint offset, bool mostSignificantByteFirst) const
+{
+  return toNumber<uint>(*this, offset, mostSignificantByteFirst);
+}
+
+TagLib::uint ByteVector::toUInt(uint offset, uint length, bool mostSignificantByteFirst) const
+{
+  return toNumber<uint>(*this, offset, length, mostSignificantByteFirst);
 }
 
 short ByteVector::toShort(bool mostSignificantByteFirst) const
 {
-  return toNumber<unsigned short>(d->data, mostSignificantByteFirst);
+  return toNumber<unsigned short>(*this, 0, mostSignificantByteFirst);
+}
+
+short ByteVector::toShort(uint offset, bool mostSignificantByteFirst) const
+{
+  return toNumber<unsigned short>(*this, offset, mostSignificantByteFirst);
 }
 
 unsigned short ByteVector::toUShort(bool mostSignificantByteFirst) const
 {
-  return toNumber<unsigned short>(d->data, mostSignificantByteFirst);
+  return toNumber<unsigned short>(*this, 0, mostSignificantByteFirst);
+}
+
+unsigned short ByteVector::toUShort(uint offset, bool mostSignificantByteFirst) const
+{
+  return toNumber<unsigned short>(*this, offset, mostSignificantByteFirst);
 }
 
 long long ByteVector::toLongLong(bool mostSignificantByteFirst) const
 {
-  return toNumber<unsigned long long>(d->data, mostSignificantByteFirst);
+  return toNumber<unsigned long long>(*this, 0, mostSignificantByteFirst);
+}
+
+long long ByteVector::toLongLong(uint offset, bool mostSignificantByteFirst) const
+{
+  return toNumber<unsigned long long>(*this, offset, mostSignificantByteFirst);
 }
 
 const char &ByteVector::operator[](int index) const
 {
-  return d->data[index];
+  return d->data->data[d->offset + index];
 }
 
 char &ByteVector::operator[](int index)
 {
   detach();
-
-  return d->data[index];
+  return d->data->data[d->offset + index];
 }
 
 bool ByteVector::operator==(const ByteVector &v) const
 {
-  if(d->size != v.d->size)
+  if(size() != v.size())
     return false;
 
-  return ::memcmp(data(), v.data(), size()) == 0;
+  return (::memcmp(data(), v.data(), size()) == 0);
 }
 
 bool ByteVector::operator!=(const ByteVector &v) const
@@ -636,10 +760,10 @@ bool ByteVector::operator!=(const ByteVector &v) const
 
 bool ByteVector::operator==(const char *s) const
 {
-  if(d->size != ::strlen(s))
+  if(size() != ::strlen(s))
     return false;
 
-  return ::memcmp(data(), s, d->size) == 0;
+  return (::memcmp(data(), s, size()) == 0);
 }
 
 bool ByteVector::operator!=(const char *s) const
@@ -649,8 +773,7 @@ bool ByteVector::operator!=(const char *s) const
 
 bool ByteVector::operator<(const ByteVector &v) const
 {
-  int result = ::memcmp(data(), v.data(), d->size < v.d->size ? d->size : v.d->size);
-
+  const int result = ::memcmp(data(), v.data(), std::min(size(), v.size()));
   if(result != 0)
     return result < 0;
   else
@@ -697,12 +820,12 @@ ByteVector &ByteVector::operator=(const char *data)
 ByteVector ByteVector::toHex() const
 {
   ByteVector encoded(size() * 2);
+  char *p = encoded.data();
 
-  uint j = 0;
   for(uint i = 0; i < size(); i++) {
-    unsigned char c = d->data[i];
-    encoded[j++] = hexTable[(c >> 4) & 0x0F];
-    encoded[j++] = hexTable[(c     ) & 0x0F];
+    unsigned char c = data()[i];
+    *p++ = hexTable[(c >> 4) & 0x0F];
+    *p++ = hexTable[(c     ) & 0x0F];
   }
 
   return encoded;
@@ -714,17 +837,24 @@ ByteVector ByteVector::toHex() const
 
 void ByteVector::detach()
 {
+  if(d->data->count() > 1) {
+    d->data->deref();
+    d->data = new DataPrivate(d->data->data, d->offset, d->length);
+    d->offset = 0;
+  }
+
   if(d->count() > 1) {
     d->deref();
-    d = new ByteVectorPrivate(d->data);
+    d = new ByteVectorPrivate(d->data->data, d->offset, d->length);
   }
 }
+}
 
 ////////////////////////////////////////////////////////////////////////////////
 // related functions
 ////////////////////////////////////////////////////////////////////////////////
 
-std::ostream &operator<<(std::ostream &s, const ByteVector &v)
+std::ostream &operator<<(std::ostream &s, const TagLib::ByteVector &v)
 {
   for(TagLib::uint i = 0; i < v.size(); i++)
     s << v[i];
diff --git a/3rdparty/taglib/toolkit/tbytevector.h b/3rdparty/taglib/toolkit/tbytevector.h
index 0c5832630..538565b0a 100644
--- a/3rdparty/taglib/toolkit/tbytevector.h
+++ b/3rdparty/taglib/toolkit/tbytevector.h
@@ -48,6 +48,8 @@ namespace TagLib {
 #ifndef DO_NOT_DOCUMENT
     typedef std::vector<char>::iterator Iterator;
     typedef std::vector<char>::const_iterator ConstIterator;
+    typedef std::vector<char>::reverse_iterator ReverseIterator;
+    typedef std::vector<char>::const_reverse_iterator ConstReverseIterator;
 #endif
 
     /*!
@@ -62,12 +64,17 @@ namespace TagLib {
     ByteVector(uint size, char value = 0);
 
     /*!
-     * Contructs a byte vector that is a copy of \a v.
+     * Constructs a byte vector that is a copy of \a v.
      */
     ByteVector(const ByteVector &v);
 
     /*!
-     * Contructs a byte vector that contains \a c.
+     * Constructs a byte vector that is a copy of \a v.
+     */
+    ByteVector(const ByteVector &v, uint offset, uint length);
+
+    /*!
+     * Constructs a byte vector that contains \a c.
      */
     ByteVector(char c);
 
@@ -135,6 +142,14 @@ namespace TagLib {
      */
     int find(const ByteVector &pattern, uint offset = 0, int byteAlign = 1) const;
 
+    /*!
+     * Searches the char for \a c starting at \a offset and returns
+     * the offset.  Returns \a npos if the pattern was not found.  If \a byteAlign is
+     * specified the pattern will only be matched if it starts on a byte divisible
+     * by \a byteAlign (starting from \a offset).
+     */
+    int find(char c, uint offset = 0, int byteAlign = 1) const;
+
     /*!
      * Searches the ByteVector for \a pattern starting from either the end of the
      * vector or \a offset and returns the offset.  Returns -1 if the pattern was
@@ -222,6 +237,26 @@ namespace TagLib {
      */
     ConstIterator end() const;
 
+    /*!
+     * Returns a ReverseIterator that points to the front of the vector.
+     */
+    ReverseIterator rbegin();
+
+    /*!
+     * Returns a ConstReverseIterator that points to the front of the vector.
+     */
+    ConstReverseIterator rbegin() const;
+
+    /*!
+     * Returns a ReverseIterator that points to the back of the vector.
+     */
+    ReverseIterator rend();
+
+    /*!
+     * Returns a ConstReverseIterator that points to the back of the vector.
+     */
+    ConstReverseIterator rend() const;
+
     /*!
      * Returns true if the vector is null.
      *
@@ -256,7 +291,32 @@ namespace TagLib {
     uint toUInt(bool mostSignificantByteFirst = true) const;
 
     /*!
-     * Converts the first 2 bytes of the vector to a short.
+     * Converts the 4 bytes at \a offset of the vector to an unsigned integer. 
+     *
+     * If \a mostSignificantByteFirst is true this will operate left to right
+     * evaluating the integer.  For example if \a mostSignificantByteFirst is
+     * true then $00 $00 $00 $01 == 0x00000001 == 1, if false, $01 00 00 00 ==
+     * 0x01000000 == 1.
+     *
+     * \see fromUInt()
+     */
+    uint toUInt(uint offset, bool mostSignificantByteFirst = true) const;
+
+    /*!
+     * Converts the \a length bytes at \a offset of the vector to an unsigned 
+     * integer. If \a length is larger than 4, the excess is ignored. 
+     *
+     * If \a mostSignificantByteFirst is true this will operate left to right
+     * evaluating the integer.  For example if \a mostSignificantByteFirst is
+     * true then $00 $00 $00 $01 == 0x00000001 == 1, if false, $01 00 00 00 ==
+     * 0x01000000 == 1.
+     *
+     * \see fromUInt()
+     */
+    uint toUInt(uint offset, uint length, bool mostSignificantByteFirst = true) const;
+
+    /*!
+     * Converts the first 2 bytes of the vector to a (signed) short.
      *
      * If \a mostSignificantByteFirst is true this will operate left to right
      * evaluating the integer.  For example if \a mostSignificantByteFirst is
@@ -266,6 +326,17 @@ namespace TagLib {
      */
     short toShort(bool mostSignificantByteFirst = true) const;
 
+    /*!
+     * Converts the 2 bytes at \a offset of the vector to a (signed) short.
+     *
+     * If \a mostSignificantByteFirst is true this will operate left to right
+     * evaluating the integer.  For example if \a mostSignificantByteFirst is
+     * true then $00 $01 == 0x0001 == 1, if false, $01 00 == 0x01000000 == 1.
+     *
+     * \see fromShort()
+     */
+    short toShort(uint offset, bool mostSignificantByteFirst = true) const;
+
     /*!
      * Converts the first 2 bytes of the vector to a unsigned short.
      *
@@ -277,6 +348,17 @@ namespace TagLib {
      */
     unsigned short toUShort(bool mostSignificantByteFirst = true) const;
 
+    /*!
+     * Converts the 2 bytes at \a offset of the vector to a unsigned short.
+     *
+     * If \a mostSignificantByteFirst is true this will operate left to right
+     * evaluating the integer.  For example if \a mostSignificantByteFirst is
+     * true then $00 $01 == 0x0001 == 1, if false, $01 00 == 0x01000000 == 1.
+     *
+     * \see fromShort()
+     */
+    unsigned short toUShort(uint offset, bool mostSignificantByteFirst = true) const;
+
     /*!
      * Converts the first 8 bytes of the vector to a (signed) long long.
      *
@@ -289,6 +371,18 @@ namespace TagLib {
      */
     long long toLongLong(bool mostSignificantByteFirst = true) const;
 
+    /*!
+     * Converts the 8 bytes at \a offset of the vector to a (signed) long long.
+     *
+     * If \a mostSignificantByteFirst is true this will operate left to right
+     * evaluating the integer.  For example if \a mostSignificantByteFirst is
+     * true then $00 00 00 00 00 00 00 01 == 0x0000000000000001 == 1,
+     * if false, $01 00 00 00 00 00 00 00 == 0x0100000000000000 == 1.
+     *
+     * \see fromUInt()
+     */
+    long long toLongLong(uint offset, bool mostSignificantByteFirst = true) const;
+
     /*!
      * Creates a 4 byte ByteVector based on \a value.  If
      * \a mostSignificantByteFirst is true, then this will operate left to right
@@ -413,7 +507,6 @@ namespace TagLib {
     class ByteVectorPrivate;
     ByteVectorPrivate *d;
   };
-
 }
 
 /*!
diff --git a/3rdparty/taglib/toolkit/tdebug.cpp b/3rdparty/taglib/toolkit/tdebug.cpp
index 522b68c97..71350af7c 100644
--- a/3rdparty/taglib/toolkit/tdebug.cpp
+++ b/3rdparty/taglib/toolkit/tdebug.cpp
@@ -23,33 +23,77 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
-#ifndef NDEBUG
-#include <iostream>
-#include <bitset>
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif
 
 #include "tdebug.h"
 #include "tstring.h"
+#include "tdebuglistener.h"
+
+#include <bitset>
+#include <cstdio>
+#include <cstdarg>
 
 using namespace TagLib;
 
-void TagLib::debug(const String &s)
+namespace
 {
-  std::cerr << "TagLib: " << s << std::endl;
-}
+  String format(const char *fmt, ...)
+  {
+    va_list args;
+    va_start(args, fmt);
 
-void TagLib::debugData(const ByteVector &v)
-{
-  for(uint i = 0; i < v.size(); i++) {
+    char buf[256];
 
-    std::cout << "*** [" << i << "] - '" << char(v[i]) << "' - int " << int(v[i])
-              << std::endl;
+#if defined(HAVE_SNPRINTF)
 
-    std::bitset<8> b(v[i]);
+    vsnprintf(buf, sizeof(buf), fmt, args);
 
-    for(int j = 0; j < 8; j++)
-      std::cout << i << ":" << j << " " << b.test(j) << std::endl;
+#elif defined(HAVE_SPRINTF_S)
 
-    std::cout << std::endl;
+    vsprintf_s(buf, fmt, args);
+
+#else
+
+    // Be careful. May cause a buffer overflow.  
+    vsprintf(buf, fmt, args);
+
+#endif
+
+    va_end(args);
+
+    return String(buf);
   }
 }
+
+namespace TagLib
+{
+  // The instance is defined in tdebuglistener.cpp.
+  extern DebugListener *debugListener;
+
+  void debug(const String &s)
+  {
+#if !defined(NDEBUG) || defined(TRACE_IN_RELEASE)
+  
+    debugListener->printMessage("TagLib: " + s + "\n");
+
 #endif
+  }
+
+  void debugData(const ByteVector &v)
+  {
+#if !defined(NDEBUG) || defined(TRACE_IN_RELEASE)
+
+    for(size_t i = 0; i < v.size(); ++i) 
+    {
+      std::string bits = std::bitset<8>(v[i]).to_string();
+      String msg = format("*** [%d] - char '%c' - int %d, 0x%02x, 0b%s\n", 
+        i, v[i], v[i], v[i], bits.c_str());
+
+      debugListener->printMessage(msg);
+    }
+
+#endif
+  }
+}
diff --git a/3rdparty/taglib/toolkit/tdebug.h b/3rdparty/taglib/toolkit/tdebug.h
index 5204fe707..bd94d159f 100644
--- a/3rdparty/taglib/toolkit/tdebug.h
+++ b/3rdparty/taglib/toolkit/tdebug.h
@@ -32,11 +32,11 @@ namespace TagLib {
   class ByteVector;
 
 #ifndef DO_NOT_DOCUMENT
-#ifndef NDEBUG
 
   /*!
-   * A simple function that prints debugging output to cerr if debugging is
-   * not disabled.
+   * A simple function that outputs the debug messages to the listener. 
+   * The default listener redirects the messages to \a stderr when NDEBUG is 
+   * not defined.
    *
    * \warning Do not use this outside of TagLib, it could lead to undefined
    * symbols in your build if TagLib is built with NDEBUG defined and your
@@ -45,7 +45,7 @@ namespace TagLib {
    * \internal
    */
   void debug(const String &s);
-
+  
   /*!
    * For debugging binary data.
    *
@@ -56,16 +56,7 @@ namespace TagLib {
    * \internal
    */
   void debugData(const ByteVector &v);
-
-#else
-
-  // Define these to an empty statement if debugging is disabled.
-
-#define debug(x)
-#define debugData(x)
-
-#endif
-#endif
 }
 
 #endif
+#endif
diff --git a/3rdparty/taglib/toolkit/tdebuglistener.cpp b/3rdparty/taglib/toolkit/tdebuglistener.cpp
new file mode 100644
index 000000000..48912222d
--- /dev/null
+++ b/3rdparty/taglib/toolkit/tdebuglistener.cpp
@@ -0,0 +1,85 @@
+/***************************************************************************
+    copyright            : (C) 2013 by Tsuda Kageyu
+    email                : tsuda.kageyu@gmail.com
+ ***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#include "tdebuglistener.h"
+
+#include <iostream>
+#include <bitset>
+
+#ifdef _WIN32
+# include <windows.h>
+#endif
+
+using namespace TagLib;
+
+namespace
+{
+  class DefaultListener : public DebugListener
+  {
+  public:
+    virtual void printMessage(const String &msg)
+    {
+#ifdef _WIN32
+
+      const wstring wstr = msg.toWString();
+      const int len = WideCharToMultiByte(CP_ACP, 0, wstr.c_str(), -1, NULL, 0, NULL, NULL);
+      if(len != 0) {
+        std::vector<char> buf(len);
+        WideCharToMultiByte(CP_ACP, 0, wstr.c_str(), -1, &buf[0], len, NULL, NULL);
+
+        std::cerr << std::string(&buf[0]);
+      }
+
+#else
+
+      std::cerr << msg;
+
+#endif 
+    }
+  };
+
+  DefaultListener defaultListener;
+}
+
+namespace TagLib
+{
+  DebugListener *debugListener = &defaultListener;
+
+  DebugListener::DebugListener()
+  {
+  }
+
+  DebugListener::~DebugListener()
+  {
+  }
+
+  void setDebugListener(DebugListener *listener)
+  {
+    if(listener)
+      debugListener = listener;
+    else
+      debugListener = &defaultListener;
+  }
+}
diff --git a/3rdparty/taglib/toolkit/tdebuglistener.h b/3rdparty/taglib/toolkit/tdebuglistener.h
new file mode 100644
index 000000000..a32f285f4
--- /dev/null
+++ b/3rdparty/taglib/toolkit/tdebuglistener.h
@@ -0,0 +1,74 @@
+/***************************************************************************
+    copyright            : (C) 2013 by Tsuda Kageyu
+    email                : tsuda.kageyu@gmail.com
+ ***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#ifndef TAGLIB_DEBUGLISTENER_H
+#define TAGLIB_DEBUGLISTENER_H
+
+#include "taglib_export.h"
+#include "tstring.h"
+
+namespace TagLib 
+{
+  //! An abstraction for the listener to the debug messages.
+
+  /*!
+   * This class enables you to handle the debug messages in your preferred 
+   * way by subclassing this class, reimplementing printMessage() and setting 
+   * your reimplementation as the default with setDebugListener().
+   *
+   * \see setDebugListener()
+   */  
+  class TAGLIB_EXPORT DebugListener
+  {
+  public:
+    DebugListener();
+    virtual ~DebugListener();
+
+    /*!
+     * When overridden in a derived class, redirects \a msg to your preferred
+     * channel such as stderr, Windows debugger or so forth.
+     */
+    virtual void printMessage(const String &msg) = 0;
+
+  private:
+    // Noncopyable
+    DebugListener(const DebugListener &);
+    DebugListener &operator=(const DebugListener &);
+  };
+
+  /*!
+   * Sets the listener that decides how the debug messages are redirected.
+   * If the parameter \a listener is null, the previous listener is released 
+   * and default stderr listener is restored.   
+   *
+   * \note The caller is responsible for deleting the previous listener
+   * as needed after it is released.
+   *
+   * \see DebugListener
+   */
+  TAGLIB_EXPORT void setDebugListener(DebugListener *listener);
+}
+
+#endif
diff --git a/3rdparty/taglib/toolkit/tfile.cpp b/3rdparty/taglib/toolkit/tfile.cpp
index d0a6116fb..4a05b05f0 100644
--- a/3rdparty/taglib/toolkit/tfile.cpp
+++ b/3rdparty/taglib/toolkit/tfile.cpp
@@ -29,21 +29,14 @@
 #include "tdebug.h"
 #include "tpropertymap.h"
 
-#include <stdio.h>
-#include <string.h>
-#include <sys/stat.h>
-
 #ifdef _WIN32
-# include <wchar.h>
 # include <windows.h>
 # include <io.h>
-# define ftruncate _chsize
 #else
+# include <stdio.h>
 # include <unistd.h>
 #endif
 
-#include <stdlib.h>
-
 #ifndef R_OK
 # define R_OK 4
 #endif
@@ -60,6 +53,7 @@
 #include "mp4file.h"
 #include "wavpackfile.h"
 #include "speexfile.h"
+#include "opusfile.h"
 #include "trueaudiofile.h"
 #include "aifffile.h"
 #include "wavfile.h"
@@ -68,9 +62,19 @@
 #include "s3mfile.h"
 #include "itfile.h"
 #include "xmfile.h"
+#include "mp4file.h"
 
 using namespace TagLib;
 
+namespace
+{
+#ifdef _WIN32
+  const TagLib::uint BufferSize = 8192;
+#else
+  const TagLib::uint BufferSize = 1024;
+#endif
+}
+
 class File::FilePrivate
 {
 public:
@@ -79,7 +83,6 @@ public:
   IOStream *stream;
   bool streamOwner;
   bool valid;
-  static const uint bufferSize = 1024;
 };
 
 File::FilePrivate::FilePrivate(IOStream *stream, bool owner) :
@@ -135,6 +138,8 @@ PropertyMap File::properties() const
     return dynamic_cast<const Ogg::FLAC::File* >(this)->properties();
   if(dynamic_cast<const Ogg::Speex::File* >(this))
     return dynamic_cast<const Ogg::Speex::File* >(this)->properties();
+  if(dynamic_cast<const Ogg::Opus::File* >(this))
+    return dynamic_cast<const Ogg::Opus::File* >(this)->properties();
   if(dynamic_cast<const Ogg::Vorbis::File* >(this))
     return dynamic_cast<const Ogg::Vorbis::File* >(this)->properties();
   if(dynamic_cast<const RIFF::AIFF::File* >(this))
@@ -149,12 +154,10 @@ PropertyMap File::properties() const
     return dynamic_cast<const WavPack::File* >(this)->properties();
   if(dynamic_cast<const XM::File* >(this))
     return dynamic_cast<const XM::File* >(this)->properties();
-  // no specialized implementation available -> use generic one
-  // - ASF: ugly format, largely undocumented, not worth implementing
-  //   dict interface ...
-  // - MP4: taglib's MP4::Tag does not really support anything beyond
-  //   the basic implementation, therefor we use just the default Tag
-  //   interface
+  if(dynamic_cast<const MP4::File* >(this))
+    return dynamic_cast<const MP4::File* >(this)->properties();
+  if(dynamic_cast<const ASF::File* >(this))
+    return dynamic_cast<const ASF::File* >(this)->properties();
   return tag()->properties();
 }
 
@@ -170,24 +173,20 @@ void File::removeUnsupportedProperties(const StringList &properties)
     dynamic_cast<MPC::File* >(this)->removeUnsupportedProperties(properties);
   else if(dynamic_cast<MPEG::File* >(this))
     dynamic_cast<MPEG::File* >(this)->removeUnsupportedProperties(properties);
-  else if(dynamic_cast<Ogg::FLAC::File* >(this))
-    dynamic_cast<Ogg::FLAC::File* >(this)->removeUnsupportedProperties(properties);
-  else if(dynamic_cast<Ogg::Speex::File* >(this))
-    dynamic_cast<Ogg::Speex::File* >(this)->removeUnsupportedProperties(properties);
   else if(dynamic_cast<Ogg::Vorbis::File* >(this))
     dynamic_cast<Ogg::Vorbis::File* >(this)->removeUnsupportedProperties(properties);
   else if(dynamic_cast<RIFF::AIFF::File* >(this))
     dynamic_cast<RIFF::AIFF::File* >(this)->removeUnsupportedProperties(properties);
   else if(dynamic_cast<RIFF::WAV::File* >(this))
     dynamic_cast<RIFF::WAV::File* >(this)->removeUnsupportedProperties(properties);
-  else if(dynamic_cast<S3M::File* >(this))
-    dynamic_cast<S3M::File* >(this)->removeUnsupportedProperties(properties);
   else if(dynamic_cast<TrueAudio::File* >(this))
     dynamic_cast<TrueAudio::File* >(this)->removeUnsupportedProperties(properties);
   else if(dynamic_cast<WavPack::File* >(this))
     dynamic_cast<WavPack::File* >(this)->removeUnsupportedProperties(properties);
-  else if(dynamic_cast<XM::File* >(this))
-    dynamic_cast<XM::File* >(this)->removeUnsupportedProperties(properties);
+  else if(dynamic_cast<MP4::File* >(this))
+    dynamic_cast<MP4::File* >(this)->removeUnsupportedProperties(properties);
+  else if(dynamic_cast<ASF::File* >(this))
+    dynamic_cast<ASF::File* >(this)->removeUnsupportedProperties(properties);
   else
     tag()->removeUnsupportedProperties(properties);
 }
@@ -210,6 +209,8 @@ PropertyMap File::setProperties(const PropertyMap &properties)
     return dynamic_cast<Ogg::FLAC::File* >(this)->setProperties(properties);
   else if(dynamic_cast<Ogg::Speex::File* >(this))
     return dynamic_cast<Ogg::Speex::File* >(this)->setProperties(properties);
+  else if(dynamic_cast<Ogg::Opus::File* >(this))
+    return dynamic_cast<Ogg::Opus::File* >(this)->setProperties(properties);
   else if(dynamic_cast<Ogg::Vorbis::File* >(this))
     return dynamic_cast<Ogg::Vorbis::File* >(this)->setProperties(properties);
   else if(dynamic_cast<RIFF::AIFF::File* >(this))
@@ -224,6 +225,10 @@ PropertyMap File::setProperties(const PropertyMap &properties)
     return dynamic_cast<WavPack::File* >(this)->setProperties(properties);
   else if(dynamic_cast<XM::File* >(this))
     return dynamic_cast<XM::File* >(this)->setProperties(properties);
+  else if(dynamic_cast<MP4::File* >(this))
+    return dynamic_cast<MP4::File* >(this)->setProperties(properties);
+  else if(dynamic_cast<ASF::File* >(this))
+    return dynamic_cast<ASF::File* >(this)->setProperties(properties);
   else
     return tag()->setProperties(properties);
 }
@@ -240,7 +245,7 @@ void File::writeBlock(const ByteVector &data)
 
 long File::find(const ByteVector &pattern, long fromOffset, const ByteVector &before)
 {
-  if(!d->stream || pattern.size() > d->bufferSize)
+  if(!d->stream || pattern.size() > bufferSize())
       return -1;
 
   // The position in the file that the current buffer starts at.
@@ -281,20 +286,20 @@ long File::find(const ByteVector &pattern, long fromOffset, const ByteVector &be
   // then check for "before".  The order is important because it gives priority
   // to "real" matches.
 
-  for(buffer = readBlock(d->bufferSize); buffer.size() > 0; buffer = readBlock(d->bufferSize)) {
+  for(buffer = readBlock(bufferSize()); buffer.size() > 0; buffer = readBlock(bufferSize())) {
 
     // (1) previous partial match
 
-    if(previousPartialMatch >= 0 && int(d->bufferSize) > previousPartialMatch) {
-      const int patternOffset = (d->bufferSize - previousPartialMatch);
+    if(previousPartialMatch >= 0 && int(bufferSize()) > previousPartialMatch) {
+      const int patternOffset = (bufferSize() - previousPartialMatch);
       if(buffer.containsAt(pattern, 0, patternOffset)) {
         seek(originalPosition);
-        return bufferOffset - d->bufferSize + previousPartialMatch;
+        return bufferOffset - bufferSize() + previousPartialMatch;
       }
     }
 
-    if(!before.isNull() && beforePreviousPartialMatch >= 0 && int(d->bufferSize) > beforePreviousPartialMatch) {
-      const int beforeOffset = (d->bufferSize - beforePreviousPartialMatch);
+    if(!before.isNull() && beforePreviousPartialMatch >= 0 && int(bufferSize()) > beforePreviousPartialMatch) {
+      const int beforeOffset = (bufferSize() - beforePreviousPartialMatch);
       if(buffer.containsAt(before, 0, beforeOffset)) {
         seek(originalPosition);
         return -1;
@@ -321,7 +326,7 @@ long File::find(const ByteVector &pattern, long fromOffset, const ByteVector &be
     if(!before.isNull())
       beforePreviousPartialMatch = buffer.endsWithPartialMatch(before);
 
-    bufferOffset += d->bufferSize;
+    bufferOffset += bufferSize();
   }
 
   // Since we hit the end of the file, reset the status before continuing.
@@ -336,7 +341,7 @@ long File::find(const ByteVector &pattern, long fromOffset, const ByteVector &be
 
 long File::rfind(const ByteVector &pattern, long fromOffset, const ByteVector &before)
 {
-  if(!d->stream || pattern.size() > d->bufferSize)
+  if(!d->stream || pattern.size() > bufferSize())
       return -1;
 
   // The position in the file that the current buffer starts at.
@@ -360,17 +365,17 @@ long File::rfind(const ByteVector &pattern, long fromOffset, const ByteVector &b
 
   long bufferOffset;
   if(fromOffset == 0) {
-    seek(-1 * int(d->bufferSize), End);
+    seek(-1 * int(bufferSize()), End);
     bufferOffset = tell();
   }
   else {
-    seek(fromOffset + -1 * int(d->bufferSize), Beginning);
+    seek(fromOffset + -1 * int(bufferSize()), Beginning);
     bufferOffset = tell();
   }
 
   // See the notes in find() for an explanation of this algorithm.
 
-  for(buffer = readBlock(d->bufferSize); buffer.size() > 0; buffer = readBlock(d->bufferSize)) {
+  for(buffer = readBlock(bufferSize()); buffer.size() > 0; buffer = readBlock(bufferSize())) {
 
     // TODO: (1) previous partial match
 
@@ -389,7 +394,7 @@ long File::rfind(const ByteVector &pattern, long fromOffset, const ByteVector &b
 
     // TODO: (3) partial match
 
-    bufferOffset -= d->bufferSize;
+    bufferOffset -= bufferSize();
     seek(bufferOffset);
   }
 
@@ -488,7 +493,7 @@ bool File::isWritable(const char *file)
 
 TagLib::uint File::bufferSize()
 {
-  return FilePrivate::bufferSize;
+  return BufferSize;
 }
 
 void File::setValid(bool valid)
diff --git a/3rdparty/taglib/toolkit/tfile.h b/3rdparty/taglib/toolkit/tfile.h
index 7e6f2b936..67f6f80f5 100644
--- a/3rdparty/taglib/toolkit/tfile.h
+++ b/3rdparty/taglib/toolkit/tfile.h
@@ -80,12 +80,14 @@ namespace TagLib {
 
     /*!
      * Exports the tags of the file as dictionary mapping (human readable) tag
-     * names (Strings) to StringLists of tag values. Calls the according specialization
-     * in the File subclasses.
+     * names (uppercase Strings) to StringLists of tag values. Calls the according
+     * specialization in the File subclasses.
      * For each metadata object of the file that could not be parsed into the PropertyMap
      * format, the returend map's unsupportedData() list will contain one entry identifying
      * that object (e.g. the frame type for ID3v2 tags). Use removeUnsupportedProperties()
      * to remove (a subset of) them.
+     * For files that contain more than one tag (e.g. an MP3 with both an ID3v2 and an ID3v2
+     * tag) only the most "modern" one will be exported (ID3v2 in this case).
      * BIC: Will be made virtual in future releases.
      */
     PropertyMap properties() const;
@@ -105,9 +107,15 @@ namespace TagLib {
      * If some value(s) could not be written imported to the specific metadata format,
      * the returned PropertyMap will contain those value(s). Otherwise it will be empty,
      * indicating that no problems occured.
+     * With file types that support several tag formats (for instance, MP3 files can have
+     * ID3v1, ID3v2, and APEv2 tags), this function will create the most appropriate one
+     * (ID3v2 for MP3 files). Older formats will be updated as well, if they exist, but won't
+     * be taken into account for the return value of this function.
+     * See the documentation of the subclass implementations for detailed descriptions.
      * BIC: will become pure virtual in the future
      */
     PropertyMap setProperties(const PropertyMap &properties);
+    
     /*!
      * Returns a pointer to this file's audio properties.  This should be
      * reimplemented in the concrete subclasses.  If no audio properties were
@@ -205,7 +213,7 @@ namespace TagLib {
     bool isOpen() const;
 
     /*!
-     * Returns true if the file is open and readble.
+     * Returns true if the file is open and readable.
      */
     bool isValid() const;
 
diff --git a/3rdparty/taglib/toolkit/tfilestream.cpp b/3rdparty/taglib/toolkit/tfilestream.cpp
index b1fd9f80a..eae525733 100644
--- a/3rdparty/taglib/toolkit/tfilestream.cpp
+++ b/3rdparty/taglib/toolkit/tfilestream.cpp
@@ -27,137 +27,145 @@
 #include "tstring.h"
 #include "tdebug.h"
 
-#include <stdio.h>
-#include <string.h>
-#include <sys/stat.h>
-
 #ifdef _WIN32
-# include <wchar.h>
 # include <windows.h>
-# include <io.h>
-# define ftruncate _chsize
 #else
+# include <stdio.h>
 # include <unistd.h>
 #endif
 
-#include <stdlib.h>
-
-#ifndef R_OK
-# define R_OK 4
-#endif
-#ifndef W_OK
-# define W_OK 2
-#endif
-
 using namespace TagLib;
 
+namespace 
+{
 #ifdef _WIN32
 
-typedef FileName FileNameHandle;
+  // Uses Win32 native API instead of POSIX API to reduce the resource consumption.
 
-#else
+  typedef FileName FileNameHandle;
+  typedef HANDLE FileHandle;
 
-struct FileNameHandle : public std::string
-{
-  FileNameHandle(FileName name) : std::string(name) {}
-  operator FileName () const { return c_str(); }
-};
+  const TagLib::uint BufferSize = 8192;
+  const FileHandle InvalidFileHandle = INVALID_HANDLE_VALUE;
 
-#endif
-
-namespace {
-  FILE *openFile(const FileName &path, bool readOnly)
+  inline FileHandle openFile(const FileName &path, bool readOnly)
   {
-    // Calls a proper variation of fopen() depending on the compiling environment.
+    const DWORD access = readOnly ? GENERIC_READ : (GENERIC_READ | GENERIC_WRITE);
 
-#if defined(_WIN32)
-  
-# if defined(_MSC_VER) && (_MSC_VER >= 1400) 
-
-    // Visual C++ 2005 or later.
-
-    FILE *file;
-    errno_t err;
-
-    if(wcslen(path) > 0)
-      err = _wfopen_s(&file, path, readOnly ? L"rb" : L"rb+");
+    if(!path.wstr().empty())
+      return CreateFileW(path.wstr().c_str(), access, FILE_SHARE_READ, NULL, OPEN_EXISTING, 0, NULL);
+    else if(!path.str().empty())
+      return CreateFileA(path.str().c_str(), access, FILE_SHARE_READ, NULL, OPEN_EXISTING, 0, NULL);
     else
-      err = fopen_s(&file, path, readOnly ? "rb" : "rb+");
-  
-    if(err == 0)
-      return file;
-    else
-      return NULL;
-  
-# else   // defined(_MSC_VER) && (_MSC_VER >= 1400)
-
-    // Visual C++.NET 2003 or earlier.
-
-    if(wcslen(path) > 0)
-      return _wfopen(path, readOnly ? L"rb" : L"rb+");
-    else
-      return fopen(path, readOnly ? "rb" : "rb+");
-
-# endif  // defined(_MSC_VER) && (_MSC_VER >= 1400)
-
-#else  // defined(_WIN32)
-
-    // Non-Win32
-
-    return fopen(path, readOnly ? "rb" : "rb+");
-
-#endif // defined(_WIN32)
+      return InvalidFileHandle;
   }
+
+  inline void closeFile(FileHandle file)
+  {
+    CloseHandle(file);
+  }
+
+  inline size_t readFile(FileHandle file, ByteVector &buffer)
+  {
+    DWORD length;
+    if(ReadFile(file, buffer.data(), static_cast<DWORD>(buffer.size()), &length, NULL))
+      return static_cast<size_t>(length);
+    else
+      return 0;
+  }
+
+  inline size_t writeFile(FileHandle file, const ByteVector &buffer)
+  {
+    DWORD length;
+    if(WriteFile(file, buffer.data(), static_cast<DWORD>(buffer.size()), &length, NULL))
+      return static_cast<size_t>(length);
+    else 
+      return 0;
+  }
+
+#else   // _WIN32
+
+  struct FileNameHandle : public std::string
+  {
+    FileNameHandle(FileName name) : std::string(name) {}
+    operator FileName () const { return c_str(); }
+  };
+
+  typedef FILE* FileHandle;
+
+  const TagLib::uint BufferSize = 8192;
+  const FileHandle InvalidFileHandle = 0;
+
+  inline FileHandle openFile(const FileName &path, bool readOnly)
+  {
+    return fopen(path, readOnly ? "rb" : "rb+");
+  }
+
+  inline void closeFile(FileHandle file)
+  {
+    fclose(file);
+  }
+
+  inline size_t readFile(FileHandle file, ByteVector &buffer)
+  {
+    return fread(buffer.data(), sizeof(char), buffer.size(), file);
+  }
+
+  inline size_t writeFile(FileHandle file, const ByteVector &buffer)
+  {
+    return fwrite(buffer.data(), sizeof(char), buffer.size(), file);
+  }
+
+#endif  // _WIN32
 }
 
 class FileStream::FileStreamPrivate
 {
 public:
-  FileStreamPrivate(FileName fileName, bool openReadOnly);
-
-  FILE *file;
-
-  FileNameHandle name;
-
-  bool readOnly;
-  ulong size;
-  static const uint bufferSize = 1024;
-};
-
-FileStream::FileStreamPrivate::FileStreamPrivate(FileName fileName, bool openReadOnly) :
-  file(0),
-  name(fileName),
-  readOnly(true),
-  size(0)
-{
-  // First try with read / write mode, if that fails, fall back to read only.
-
-  if(!openReadOnly)
-    file = openFile(name, false);
-
-  if(file)
-    readOnly = false;
-  else
-    file = openFile(name, true);
-
-  if(!file) {
-    debug("Could not open file " + String((const char *) name));
+  FileStreamPrivate(const FileName &fileName)
+    : file(InvalidFileHandle)
+    , name(fileName)
+    , readOnly(true)
+  {
   }
-}
+
+  FileHandle file;
+  FileNameHandle name;
+  bool readOnly;
+};
 
 ////////////////////////////////////////////////////////////////////////////////
 // public members
 ////////////////////////////////////////////////////////////////////////////////
 
-FileStream::FileStream(FileName file, bool openReadOnly)
+FileStream::FileStream(FileName fileName, bool openReadOnly)
+  : d(new FileStreamPrivate(fileName))
 {
-  d = new FileStreamPrivate(file, openReadOnly);
+  // First try with read / write mode, if that fails, fall back to read only.
+
+  if(!openReadOnly)
+    d->file = openFile(fileName, false);
+
+  if(d->file != InvalidFileHandle)
+    d->readOnly = false;
+  else
+    d->file = openFile(fileName, true);
+
+  if(d->file == InvalidFileHandle) 
+  {
+# ifdef _WIN32
+    debug("Could not open file " + fileName.toString());
+# else
+    debug("Could not open file " + String(static_cast<const char *>(d->name)));
+# endif 
+  }
 }
 
 FileStream::~FileStream()
 {
-  if(d->file)
-    fclose(d->file);
+  if(isOpen())
+    closeFile(d->file);
+
   delete d;
 }
 
@@ -168,43 +176,52 @@ FileName FileStream::name() const
 
 ByteVector FileStream::readBlock(ulong length)
 {
-  if(!d->file) {
-    debug("FileStream::readBlock() -- Invalid File");
+  if(!isOpen()) {
+    debug("File::readBlock() -- invalid file.");
     return ByteVector::null;
   }
 
   if(length == 0)
     return ByteVector::null;
 
-  if(length > FileStreamPrivate::bufferSize &&
-     length > ulong(FileStream::length()))
-  {
-    length = FileStream::length();
-  }
+  const ulong streamLength = static_cast<ulong>(FileStream::length());
+  if(length > bufferSize() && length > streamLength)
+    length = streamLength;
 
-  ByteVector v(static_cast<uint>(length));
-  const int count = fread(v.data(), sizeof(char), length, d->file);
-  v.resize(count);
-  return v;
+  ByteVector buffer(static_cast<uint>(length));
+
+  const size_t count = readFile(d->file, buffer);
+  buffer.resize(static_cast<uint>(count));
+  
+  return buffer;
 }
 
 void FileStream::writeBlock(const ByteVector &data)
 {
-  if(!d->file)
-    return;
-
-  if(d->readOnly) {
-    debug("File::writeBlock() -- attempted to write to a file that is not writable");
+  if(!isOpen()) {
+    debug("File::writeBlock() -- invalid file.");
     return;
   }
 
-  fwrite(data.data(), sizeof(char), data.size(), d->file);
+  if(readOnly()) {
+    debug("File::writeBlock() -- read only file.");
+    return;
+  }
+
+  writeFile(d->file, data);
 }
 
 void FileStream::insert(const ByteVector &data, ulong start, ulong replace)
 {
-  if(!d->file)
+  if(!isOpen()) {
+    debug("File::insert() -- invalid file.");
     return;
+  }
+
+  if(readOnly()) {
+    debug("File::insert() -- read only file.");
+    return;
+  }
 
   if(data.size() == replace) {
     seek(start);
@@ -212,10 +229,10 @@ void FileStream::insert(const ByteVector &data, ulong start, ulong replace)
     return;
   }
   else if(data.size() < replace) {
-      seek(start);
-      writeBlock(data);
-      removeBlock(start + data.size(), replace - data.size());
-      return;
+    seek(start);
+    writeBlock(data);
+    removeBlock(start + data.size(), replace - data.size());
+    return;
   }
 
   // Woohoo!  Faster (about 20%) than id3lib at last.  I had to get hardcore
@@ -238,71 +255,50 @@ void FileStream::insert(const ByteVector &data, ulong start, ulong replace)
   long readPosition = start + replace;
   long writePosition = start;
 
-  ByteVector buffer;
+  ByteVector buffer = data;
   ByteVector aboutToOverwrite(static_cast<uint>(bufferLength));
 
-  // This is basically a special case of the loop below.  Here we're just
-  // doing the same steps as below, but since we aren't using the same buffer
-  // size -- instead we're using the tag size -- this has to be handled as a
-  // special case.  We're also using File::writeBlock() just for the tag.
-  // That's a bit slower than using char *'s so, we're only doing it here.
-
-  seek(readPosition);
-  int bytesRead = fread(aboutToOverwrite.data(), sizeof(char), bufferLength, d->file);
-  readPosition += bufferLength;
-
-  seek(writePosition);
-  writeBlock(data);
-  writePosition += data.size();
-
-  buffer = aboutToOverwrite;
-
-  // In case we've already reached the end of file...
-
-  buffer.resize(bytesRead);
-
-  // Ok, here's the main loop.  We want to loop until the read fails, which
-  // means that we hit the end of the file.
-
-  while(!buffer.isEmpty()) {
-
+  while(true)
+  {
     // Seek to the current read position and read the data that we're about
     // to overwrite.  Appropriately increment the readPosition.
-
+    
     seek(readPosition);
-    bytesRead = fread(aboutToOverwrite.data(), sizeof(char), bufferLength, d->file);
+    const size_t bytesRead = readFile(d->file, aboutToOverwrite);
     aboutToOverwrite.resize(bytesRead);
     readPosition += bufferLength;
 
     // Check to see if we just read the last block.  We need to call clear()
     // if we did so that the last write succeeds.
 
-    if(ulong(bytesRead) < bufferLength)
+    if(bytesRead < bufferLength)
       clear();
 
     // Seek to the write position and write our buffer.  Increment the
     // writePosition.
 
     seek(writePosition);
-    fwrite(buffer.data(), sizeof(char), buffer.size(), d->file);
+    writeBlock(buffer);
+
+    // We hit the end of the file.
+
+    if(bytesRead == 0)
+      break;
+
     writePosition += buffer.size();
 
     // Make the current buffer the data that we read in the beginning.
-
+    
     buffer = aboutToOverwrite;
-
-    // Again, we need this for the last write.  We don't want to write garbage
-    // at the end of our file, so we need to set the buffer size to the amount
-    // that we actually read.
-
-    bufferLength = bytesRead;
   }
 }
 
 void FileStream::removeBlock(ulong start, ulong length)
 {
-  if(!d->file)
+  if(!isOpen()) {
+    debug("File::removeBlock() -- invalid file.");
     return;
+  }
 
   ulong bufferLength = bufferSize();
 
@@ -311,23 +307,26 @@ void FileStream::removeBlock(ulong start, ulong length)
 
   ByteVector buffer(static_cast<uint>(bufferLength));
 
-  ulong bytesRead = 1;
-
-  while(bytesRead != 0) {
+  for(size_t bytesRead = -1; bytesRead != 0;)
+  {
     seek(readPosition);
-    bytesRead = fread(buffer.data(), sizeof(char), bufferLength, d->file);
+    bytesRead = readFile(d->file, buffer);
     readPosition += bytesRead;
 
     // Check to see if we just read the last block.  We need to call clear()
     // if we did so that the last write succeeds.
 
-    if(bytesRead < bufferLength)
+    if(bytesRead < buffer.size()) {
       clear();
+      buffer.resize(bytesRead);
+    }
 
     seek(writePosition);
-    fwrite(buffer.data(), sizeof(char), bytesRead, d->file);
+    writeFile(d->file, buffer);
+
     writePosition += bytesRead;
   }
+
   truncate(writePosition);
 }
 
@@ -338,58 +337,125 @@ bool FileStream::readOnly() const
 
 bool FileStream::isOpen() const
 {
-  return (d->file != NULL);
+  return (d->file != InvalidFileHandle);
 }
 
 void FileStream::seek(long offset, Position p)
 {
-  if(!d->file) {
-    debug("File::seek() -- trying to seek in a file that isn't opened.");
+  if(!isOpen()) {
+    debug("File::seek() -- invalid file.");
     return;
   }
 
+#ifdef _WIN32
+
+  DWORD whence;
   switch(p) {
   case Beginning:
-    fseek(d->file, offset, SEEK_SET);
+    whence = FILE_BEGIN;
     break;
   case Current:
-    fseek(d->file, offset, SEEK_CUR);
+    whence = FILE_CURRENT;
     break;
   case End:
-    fseek(d->file, offset, SEEK_END);
+    whence = FILE_END;
     break;
+  default:
+    debug("FileStream::seek() -- Invalid Position value.");
+    return;
   }
+
+  SetFilePointer(d->file, offset, NULL, whence);
+  if(GetLastError() != NO_ERROR) {
+    debug("File::seek() -- Failed to set the file pointer.");
+  }
+
+#else
+
+  int whence;
+  switch(p) {
+  case Beginning:
+    whence = SEEK_SET;
+    break;
+  case Current:
+    whence = SEEK_CUR;
+    break;
+  case End:
+    whence = SEEK_END;
+    break;
+  default:
+    debug("FileStream::seek() -- Invalid Position value.");
+    return;
+  }
+
+  fseek(d->file, offset, whence);
+
+#endif
 }
 
 void FileStream::clear()
 {
+#ifdef _WIN32
+
+  // NOP
+
+#else
+
   clearerr(d->file);
+
+#endif
 }
 
 long FileStream::tell() const
 {
+#ifdef _WIN32
+
+  const DWORD position = SetFilePointer(d->file, 0, NULL, FILE_CURRENT);
+  if(GetLastError() == NO_ERROR) {
+    return static_cast<long>(position);
+  }
+  else {
+    debug("File::tell() -- Failed to get the file pointer.");
+    return 0;
+  }
+
+#else
+
   return ftell(d->file);
+
+#endif
 }
 
 long FileStream::length()
 {
-  // Do some caching in case we do multiple calls.
-
-  if(d->size > 0)
-    return d->size;
-
-  if(!d->file)
+  if(!isOpen()) {
+    debug("File::length() -- invalid file.");
     return 0;
+  }
 
-  long curpos = tell();
+#ifdef _WIN32
+
+  const DWORD fileSize = GetFileSize(d->file, NULL);
+  if(GetLastError() == NO_ERROR) {
+    return static_cast<ulong>(fileSize);
+  }
+  else {
+    debug("File::length() -- Failed to get the file size.");
+    return 0;
+  }
+
+#else
+
+  const long curpos = tell();
 
   seek(0, End);
-  long endpos = tell();
+  const long endpos = tell();
 
   seek(curpos, Beginning);
 
-  d->size = endpos;
   return endpos;
+
+#endif
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -398,20 +464,29 @@ long FileStream::length()
 
 void FileStream::truncate(long length)
 {
+#ifdef _WIN32
 
-#if defined(_MSC_VER) && (_MSC_VER >= 1400)  // VC++2005 or later
+  const long currentPos = tell();
 
-  ftruncate(_fileno(d->file), length);
+  seek(length);
+  SetEndOfFile(d->file);
+  if(GetLastError() != NO_ERROR) {
+    debug("File::truncate() -- Failed to truncate the file.");
+  }
+
+  seek(currentPos);
 
 #else
 
-  ftruncate(fileno(d->file), length);
+  const int error = ftruncate(fileno(d->file), length);
+  if(error != 0) {
+    debug("FileStream::truncate() -- Coundn't truncate the file.");
+  }
 
 #endif
-
 }
 
 TagLib::uint FileStream::bufferSize()
 {
-  return FileStreamPrivate::bufferSize;
+  return BufferSize;
 }
diff --git a/3rdparty/taglib/toolkit/tiostream.cpp b/3rdparty/taglib/toolkit/tiostream.cpp
index e7b60dd75..c2ad29125 100644
--- a/3rdparty/taglib/toolkit/tiostream.cpp
+++ b/3rdparty/taglib/toolkit/tiostream.cpp
@@ -27,6 +27,111 @@
 
 using namespace TagLib;
 
+#ifdef _WIN32
+
+# include "tstring.h"
+# include "tdebug.h"
+# include <windows.h>
+
+namespace 
+{
+  // Check if the running system has CreateFileW() function.
+  // Windows9x systems don't have CreateFileW() or can't accept Unicode file names. 
+
+  bool supportsUnicode()
+  {
+    const FARPROC p = GetProcAddress(GetModuleHandleA("kernel32"), "CreateFileW");
+    return (p != NULL);
+  }
+
+  // Indicates whether the system supports Unicode file names.
+  
+  const bool SystemSupportsUnicode = supportsUnicode(); 
+
+  // Converts a UTF-16 string into a local encoding.
+  // This function should only be used in Windows9x systems which don't support 
+  // Unicode file names.
+
+  std::string unicodeToAnsi(const wchar_t *wstr)
+  {
+    if(SystemSupportsUnicode) {
+      debug("unicodeToAnsi() - Should not be used on WinNT systems.");
+    }
+
+    const int len = WideCharToMultiByte(CP_ACP, 0, wstr, -1, NULL, 0, NULL, NULL);
+    if(len == 0)
+      return std::string();
+
+    std::string str(len, '\0');
+    WideCharToMultiByte(CP_ACP, 0, wstr, -1, &str[0], len, NULL, NULL);
+
+    return str;
+  }
+}
+
+// If WinNT, stores a Unicode string into m_wname directly.
+// If Win9x, converts and stores it into m_name to avoid calling Unicode version functions.
+
+FileName::FileName(const wchar_t *name) 
+  : m_name (SystemSupportsUnicode ? "" : unicodeToAnsi(name))
+  , m_wname(SystemSupportsUnicode ? name : L"")
+{
+}
+
+FileName::FileName(const char *name) 
+  : m_name(name) 
+{
+}
+
+FileName::FileName(const FileName &name) 
+  : m_name (name.m_name) 
+  , m_wname(name.m_wname)
+{
+}
+
+FileName::operator const wchar_t *() const 
+{ 
+  return m_wname.c_str(); 
+}
+
+FileName::operator const char *() const 
+{ 
+  return m_name.c_str(); 
+}
+
+const std::wstring &FileName::wstr() const 
+{ 
+  return m_wname; 
+}
+
+const std::string &FileName::str() const 
+{ 
+  return m_name; 
+}  
+
+String FileName::toString() const
+{
+  if(!m_wname.empty()) {
+    return String(m_wname);
+  } 
+  else if(!m_name.empty()) {
+    const int len = MultiByteToWideChar(CP_ACP, 0, m_name.c_str(), -1, NULL, 0);
+    if(len == 0)
+      return String::null;
+
+    std::vector<wchar_t> buf(len);
+    MultiByteToWideChar(CP_ACP, 0, m_name.c_str(), -1, &buf[0], len);
+
+    return String(&buf[0]);
+  }
+  else {
+    return String::null;
+  }
+}
+
+
+#endif  // _WIN32
+
 ////////////////////////////////////////////////////////////////////////////////
 // public members
 ////////////////////////////////////////////////////////////////////////////////
diff --git a/3rdparty/taglib/toolkit/tiostream.h b/3rdparty/taglib/toolkit/tiostream.h
index 3e7de22aa..868269644 100644
--- a/3rdparty/taglib/toolkit/tiostream.h
+++ b/3rdparty/taglib/toolkit/tiostream.h
@@ -36,13 +36,22 @@ namespace TagLib {
   class TAGLIB_EXPORT FileName
   {
   public:
-    FileName(const wchar_t *name) : m_wname(name) {}
-    FileName(const char *name) : m_name(name) {}
-    operator const wchar_t *() const { return m_wname.c_str(); }
-    operator const char *() const { return m_name.c_str(); }
+    FileName(const wchar_t *name);
+    FileName(const char *name);
+
+    FileName(const FileName &name);
+
+    operator const wchar_t *() const;
+    operator const char *() const;
+
+    const std::wstring &wstr() const;
+    const std::string  &str() const; 
+
+    String toString() const;
+
   private:
-    std::string m_name;
-    std::wstring m_wname;
+    const std::string  m_name;
+    const std::wstring m_wname;
   };
 #else
   typedef const char *FileName;
diff --git a/3rdparty/taglib/toolkit/tlist.tcc b/3rdparty/taglib/toolkit/tlist.tcc
index 37817f059..81e291533 100644
--- a/3rdparty/taglib/toolkit/tlist.tcc
+++ b/3rdparty/taglib/toolkit/tlist.tcc
@@ -24,6 +24,7 @@
  ***************************************************************************/
 
 #include <algorithm>
+#include "trefcounter.h"
 
 namespace TagLib {
 
diff --git a/3rdparty/taglib/toolkit/tmap.tcc b/3rdparty/taglib/toolkit/tmap.tcc
index 0f2b99332..d6f51c071 100644
--- a/3rdparty/taglib/toolkit/tmap.tcc
+++ b/3rdparty/taglib/toolkit/tmap.tcc
@@ -23,6 +23,8 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
+#include "trefcounter.h"
+
 namespace TagLib {
 
 ////////////////////////////////////////////////////////////////////////////////
diff --git a/3rdparty/taglib/toolkit/tpropertymap.h b/3rdparty/taglib/toolkit/tpropertymap.h
index 7f59b21ea..2be49ddb0 100644
--- a/3rdparty/taglib/toolkit/tpropertymap.h
+++ b/3rdparty/taglib/toolkit/tpropertymap.h
@@ -40,6 +40,65 @@ namespace TagLib {
    * Note that most metadata formats pose additional conditions on the tag keys. The
    * most popular ones (Vorbis, APE, ID3v2) should support all ASCII only words of
    * length between 2 and 16.
+   * 
+   * This class can contain any tags, but here is a list of "well-known" tags that
+   * you might want to use:
+   *
+   * Basic tags:
+   *
+   *  - TITLE
+   *  - ALBUM
+   *  - ARTIST
+   *  - ALBUMARTIST
+   *  - SUBTITLE
+   *  - TRACKNUMBER
+   *  - DISCNUMBER
+   *  - DATE
+   *  - ORIGINALDATE
+   *  - GENRE
+   *  - COMMENT
+   *
+   * Sort names:
+   *
+   *  - TITLESORT
+   *  - ALBUMSORT
+   *  - ARTISTSORT
+   *  - ALBUMARTISTSORT
+   *
+   * Credits:
+   *
+   *  - COMPOSER
+   *  - LYRICIST
+   *  - CONDUCTOR
+   *  - REMIXER
+   *  - PERFORMER:<XXXX>
+   *
+   * Other tags:
+   *
+   *  - ISRC
+   *  - ASIN
+   *  - BPM
+   *  - COPYRIGHT
+   *  - ENCODEDBY
+   *  - MOOD
+   *  - COMMENT 
+   *  - MEDIA
+   *  - LABEL
+   *  - CATALOGNUMBER
+   *  - BARCODE
+   *
+   * MusicBrainz identifiers:
+   * 
+   *  - MUSICBRAINZ_TRACKID
+   *  - MUSICBRAINZ_ALBUMID
+   *  - MUSICBRAINZ_RELEASEGROUPID
+   *  - MUSICBRAINZ_WORKID
+   *  - MUSICBRAINZ_ARTISTID
+   *  - MUSICBRAINZ_ALBUMARTISTID
+   *  - ACOUSTID_ID
+   *  - ACOUSTID_FINGERPRINT
+   *  - MUSICIP_PUID
+   *
    */
 
   class TAGLIB_EXPORT PropertyMap: public SimplePropertyMap
diff --git a/3rdparty/taglib/toolkit/trefcounter.cpp b/3rdparty/taglib/toolkit/trefcounter.cpp
new file mode 100644
index 000000000..71f3c2f28
--- /dev/null
+++ b/3rdparty/taglib/toolkit/trefcounter.cpp
@@ -0,0 +1,108 @@
+/***************************************************************************
+    copyright            : (C) 2013 by Tsuda Kageyu
+    email                : tsuda.kageyu@gmail.com
+ ***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif
+
+#include "trefcounter.h"
+
+#if defined(HAVE_STD_ATOMIC)
+# include <atomic>
+# define ATOMIC_INT std::atomic<unsigned int>
+# define ATOMIC_INC(x) x.fetch_add(1)
+# define ATOMIC_DEC(x) (x.fetch_sub(1) - 1)
+#elif defined(HAVE_BOOST_ATOMIC)
+# include <boost/atomic.hpp>
+# define ATOMIC_INT boost::atomic<unsigned int>
+# define ATOMIC_INC(x) x.fetch_add(1)
+# define ATOMIC_DEC(x) (x.fetch_sub(1) - 1)
+#elif defined(HAVE_GCC_ATOMIC)
+# define ATOMIC_INT int
+# define ATOMIC_INC(x) __sync_add_and_fetch(&x, 1)
+# define ATOMIC_DEC(x) __sync_sub_and_fetch(&x, 1)
+#elif defined(HAVE_WIN_ATOMIC)
+# if !defined(NOMINMAX)
+#   define NOMINMAX
+# endif
+# include <windows.h>
+# define ATOMIC_INT long
+# define ATOMIC_INC(x) InterlockedIncrement(&x)
+# define ATOMIC_DEC(x) InterlockedDecrement(&x)
+#elif defined(HAVE_MAC_ATOMIC)
+# include <libkern/OSAtomic.h>
+# define ATOMIC_INT int32_t
+# define ATOMIC_INC(x) OSAtomicIncrement32Barrier(&x)
+# define ATOMIC_DEC(x) OSAtomicDecrement32Barrier(&x)
+#elif defined(HAVE_IA64_ATOMIC)
+# include <ia64intrin.h>
+# define ATOMIC_INT int
+# define ATOMIC_INC(x) __sync_add_and_fetch(&x, 1)
+# define ATOMIC_DEC(x) __sync_sub_and_fetch(&x, 1)
+#else
+# define ATOMIC_INT int
+# define ATOMIC_INC(x) (++x)
+# define ATOMIC_DEC(x) (--x)
+#endif
+
+namespace TagLib
+{
+  class RefCounter::RefCounterPrivate
+  {
+  public:
+    RefCounterPrivate() : refCount(1) {}
+
+    void ref() { ATOMIC_INC(refCount); }
+    bool deref() { return (ATOMIC_DEC(refCount) == 0); }
+    int count() const { return refCount; }
+
+    volatile ATOMIC_INT refCount;
+  };
+
+  RefCounter::RefCounter()
+    : d(new RefCounterPrivate())
+  {
+  }
+
+  RefCounter::~RefCounter()
+  {
+    delete d;
+  }
+
+  void RefCounter::ref() 
+  { 
+    d->ref(); 
+  }
+
+  bool RefCounter::deref() 
+  { 
+    return d->deref(); 
+  }
+
+  int RefCounter::count() const 
+  { 
+    return d->count(); 
+  }
+}
diff --git a/3rdparty/taglib/toolkit/trefcounter.h b/3rdparty/taglib/toolkit/trefcounter.h
new file mode 100644
index 000000000..53c6f1c05
--- /dev/null
+++ b/3rdparty/taglib/toolkit/trefcounter.h
@@ -0,0 +1,58 @@
+/***************************************************************************
+    copyright            : (C) 2013 by Tsuda Kageyu
+    email                : tsuda.kageyu@gmail.com
+ ***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#ifndef TAGLIB_REFCOUNTER_H
+#define TAGLIB_REFCOUNTER_H
+
+#include "taglib_export.h"
+#include "taglib.h"
+
+#ifndef DO_NOT_DOCUMENT // Tell Doxygen to skip this class.
+/*!
+  * \internal
+  * This is just used as a base class for shared classes in TagLib.
+  *
+  * \warning This <b>is not</b> part of the TagLib public API!
+  */
+namespace TagLib
+{
+  class TAGLIB_EXPORT RefCounter
+  {
+  public:
+    RefCounter();
+    virtual ~RefCounter();
+
+    void ref();
+    bool deref();
+    int count() const;
+
+  private:
+    class RefCounterPrivate;
+    RefCounterPrivate *d;
+  };
+}
+
+#endif // DO_NOT_DOCUMENT
+#endif
diff --git a/3rdparty/taglib/toolkit/tstring.cpp b/3rdparty/taglib/toolkit/tstring.cpp
index 292f35333..c83118d12 100644
--- a/3rdparty/taglib/toolkit/tstring.cpp
+++ b/3rdparty/taglib/toolkit/tstring.cpp
@@ -23,176 +23,251 @@
  *   http://www.mozilla.org/MPL/                                           *
  ***************************************************************************/
 
+// This class assumes that std::basic_string<T> has a contiguous and null-terminated buffer.
+
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif
+
 #include "tstring.h"
-#include "unicode.h"
 #include "tdebug.h"
 #include "tstringlist.h"
+#include "trefcounter.h"
+#include "tutils.h"
 
 #include <iostream>
+#include <cstdio>
+#include <cstring>
 
-#include <string.h>
+#ifdef HAVE_STD_CODECVT
+# include <codecvt>
+#else
+# include "unicode.h"
+#endif
 
-namespace TagLib {
-
-  inline unsigned short byteSwap(unsigned short x)
-  {
-    return (((x) >> 8) & 0xff) | (((x) & 0xff) << 8);
-  }
+namespace 
+{
 
   inline unsigned short combine(unsigned char c1, unsigned char c2)
   {
     return (c1 << 8) | c2;
   }
+
+  void UTF16toUTF8(const wchar_t *src, size_t srcLength, char *dst, size_t dstLength)
+  {
+#ifdef HAVE_STD_CODECVT
+
+    typedef std::codecvt_utf8_utf16<wchar_t> utf8_utf16_t;
+
+    using namespace TagLib;
+
+    const wchar_t *srcBegin = src;
+    const wchar_t *srcEnd   = srcBegin + srcLength;
+
+    char *dstBegin = dst;
+    char *dstEnd   = dstBegin + dstLength;
+
+    std::mbstate_t st;
+    const wchar_t *source;
+    char *target;
+    memset(&st, 0, sizeof(st));
+    std::codecvt_base::result result = utf8_utf16_t().out(
+      st, srcBegin, srcEnd, source, dstBegin, dstEnd, target);
+
+    if(result != utf8_utf16_t::ok) {
+      debug("String::copyFromUTF8() - Unicode conversion error.");
+    }
+
+#else
+
+    using namespace Unicode;
+    using namespace TagLib;
+
+    const Unicode::UTF16 *srcBegin = src;
+    const Unicode::UTF16 *srcEnd   = srcBegin + srcLength;
+
+    Unicode::UTF8 *dstBegin = reinterpret_cast<Unicode::UTF8*>(dst);
+    Unicode::UTF8 *dstEnd   = dstBegin + dstLength;
+
+    Unicode::ConversionResult result = Unicode::ConvertUTF16toUTF8(
+      &srcBegin, srcEnd, &dstBegin, dstEnd, Unicode::lenientConversion);
+
+    if(result != Unicode::conversionOK) {
+      debug("String::to8Bit() - Unicode conversion error.");
+    }
+
+#endif
+  }
+
+  void UTF8toUTF16(const char *src, size_t srcLength, wchar_t *dst, size_t dstLength)
+  {
+#ifdef HAVE_STD_CODECVT
+
+    typedef std::codecvt_utf8_utf16<wchar_t> utf8_utf16_t;
+
+    using namespace TagLib;
+
+    const char *srcBegin = src;
+    const char *srcEnd   = srcBegin + srcLength;
+
+    wchar_t *dstBegin = dst;
+    wchar_t *dstEnd   = dstBegin + dstLength;
+
+    std::mbstate_t st;
+    const char *source;
+    wchar_t *target;
+    memset(&st, 0, sizeof(st));
+    std::codecvt_base::result result = utf8_utf16_t().in(
+      st, srcBegin, srcEnd, source, dstBegin, dstEnd, target);
+
+    if(result != utf8_utf16_t::ok) {
+      debug("String::copyFromUTF8() - Unicode conversion error.");
+    }
+
+#else
+
+    using namespace Unicode;
+    using namespace TagLib;
+
+    const Unicode::UTF8 *srcBegin = reinterpret_cast<const Unicode::UTF8*>(src);
+    const Unicode::UTF8 *srcEnd   = srcBegin + srcLength;
+
+    Unicode::UTF16 *dstBegin = dst;
+    Unicode::UTF16 *dstEnd   = dstBegin + dstLength;
+
+    Unicode::ConversionResult result = Unicode::ConvertUTF8toUTF16(
+      &srcBegin, srcEnd, &dstBegin, dstEnd, Unicode::lenientConversion);
+
+    if(result != Unicode::conversionOK) {
+      debug("String::copyFromUTF8() - Unicode conversion error.");
+    }
+
+#endif 
+  }
 }
 
-using namespace TagLib;
+namespace TagLib {
 
 class String::StringPrivate : public RefCounter
 {
 public:
-  StringPrivate(const wstring &s) :
-    RefCounter(),
-    data(s),
-    CString(0) {}
-
-  StringPrivate() :
-    RefCounter(),
-    CString(0) {}
-
-  ~StringPrivate() {
-    delete [] CString;
+  StringPrivate() 
+    : RefCounter() 
+  {
   }
 
-  wstring data;
+  StringPrivate(const wstring &s) 
+    : RefCounter()
+    , data(s) 
+  {
+  }
+  
+  StringPrivate(uint n, wchar_t c) 
+    : RefCounter()
+    , data(static_cast<size_t>(n), c) 
+  {
+  }
 
   /*!
-   * This is only used to hold the a pointer to the most recent value of
-   * toCString.
+   * Stores string in UTF-16. The byte order depends on the CPU endian. 
    */
-  char *CString;
+  TagLib::wstring data;
+
+  /*!
+   * This is only used to hold the the most recent value of toCString().
+   */
+  std::string cstring;
 };
 
 String String::null;
 
 ////////////////////////////////////////////////////////////////////////////////
 
-String::String()
+String::String() 
+  : d(new StringPrivate())
 {
-  d = new StringPrivate;
 }
 
-String::String(const String &s) : d(s.d)
+String::String(const String &s) 
+  : d(s.d)
 {
   d->ref();
 }
 
 String::String(const std::string &s, Type t)
+  : d(new StringPrivate())
 {
-  d = new StringPrivate;
-
-  if(t == UTF16 || t == UTF16BE || t == UTF16LE) {
+  if(t == Latin1)
+    copyFromLatin1(&s[0], s.length());
+  else if(t == String::UTF8)
+    copyFromUTF8(&s[0], s.length());
+  else {
     debug("String::String() -- A std::string should not contain UTF16.");
-    return;
   }
-
-  int length = s.length();
-  d->data.resize(length);
-  wstring::iterator targetIt = d->data.begin();
-
-  for(std::string::const_iterator it = s.begin(); it != s.end(); it++) {
-    *targetIt = uchar(*it);
-    ++targetIt;
-  }
-
-  prepare(t);
 }
 
 String::String(const wstring &s, Type t)
+  : d(new StringPrivate())
 {
-  d = new StringPrivate(s);
-  prepare(t);
+  if(t == UTF16 || t == UTF16BE || t == UTF16LE)
+    copyFromUTF16(s.c_str(), s.length(), t);
+  else {
+    debug("String::String() -- A TagLib::wstring should not contain Latin1 or UTF-8.");
+  }
 }
 
 String::String(const wchar_t *s, Type t)
+  : d(new StringPrivate())
 {
-  d = new StringPrivate(s);
-  prepare(t);
+  if(t == UTF16 || t == UTF16BE || t == UTF16LE)
+    copyFromUTF16(s, ::wcslen(s), t);
+  else {
+    debug("String::String() -- A const wchar_t * should not contain Latin1 or UTF-8.");
+  }
 }
 
 String::String(const char *s, Type t)
+  : d(new StringPrivate())
 {
-  d = new StringPrivate;
-
-  if(t == UTF16 || t == UTF16BE || t == UTF16LE) {
+  if(t == Latin1)
+    copyFromLatin1(s, ::strlen(s));
+  else if(t == String::UTF8)
+    copyFromUTF8(s, ::strlen(s));
+  else {
     debug("String::String() -- A const char * should not contain UTF16.");
-    return;
   }
-
-  int length = ::strlen(s);
-  d->data.resize(length);
-
-  wstring::iterator targetIt = d->data.begin();
-
-  for(int i = 0; i < length; i++) {
-    *targetIt = uchar(s[i]);
-    ++targetIt;
-  }
-
-  prepare(t);
 }
 
 String::String(wchar_t c, Type t)
+  : d(new StringPrivate())
 {
-  d = new StringPrivate;
-  d->data += c;
-  prepare(t);
+  if(t == UTF16 || t == UTF16BE || t == UTF16LE)
+    copyFromUTF16(&c, 1, t);
+  else {
+    debug("String::String() -- A const wchar_t should not contain Latin1 or UTF-8.");
+  }
 }
 
 String::String(char c, Type t)
+  : d(new StringPrivate(1, static_cast<uchar>(c)))
 {
-  d = new StringPrivate;
-
-  if(t == UTF16 || t == UTF16BE || t == UTF16LE) {
-    debug("String::String() -- A std::string should not contain UTF16.");
-    return;
+  if(t != Latin1 && t != UTF8) {
+    debug("String::String() -- A char should not contain UTF16.");
   }
-
-  d->data += uchar(c);
-  prepare(t);
 }
 
 String::String(const ByteVector &v, Type t)
+  : d(new StringPrivate())
 {
-  d = new StringPrivate;
-
   if(v.isEmpty())
     return;
 
-  if(t == Latin1 || t == UTF8) {
-
-    int length = 0;
-    d->data.resize(v.size());
-    wstring::iterator targetIt = d->data.begin();
-    for(ByteVector::ConstIterator it = v.begin(); it != v.end() && (*it); ++it) {
-      *targetIt = uchar(*it);
-      ++targetIt;
-      ++length;
-    }
-    d->data.resize(length);
-  }
-  else  {
-    d->data.resize(v.size() / 2);
-    wstring::iterator targetIt = d->data.begin();
-
-    for(ByteVector::ConstIterator it = v.begin();
-        it != v.end() && it + 1 != v.end() && combine(*it, *(it + 1));
-        it += 2)
-    {
-      *targetIt = combine(*it, *(it + 1));
-      ++targetIt;
-    }
-  }
-  prepare(t);
+  if(t == Latin1) 
+    copyFromLatin1(v.data(), v.size());
+  else if(t == UTF8) 
+    copyFromUTF8(v.data(), v.size());
+  else 
+    copyFromUTF16(v.data(), v.size(), t);
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -206,46 +281,23 @@ String::~String()
 std::string String::to8Bit(bool unicode) const
 {
   std::string s;
-  s.resize(d->data.size());
 
   if(!unicode) {
+    s.resize(d->data.size());
+
     std::string::iterator targetIt = s.begin();
     for(wstring::const_iterator it = d->data.begin(); it != d->data.end(); it++) {
-      *targetIt = char(*it);
+      *targetIt = static_cast<char>(*it);
       ++targetIt;
     }
-    return s;
   }
+  else {
+    s.resize(d->data.size() * 4 + 1);
 
-  const int outputBufferSize = d->data.size() * 3 + 1;
-
-  Unicode::UTF16 *sourceBuffer = new Unicode::UTF16[d->data.size() + 1];
-  Unicode::UTF8  *targetBuffer = new Unicode::UTF8[outputBufferSize];
-
-  for(unsigned int i = 0; i < d->data.size(); i++)
-    sourceBuffer[i] = Unicode::UTF16(d->data[i]);
-
-  const Unicode::UTF16 *source = sourceBuffer;
-  Unicode::UTF8 *target = targetBuffer;
-
-  Unicode::ConversionResult result =
-    Unicode::ConvertUTF16toUTF8(&source, sourceBuffer + d->data.size(),
-                                &target, targetBuffer + outputBufferSize,
-                                Unicode::lenientConversion);
-
-  if(result != Unicode::conversionOK) {
-    debug("String::to8Bit() - Unicode conversion error.");
+    UTF16toUTF8(&d->data[0], d->data.size(), &s[0], s.size());
+    s.resize(::strlen(s.c_str()));
   }
 
-  int newSize = target - targetBuffer;
-  s.resize(newSize);
-  targetBuffer[newSize] = 0;
-
-  s = (char *) targetBuffer;
-
-  delete [] sourceBuffer;
-  delete [] targetBuffer;
-
   return s;
 }
 
@@ -256,22 +308,13 @@ TagLib::wstring String::toWString() const
 
 const char *String::toCString(bool unicode) const
 {
-  delete [] d->CString;
+  d->cstring = to8Bit(unicode);
+  return d->cstring.c_str();
+}
 
-  std::string buffer = to8Bit(unicode);
-  d->CString = new char[buffer.size() + 1];
-
-#if defined(_MSC_VER) && (_MSC_VER >= 1400)  // VC++2005 or later
-
-  strcpy_s(d->CString, buffer.size() + 1, buffer.c_str());
-
-#else
-
-  strcpy(d->CString, buffer.c_str());
-
-#endif                                          
-
-  return d->CString;
+const wchar_t *String::toCWString() const
+{
+  return d->data.c_str();
 }
 
 String::Iterator String::begin()
@@ -296,23 +339,12 @@ String::ConstIterator String::end() const
 
 int String::find(const String &s, int offset) const
 {
-  wstring::size_type position = d->data.find(s.d->data, offset);
-
-  if(position != wstring::npos)
-    return position;
-  else
-    return -1;
+  return d->data.find(s.d->data, offset);
 }
 
 int String::rfind(const String &s, int offset) const
 {
-  wstring::size_type position =
-    d->data.rfind(s.d->data, offset == -1 ? wstring::npos : offset);
-
-  if(position != wstring::npos)
-    return position;
-  else
-    return -1;
+  return d->data.rfind(s.d->data, offset);
 }
 
 StringList String::split(const String &separator) const
@@ -345,9 +377,7 @@ bool String::startsWith(const String &s) const
 
 String String::substr(uint position, uint n) const
 {
-  String s;
-  s.d->data = d->data.substr(position, n);
-  return s;
+  return String(d->data.substr(position, n));
 }
 
 String &String::append(const String &s)
@@ -395,67 +425,75 @@ bool String::isNull() const
 
 ByteVector String::data(Type t) const
 {
-  ByteVector v;
-
-  switch(t) {
-
+  switch(t) 
+  {
   case Latin1:
-  {
-    for(wstring::const_iterator it = d->data.begin(); it != d->data.end(); it++)
-      v.append(char(*it));
-    break;
-  }
+    {
+      ByteVector v(size(), 0);
+      char *p = v.data();
+
+      for(wstring::const_iterator it = d->data.begin(); it != d->data.end(); it++)
+        *p++ = static_cast<char>(*it);
+
+      return v;
+    }
   case UTF8:
-  {
-    std::string s = to8Bit(true);
-    v.setData(s.c_str(), s.length());
-    break;
-  }
+    {
+      ByteVector v(size() * 4 + 1, 0);
+
+      UTF16toUTF8(&d->data[0], d->data.size(), v.data(), v.size());
+      v.resize(::strlen(v.data()));
+
+      return v;
+    }
   case UTF16:
-  {
-    // Assume that if we're doing UTF16 and not UTF16BE that we want little
-    // endian encoding.  (Byte Order Mark)
+    {
+      ByteVector v(2 + size() * 2, 0);
+      char *p = v.data();
 
-    v.append(char(0xff));
-    v.append(char(0xfe));
+      // Assume that if we're doing UTF16 and not UTF16BE that we want little
+      // endian encoding.  (Byte Order Mark)
 
-    for(wstring::const_iterator it = d->data.begin(); it != d->data.end(); it++) {
+      *p++ = '\xff';
+      *p++ = '\xfe';
 
-      char c1 = *it & 0xff;
-      char c2 = *it >> 8;
+      for(wstring::const_iterator it = d->data.begin(); it != d->data.end(); it++) {
+        *p++ = static_cast<char>(*it & 0xff);
+        *p++ = static_cast<char>(*it >> 8);
+      }
 
-      v.append(c1);
-      v.append(c2);
+      return v;
     }
-    break;
-  }
   case UTF16BE:
-  {
-    for(wstring::const_iterator it = d->data.begin(); it != d->data.end(); it++) {
+    {
+      ByteVector v(size() * 2, 0);
+      char *p = v.data();
 
-      char c1 = *it >> 8;
-      char c2 = *it & 0xff;
+      for(wstring::const_iterator it = d->data.begin(); it != d->data.end(); it++) {
+        *p++ = static_cast<char>(*it >> 8);
+        *p++ = static_cast<char>(*it & 0xff);
+      }
 
-      v.append(c1);
-      v.append(c2);
+      return v;
     }
-    break;
-  }
   case UTF16LE:
-  {
-    for(wstring::const_iterator it = d->data.begin(); it != d->data.end(); it++) {
+    {
+      ByteVector v(size() * 2, 0);
+      char *p = v.data();
 
-      char c1 = *it & 0xff;
-      char c2 = *it >> 8;
+      for(wstring::const_iterator it = d->data.begin(); it != d->data.end(); it++) {
+        *p++ = static_cast<char>(*it & 0xff);
+        *p++ = static_cast<char>(*it >> 8);
+      }
 
-      v.append(c1);
-      v.append(c2);
+      return v;
+    }
+  default:
+    {
+      debug("String::data() - Invalid Type value.");
+      return ByteVector();
     }
-    break;
   }
-  }
-
-  return v;
 }
 
 int String::toInt() const
@@ -530,37 +568,35 @@ bool String::isAscii() const
 
 String String::number(int n) // static
 {
-  if(n == 0)
-    return String("0");
+  static const size_t BufferSize = 11; // Sufficient to store "-214748364".
+  static const char *Format = "%d";
 
-  String charStack;
+  char buffer[BufferSize];
+  int length;
 
-  bool negative = n < 0;
+#if defined(HAVE_SNPRINTF)
 
-  if(negative)
-    n = n * -1;
+  length = snprintf(buffer, BufferSize, Format, n);
 
-  while(n > 0) {
-    int remainder = n % 10;
-    charStack += char(remainder + '0');
-    n = (n - remainder) / 10;
-  }
+#elif defined(HAVE_SPRINTF_S)
 
-  String s;
+  length = sprintf_s(buffer, Format, n);
 
-  if(negative)
-    s += '-';
+#else
 
-  for(int i = charStack.d->data.size() - 1; i >= 0; i--)
-    s += charStack.d->data[i];
+  length = sprintf(buffer, Format, n);
 
-  return s;
+#endif
+
+  if(length > 0)
+    return String(buffer);
+  else
+    return String::null;
 }
 
 TagLib::wchar &String::operator[](int i)
 {
   detach();
-
   return d->data[i];
 }
 
@@ -638,14 +674,7 @@ String &String::operator=(const std::string &s)
     delete d;
 
   d = new StringPrivate;
-
-  d->data.resize(s.size());
-
-  wstring::iterator targetIt = d->data.begin();
-  for(std::string::const_iterator it = s.begin(); it != s.end(); it++) {
-    *targetIt = uchar(*it);
-    ++targetIt;
-  }
+  copyFromLatin1(s.c_str(), s.length());
 
   return *this;
 }
@@ -662,6 +691,7 @@ String &String::operator=(const wchar_t *s)
 {
   if(d->deref())
     delete d;
+
   d = new StringPrivate(s);
   return *this;
 }
@@ -670,8 +700,8 @@ String &String::operator=(char c)
 {
   if(d->deref())
     delete d;
-  d = new StringPrivate;
-  d->data += uchar(c);
+
+  d = new StringPrivate(1, static_cast<uchar>(c));
   return *this;
 }
 
@@ -679,8 +709,8 @@ String &String::operator=(wchar_t c)
 {
   if(d->deref())
     delete d;
-  d = new StringPrivate;
-  d->data += c;
+
+  d = new StringPrivate(1, c);
   return *this;
 }
 
@@ -690,15 +720,7 @@ String &String::operator=(const char *s)
     delete d;
 
   d = new StringPrivate;
-
-  int length = ::strlen(s);
-  d->data.resize(length);
-
-  wstring::iterator targetIt = d->data.begin();
-  for(int i = 0; i < length; i++) {
-    *targetIt = uchar(s[i]);
-    ++targetIt;
-  }
+  copyFromLatin1(s, ::strlen(s));
 
   return *this;
 }
@@ -709,20 +731,10 @@ String &String::operator=(const ByteVector &v)
     delete d;
 
   d = new StringPrivate;
-  d->data.resize(v.size());
-  wstring::iterator targetIt = d->data.begin();
-
-  uint i = 0;
-
-  for(ByteVector::ConstIterator it = v.begin(); it != v.end() && (*it); ++it) {
-    *targetIt = uchar(*it);
-    ++targetIt;
-    ++i;
-  }
+  copyFromLatin1(v.data(), v.size());
 
   // If we hit a null in the ByteVector, shrink the string again.
-
-  d->data.resize(i);
+  d->data.resize(::wcslen(d->data.c_str()));
 
   return *this;
 }
@@ -748,68 +760,95 @@ void String::detach()
 // private members
 ////////////////////////////////////////////////////////////////////////////////
 
-void String::prepare(Type t)
+void String::copyFromLatin1(const char *s, size_t length)
 {
-  switch(t) {
-  case UTF16:
-  {
-    if(d->data.size() >= 1 && (d->data[0] == 0xfeff || d->data[0] == 0xfffe)) {
-      bool swap = d->data[0] != 0xfeff;
-      d->data.erase(d->data.begin(), d->data.begin() + 1);
-      if(swap) {
-        for(uint i = 0; i < d->data.size(); i++)
-          d->data[i] = byteSwap((unsigned short)d->data[i]);
-      }
-    }
+  d->data.resize(length);
+
+  for(size_t i = 0; i < length; ++i)
+    d->data[i] = static_cast<uchar>(s[i]);
+}
+
+void String::copyFromUTF8(const char *s, size_t length)
+{
+  d->data.resize(length);
+
+  UTF8toUTF16(s, length, &d->data[0], d->data.size());
+  d->data.resize(::wcslen(d->data.c_str()));
+}
+
+void String::copyFromUTF16(const wchar_t *s, size_t length, Type t)
+{
+  bool swap;
+  if(t == UTF16) {
+    if(length >= 1 && s[0] == 0xfeff) 
+      swap = false; // Same as CPU endian. No need to swap bytes.
+    else if(length >= 1 && s[0] == 0xfffe) 
+      swap = true;  // Not same as CPU endian. Need to swap bytes.
     else {
-      debug("String::prepare() - Invalid UTF16 string.");
-      d->data.erase(d->data.begin(), d->data.end());
-    }
-    break;
-  }
-  case UTF8:
-  {
-    int bufferSize = d->data.size() + 1;
-    Unicode::UTF8  *sourceBuffer = new Unicode::UTF8[bufferSize];
-    Unicode::UTF16 *targetBuffer = new Unicode::UTF16[bufferSize];
-
-    unsigned int i = 0;
-    for(; i < d->data.size(); i++)
-      sourceBuffer[i] = Unicode::UTF8(d->data[i]);
-    sourceBuffer[i] = 0;
-
-    const Unicode::UTF8 *source = sourceBuffer;
-    Unicode::UTF16 *target = targetBuffer;
-
-    Unicode::ConversionResult result =
-      Unicode::ConvertUTF8toUTF16(&source, sourceBuffer + bufferSize,
-                                  &target, targetBuffer + bufferSize,
-                                  Unicode::lenientConversion);
-
-    if(result != Unicode::conversionOK) {
-      debug("String::prepare() - Unicode conversion error.");
+      debug("String::copyFromUTF16() - Invalid UTF16 string.");
+      return;
     }
 
-    int newSize = target != targetBuffer ? target - targetBuffer - 1 : 0;
-    d->data.resize(newSize);
-
-    for(int i = 0; i < newSize; i++)
-      d->data[i] = targetBuffer[i];
-
-    delete [] sourceBuffer;
-    delete [] targetBuffer;
-
-    break;
+    s++;
+    length--;
   }
-  case UTF16LE:
-  {
-    for(uint i = 0; i < d->data.size(); i++)
-      d->data[i] = byteSwap((unsigned short)d->data[i]);
-    break;
+  else 
+    swap = (t != WCharByteOrder);
+
+  d->data.resize(length);
+  memcpy(&d->data[0], s, length * sizeof(wchar_t));
+
+  if(swap) {
+    for(size_t i = 0; i < length; ++i)
+      d->data[i] = byteSwap(static_cast<ushort>(s[i]));
   }
-  default:
-    break;
+}
+
+void String::copyFromUTF16(const char *s, size_t length, Type t)
+{
+  bool swap;
+  if(t == UTF16) {
+    if(length < 2) {
+      debug("String::copyFromUTF16() - Invalid UTF16 string.");
+      return;
+    }
+
+    // Uses memcpy instead of reinterpret_cast to avoid an alignment exception.
+    ushort bom;
+    ::memcpy(&bom, s, 2);
+
+    if(bom == 0xfeff) 
+      swap = false; // Same as CPU endian. No need to swap bytes.
+    else if(bom == 0xfffe) 
+      swap = true;  // Not same as CPU endian. Need to swap bytes.
+    else {
+      debug("String::copyFromUTF16() - Invalid UTF16 string.");
+      return;
+    }
+
+    s += 2;
+    length -= 2;
   }
+  else 
+    swap = (t != WCharByteOrder);
+
+  d->data.resize(length / 2);
+  for(size_t i = 0; i < length / 2; ++i) {
+    d->data[i] = swap ? combine(*s, *(s + 1)) : combine(*(s + 1), *s);
+    s += 2;
+  }
+}
+
+#if SYSTEM_BYTEORDER == 1
+
+const String::Type String::WCharByteOrder = String::UTF16LE;
+
+#else
+
+const String::Type String::WCharByteOrder = String::UTF16BE;
+
+#endif
+
 }
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -818,27 +857,28 @@ void String::prepare(Type t)
 
 const TagLib::String operator+(const TagLib::String &s1, const TagLib::String &s2)
 {
-  String s(s1);
+  TagLib::String s(s1);
   s.append(s2);
   return s;
 }
 
 const TagLib::String operator+(const char *s1, const TagLib::String &s2)
 {
-  String s(s1);
+  TagLib::String s(s1);
   s.append(s2);
   return s;
 }
 
 const TagLib::String operator+(const TagLib::String &s1, const char *s2)
 {
-  String s(s1);
+  TagLib::String s(s1);
   s.append(s2);
   return s;
 }
 
-std::ostream &operator<<(std::ostream &s, const String &str)
+std::ostream &operator<<(std::ostream &s, const TagLib::String &str)
 {
   s << str.to8Bit();
   return s;
 }
+
diff --git a/3rdparty/taglib/toolkit/tstring.h b/3rdparty/taglib/toolkit/tstring.h
index 759a175ae..57945beed 100644
--- a/3rdparty/taglib/toolkit/tstring.h
+++ b/3rdparty/taglib/toolkit/tstring.h
@@ -63,8 +63,8 @@ namespace TagLib {
   /*!
    * This is an implicitly shared \e wide string.  For storage it uses
    * TagLib::wstring, but as this is an <i>implementation detail</i> this of
-   * course could change.  Strings are stored internally as UTF-16BE.  (Without
-   * the BOM (Byte Order Mark)
+   * course could change.  Strings are stored internally as UTF-16(without BOM/
+   * CPU byte order)
    *
    * The use of implicit sharing means that copying a string is cheap, the only
    * \e cost comes into play when the copy is modified.  Prior to that the string
@@ -135,12 +135,12 @@ namespace TagLib {
     /*!
      * Makes a deep copy of the data in \a s.
      */
-    String(const wstring &s, Type t = UTF16BE);
+    String(const wstring &s, Type t = WCharByteOrder);
 
     /*!
      * Makes a deep copy of the data in \a s.
      */
-    String(const wchar_t *s, Type t = UTF16BE);
+    String(const wchar_t *s, Type t = WCharByteOrder);
 
     /*!
      * Makes a deep copy of the data in \a c.
@@ -178,34 +178,59 @@ namespace TagLib {
     virtual ~String();
 
     /*!
-     * If \a unicode if false (the default) this will return a \e Latin1 encoded
-     * std::string.  If it is true the returned std::wstring will be UTF-8
-     * encoded.
+     * Returns a deep copy of this String as an std::string.  The returned string 
+     * is encoded in UTF8 if \a unicode is true, otherwise Latin1.
+     *
+     * \see toCString()
      */
     std::string to8Bit(bool unicode = false) const;
 
     /*!
-     * Returns a wstring version of the TagLib string as a wide string.
+     * Returns a deep copy of this String as a wstring.  The returned string is 
+     * encoded in UTF-16 (without BOM/CPU byte order).
+     *
+     * \see toCWString()
      */
     wstring toWString() const;
 
     /*!
-     * Creates and returns a C-String based on the data.  This string is still
-     * owned by the String (class) and as such should not be deleted by the user.
+     * Creates and returns a standard C-style (null-terminated) version of this 
+     * String.  The returned string is encoded in UTF8 if \a unicode is true, 
+     * otherwise Latin1.
+     * 
+     * The returned string is still owned by this String and should not be deleted 
+     * by the user.
      *
-     * If \a unicode if false (the default) this string will be encoded in
-     * \e Latin1.  If it is true the returned C-String will be UTF-8 encoded.
+     * The returned pointer remains valid until this String instance is destroyed 
+     * or toCString() is called again.
      *
-     * This string remains valid until the String instance is destroyed or
-     * another export method is called.
-     *
-     * \warning This however has the side effect that this C-String will remain
-     * in memory <b>in addition to</b> other memory that is consumed by the
+     * \warning This however has the side effect that the returned string will remain
+     * in memory <b>in addition to</b> other memory that is consumed by this 
      * String instance.  So, this method should not be used on large strings or
-     * where memory is critical.
+     * where memory is critical.  Consider using to8Bit() instead to avoid it.
+     *
+     * \see to8Bit()
      */
     const char *toCString(bool unicode = false) const;
-
+    
+    /*!
+     * Returns a standard C-style (null-terminated) wide character version of 
+     * this String.  The returned string is encoded in UTF-16 (without BOM/CPU byte 
+     * order).
+     * 
+     * The returned string is still owned by this String and should not be deleted 
+     * by the user.
+     *
+     * The returned pointer remains valid until this String instance is destroyed 
+     * or any other method of this String is called.
+     *
+     * /note This returns a pointer to the String's internal data without any 
+     * conversions.
+     *
+     * \see toWString()
+     */
+    const wchar_t *toCWString() const;
+    
     /*!
      * Returns an iterator pointing to the beginning of the string.
      */
@@ -306,7 +331,7 @@ namespace TagLib {
     /*!
      * Convert the string to an integer.
      *
-     * Returns the integer if the conversion was successfull or 0 if the
+     * Returns the integer if the conversion was successful or 0 if the
      * string does not represent a number.
      */
     // BIC: merge with the method below
@@ -451,17 +476,39 @@ namespace TagLib {
 
   private:
     /*!
-     * This checks to see if the string is in \e UTF-16 (with BOM) or \e UTF-8
-     * format and if so converts it to \e UTF-16BE for internal use.  \e Latin1
-     * does not require conversion since it is a subset of \e UTF-16BE and
-     * \e UTF16-BE requires no conversion since it is used internally.
+     * Converts a \e Latin-1 string into \e UTF-16(without BOM/CPU byte order) 
+     * and copies it to the internal buffer.
      */
-    void prepare(Type t);
+    void copyFromLatin1(const char *s, size_t length);
+
+    /*!
+     * Converts a \e UTF-8 string into \e UTF-16(without BOM/CPU byte order) 
+     * and copies it to the internal buffer.
+     */
+    void copyFromUTF8(const char *s, size_t length);
+
+    /*!
+     * Converts a \e UTF-16 (with BOM), UTF-16LE or UTF16-BE string into 
+     * \e UTF-16(without BOM/CPU byte order) and copies it to the internal buffer.
+     */
+    void copyFromUTF16(const wchar_t *s, size_t length, Type t);
+
+    /*!
+     * Converts a \e UTF-16 (with BOM), UTF-16LE or UTF16-BE string into 
+     * \e UTF-16(without BOM/CPU byte order) and copies it to the internal buffer.
+     */
+    void copyFromUTF16(const char *s, size_t length, Type t);
+    
+    /*!
+     * Indicates which byte order of UTF-16 is used to store strings internally. 
+     *
+     * \note \e String::UTF16BE or \e String::UTF16LE
+     */
+    static const Type WCharByteOrder;
 
     class StringPrivate;
     StringPrivate *d;
   };
-
 }
 
 /*!
diff --git a/3rdparty/taglib/toolkit/tutils.h b/3rdparty/taglib/toolkit/tutils.h
new file mode 100644
index 000000000..73c35716d
--- /dev/null
+++ b/3rdparty/taglib/toolkit/tutils.h
@@ -0,0 +1,151 @@
+/***************************************************************************
+    copyright            : (C) 2013 by Tsuda Kageyu
+    email                : tsuda.kageyu@gmail.com
+ ***************************************************************************/
+
+/***************************************************************************
+ *   This library is free software; you can redistribute it and/or modify  *
+ *   it under the terms of the GNU Lesser General Public License version   *
+ *   2.1 as published by the Free Software Foundation.                     *
+ *                                                                         *
+ *   This library is distributed in the hope that it will be useful, but   *
+ *   WITHOUT ANY WARRANTY; without even the implied warranty of            *
+ *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU     *
+ *   Lesser General Public License for more details.                       *
+ *                                                                         *
+ *   You should have received a copy of the GNU Lesser General Public      *
+ *   License along with this library; if not, write to the Free Software   *
+ *   Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA         *
+ *   02110-1301  USA                                                       *
+ *                                                                         *
+ *   Alternatively, this file is available under the Mozilla Public        *
+ *   License Version 1.1.  You may obtain a copy of the License at         *
+ *   http://www.mozilla.org/MPL/                                           *
+ ***************************************************************************/
+
+#ifndef TAGLIB_TUTILS_H
+#define TAGLIB_TUTILS_H
+
+// THIS FILE IS NOT A PART OF THE TAGLIB API
+
+#ifndef DO_NOT_DOCUMENT  // tell Doxygen not to document this header
+
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif
+
+#if defined(HAVE_MSC_BYTESWAP)
+# include <stdlib.h>
+#elif defined(HAVE_GLIBC_BYTESWAP)
+# include <byteswap.h>
+#elif defined(HAVE_MAC_BYTESWAP)
+# include <libkern/OSByteOrder.h>
+#elif defined(HAVE_OPENBSD_BYTESWAP)
+# include <sys/endian.h>
+#endif
+
+namespace TagLib
+{
+
+  inline ushort byteSwap(ushort x)
+  {
+#if defined(HAVE_GCC_BYTESWAP_16)
+
+    return __builtin_bswap16(x);
+
+#elif defined(HAVE_MSC_BYTESWAP)
+
+    return _byteswap_ushort(x);
+
+#elif defined(HAVE_GLIBC_BYTESWAP)
+
+    return __bswap_16(x);
+
+#elif defined(HAVE_MAC_BYTESWAP)
+
+    return OSSwapInt16(x);
+
+#elif defined(HAVE_OPENBSD_BYTESWAP)
+
+    return swap16(x);
+
+#else
+
+    return ((x >> 8) & 0xff) | ((x & 0xff) << 8);
+
+#endif
+  }
+
+  inline uint byteSwap(uint x)
+  {
+#if defined(HAVE_GCC_BYTESWAP_32)
+
+    return __builtin_bswap32(x);
+
+#elif defined(HAVE_MSC_BYTESWAP)
+
+    return _byteswap_ulong(x);
+
+#elif defined(HAVE_GLIBC_BYTESWAP)
+
+    return __bswap_32(x);
+
+#elif defined(HAVE_MAC_BYTESWAP)
+
+    return OSSwapInt32(x);
+
+#elif defined(HAVE_OPENBSD_BYTESWAP)
+
+    return swap32(x);
+
+#else
+
+    return ((x & 0xff000000u) >> 24)
+      | ((x & 0x00ff0000u) >>  8)
+      | ((x & 0x0000ff00u) <<  8)
+      | ((x & 0x000000ffu) << 24);
+
+#endif
+  }
+
+  inline ulonglong byteSwap(ulonglong x)
+  {
+#if defined(HAVE_GCC_BYTESWAP_64)
+
+    return __builtin_bswap64(x);
+
+#elif defined(HAVE_MSC_BYTESWAP)
+
+    return _byteswap_uint64(x);
+
+#elif defined(HAVE_GLIBC_BYTESWAP)
+
+    return __bswap_64(x);
+
+#elif defined(HAVE_MAC_BYTESWAP)
+
+    return OSSwapInt64(x);
+
+#elif defined(HAVE_OPENBSD_BYTESWAP)
+
+    return swap64(x);
+
+#else
+
+    return ((x & 0xff00000000000000ull) >> 56)
+      | ((x & 0x00ff000000000000ull) >> 40)
+      | ((x & 0x0000ff0000000000ull) >> 24)
+      | ((x & 0x000000ff00000000ull) >> 8)
+      | ((x & 0x00000000ff000000ull) << 8)
+      | ((x & 0x0000000000ff0000ull) << 24)
+      | ((x & 0x000000000000ff00ull) << 40)
+      | ((x & 0x00000000000000ffull) << 56);
+
+#endif
+  }
+
+};
+
+#endif
+
+#endif
diff --git a/3rdparty/taglib/toolkit/unicode.h b/3rdparty/taglib/toolkit/unicode.h
index cf7eb3c56..b9de0ea21 100644
--- a/3rdparty/taglib/toolkit/unicode.h
+++ b/3rdparty/taglib/toolkit/unicode.h
@@ -115,8 +115,8 @@
 namespace Unicode {
 
 typedef unsigned long	UTF32;	/* at least 32 bits */
-typedef unsigned short	UTF16;	/* at least 16 bits */
-typedef unsigned char	UTF8;	/* typically 8 bits */
+typedef wchar_t	      UTF16;	/* TagLib assumes that wchar_t is sufficient for UTF-16. */
+typedef unsigned char	UTF8;	  /* typically 8 bits */
 typedef unsigned char	Boolean; /* 0 or 1 */
 
 typedef enum {
diff --git a/3rdparty/taglib/trueaudio/trueaudiofile.cpp b/3rdparty/taglib/trueaudio/trueaudiofile.cpp
index 9efc6e9de..f98d6addc 100644
--- a/3rdparty/taglib/trueaudio/trueaudiofile.cpp
+++ b/3rdparty/taglib/trueaudio/trueaudiofile.cpp
@@ -139,14 +139,17 @@ PropertyMap TrueAudio::File::properties() const
   return PropertyMap();
 }
 
-PropertyMap TrueAudio::File::setProperties(const PropertyMap &properties)
+void TrueAudio::File::removeUnsupportedProperties(const StringList &unsupported)
 {
   if(d->hasID3v2)
-    return d->tag.access<ID3v2::Tag>(TrueAudioID3v2Index, false)->setProperties(properties);
-  else if(d->hasID3v1)
-    return d->tag.access<ID3v1::Tag>(TrueAudioID3v1Index, false)->setProperties(properties);
-  else
-    return d->tag.access<ID3v2::Tag>(TrueAudioID3v2Index, true)->setProperties(properties);
+    d->tag.access<ID3v2::Tag>(TrueAudioID3v2Index, false)->removeUnsupportedProperties(unsupported);
+}
+
+PropertyMap TrueAudio::File::setProperties(const PropertyMap &properties)
+{
+  if(d->hasID3v1)
+    d->tag.access<ID3v1::Tag>(TrueAudioID3v1Index, false)->setProperties(properties);
+  return d->tag.access<ID3v2::Tag>(TrueAudioID3v2Index, true)->setProperties(properties);
 }
 
 TrueAudio::Properties *TrueAudio::File::audioProperties() const
@@ -233,6 +236,16 @@ void TrueAudio::File::strip(int tags)
   }
 }
 
+bool TrueAudio::File::hasID3v1Tag() const
+{
+  return d->hasID3v1;
+}
+
+bool TrueAudio::File::hasID3v2Tag() const
+{
+  return d->hasID3v2;
+}
+
 
 ////////////////////////////////////////////////////////////////////////////////
 // private members
diff --git a/3rdparty/taglib/trueaudio/trueaudiofile.h b/3rdparty/taglib/trueaudio/trueaudiofile.h
index e3e1fe628..dbaafcdb9 100644
--- a/3rdparty/taglib/trueaudio/trueaudiofile.h
+++ b/3rdparty/taglib/trueaudio/trueaudiofile.h
@@ -79,42 +79,50 @@ namespace TagLib {
       };
 
       /*!
-       * Contructs an TrueAudio file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs a TrueAudio file from \a file.  If \a readProperties is true 
+       * the file's audio properties will also be read.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(FileName file, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs an TrueAudio file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored. The frames will be created using
+       * Constructs a TrueAudio file from \a file.  If \a readProperties is true 
+       * the file's audio properties will also be read.
+       *
+       * If this file contains and ID3v2 tag the frames will be created using
        * \a frameFactory.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(FileName file, ID3v2::FrameFactory *frameFactory,
            bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs an TrueAudio file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * Constructs a TrueAudio file from \a stream.  If \a readProperties is true
+       * the file's audio properties will also be read.
        *
        * \note TagLib will *not* take ownership of the stream, the caller is
        * responsible for deleting it after the File object.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(IOStream *stream, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs an TrueAudio file from \a file.  If \a readProperties is true the
-       * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored. The frames will be created using
-       * \a frameFactory.
+       * Constructs a TrueAudio file from \a stream.  If \a readProperties is true 
+       * the file's audio properties will also be read.
        *
        * \note TagLib will *not* take ownership of the stream, the caller is
        * responsible for deleting it after the File object.
+       *
+       * If this file contains and ID3v2 tag the frames will be created using
+       * \a frameFactory.
+       *
+       * \note In the current implementation, \a propertiesStyle is ignored.
        */
       File(IOStream *stream, ID3v2::FrameFactory *frameFactory,
            bool readProperties = true,
@@ -139,11 +147,13 @@ namespace TagLib {
 
       /*!
        * Implements the unified property interface -- import function.
-       * As with the export, only one tag is taken into account. If the file
-       * has no tag at all, ID3v2 will be created.
+       * Creates in ID3v2 tag if necessary. If an ID3v1 tag exists, it will
+       * be updated as well, within the limitations of ID3v1.
        */
       PropertyMap setProperties(const PropertyMap &);
 
+      void removeUnsupportedProperties(const StringList &properties);
+
       /*!
        * Returns the TrueAudio::Properties for this file.  If no audio properties
        * were read then this will return a null pointer.
@@ -163,30 +173,40 @@ namespace TagLib {
       virtual bool save();
 
       /*!
-       * Returns a pointer to the ID3v2 tag of the file.
+       * Returns a pointer to the ID3v1 tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
-       * if there is no valid ID3v2 tag.  If \a create is true it will create
-       * an ID3v1 tag if one does not exist. If there is already an APE tag, the
-       * new ID3v1 tag will be placed after it.
+       * If \a create is false (the default) this may return a null pointer
+       * if there is no valid ID3v1 tag.  If \a create is true it will create
+       * an ID3v1 tag if one does not exist and returns a valid pointer.
        *
-       * \note The Tag <b>is still</b> owned by the TrueAudio::File and should not be
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an ID3v1 tag.  Use hasID3v1Tag() to check if the file 
+       * on disk actually has an ID3v1 tag.
+       *
+       * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasID3v1Tag()
        */
       ID3v1::Tag *ID3v1Tag(bool create = false);
 
       /*!
-       * Returns a pointer to the ID3v1 tag of the file.
+       * Returns a pointer to the ID3v2 tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
-       * if there is no valid ID3v1 tag.  If \a create is true it will create
-       * an ID3v1 tag if one does not exist. If there is already an APE tag, the
-       * new ID3v1 tag will be placed after it.
+       * If \a create is false (the default) this may return a null pointer
+       * if there is no valid ID3v2 tag.  If \a create is true it will create
+       * an ID3v2 tag if one does not exist and returns a valid pointer.
        *
-       * \note The Tag <b>is still</b> owned by the TrueAudio::File and should not be
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an ID3v2 tag.  Use hasID3v2Tag() to check if the file 
+       * on disk actually has an ID3v2 tag.
+       *
+       * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasID3v2Tag()
        */
       ID3v2::Tag *ID3v2Tag(bool create = false);
 
@@ -199,7 +219,21 @@ namespace TagLib {
        * \note In order to make the removal permanent save() still needs to be called
        */
       void strip(int tags = AllTags);
+      
+      /*!
+       * Returns whether or not the file on disk actually has an ID3v1 tag.
+       *
+       * \see ID3v1Tag()
+       */
+      bool hasID3v1Tag() const;
 
+      /*!
+       * Returns whether or not the file on disk actually has an ID3v2 tag.
+       *
+       * \see ID3v2Tag()
+       */
+      bool hasID3v2Tag() const;
+    
     private:
       File(const File &);
       File &operator=(const File &);
diff --git a/3rdparty/taglib/trueaudio/trueaudioproperties.cpp b/3rdparty/taglib/trueaudio/trueaudioproperties.cpp
index 9b251ff18..dedd74e94 100644
--- a/3rdparty/taglib/trueaudio/trueaudioproperties.cpp
+++ b/3rdparty/taglib/trueaudio/trueaudioproperties.cpp
@@ -133,16 +133,16 @@ void TrueAudio::Properties::read()
     // Skip the audio format
     pos += 2;
 
-    d->channels = d->data.mid(pos, 2).toShort(false);
+    d->channels = d->data.toShort(pos, false);
     pos += 2;
 
-    d->bitsPerSample = d->data.mid(pos, 2).toShort(false);
+    d->bitsPerSample = d->data.toShort(pos, false);
     pos += 2;
 
-    d->sampleRate = d->data.mid(pos, 4).toUInt(false);
+    d->sampleRate = d->data.toUInt(pos, false);
     pos += 4;
 
-    d->sampleFrames = d->data.mid(pos, 4).toUInt(false);
+    d->sampleFrames = d->data.toUInt(pos, false);
     d->length = d->sampleRate > 0 ? d->sampleFrames / d->sampleRate : 0;
 
     d->bitrate = d->length > 0 ? ((d->streamLength * 8L) / d->length) / 1000 : 0;
diff --git a/3rdparty/taglib/wavpack/wavpackfile.cpp b/3rdparty/taglib/wavpack/wavpackfile.cpp
index 49f7923e7..13b8c570d 100644
--- a/3rdparty/taglib/wavpack/wavpackfile.cpp
+++ b/3rdparty/taglib/wavpack/wavpackfile.cpp
@@ -86,14 +86,16 @@ WavPack::File::File(FileName file, bool readProperties,
                 Properties::ReadStyle propertiesStyle) : TagLib::File(file)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 WavPack::File::File(IOStream *stream, bool readProperties,
                 Properties::ReadStyle propertiesStyle) : TagLib::File(stream)
 {
   d = new FilePrivate;
-  read(readProperties, propertiesStyle);
+  if(isOpen())
+    read(readProperties, propertiesStyle);
 }
 
 WavPack::File::~File()
@@ -115,14 +117,19 @@ PropertyMap WavPack::File::properties() const
   return PropertyMap();
 }
 
-PropertyMap WavPack::File::setProperties(const PropertyMap &properties)
+
+void WavPack::File::removeUnsupportedProperties(const StringList &unsupported)
 {
   if(d->hasAPE)
-    return d->tag.access<APE::Tag>(WavAPEIndex, false)->setProperties(properties);
-  else if(d->hasID3v1)
-    return d->tag.access<ID3v1::Tag>(WavID3v1Index, false)->setProperties(properties);
-  else
-    return d->tag.access<APE::Tag>(APE, true)->setProperties(properties);
+    d->tag.access<APE::Tag>(WavAPEIndex, false)->removeUnsupportedProperties(unsupported);
+}
+
+
+PropertyMap WavPack::File::setProperties(const PropertyMap &properties)
+{
+  if(d->hasID3v1)
+    d->tag.access<ID3v1::Tag>(WavID3v1Index, false)->setProperties(properties);
+  return d->tag.access<APE::Tag>(WavAPEIndex, true)->setProperties(properties);
 }
 
 WavPack::Properties *WavPack::File::audioProperties() const
@@ -224,6 +231,16 @@ void WavPack::File::strip(int tags)
   }
 }
 
+bool WavPack::File::hasID3v1Tag() const
+{
+  return d->hasID3v1;
+}
+
+bool WavPack::File::hasAPETag() const
+{
+  return d->hasAPE;
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // private members
 ////////////////////////////////////////////////////////////////////////////////
diff --git a/3rdparty/taglib/wavpack/wavpackfile.h b/3rdparty/taglib/wavpack/wavpackfile.h
index 5bbbc65a1..c85c75f2b 100644
--- a/3rdparty/taglib/wavpack/wavpackfile.h
+++ b/3rdparty/taglib/wavpack/wavpackfile.h
@@ -80,15 +80,15 @@ namespace TagLib {
       };
 
       /*!
-       * Contructs an WavPack file from \a file.  If \a readProperties is true the
+       * Constructs a WavPack file from \a file.  If \a readProperties is true the
        * file's audio properties will also be read using \a propertiesStyle.  If
-       * false, \a propertiesStyle is ignored.
+       * false, \a propertiesStyle is ignored
        */
       File(FileName file, bool readProperties = true,
            Properties::ReadStyle propertiesStyle = Properties::Average);
 
       /*!
-       * Contructs an WavPack file from \a file.  If \a readProperties is true the
+       * Constructs an WavPack file from \a file.  If \a readProperties is true the
        * file's audio properties will also be read using \a propertiesStyle.  If
        * false, \a propertiesStyle is ignored.
        *
@@ -116,10 +116,12 @@ namespace TagLib {
        */
       PropertyMap properties() const;
 
+      void removeUnsupportedProperties(const StringList &properties);
+
       /*!
        * Implements the unified property interface -- import function.
-       * As for the export, only one tag is taken into account. If the file
-       * has no tag at all, APE will be created.
+       * Creates an APE tag if it does not exists and calls setProperties() on
+       * that. Any existing ID3v1 tag will be updated as well.
        */
       PropertyMap setProperties(const PropertyMap&);
 
@@ -137,27 +139,38 @@ namespace TagLib {
       /*!
        * Returns a pointer to the ID3v1 tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
+       * If \a create is false (the default) this may return a null pointer
        * if there is no valid ID3v1 tag.  If \a create is true it will create
-       * an ID3v1 tag if one does not exist. If there is already an APE tag, the
-       * new ID3v1 tag will be placed after it.
+       * an ID3v1 tag if one does not exist and returns a valid pointer.
        *
-       * \note The Tag <b>is still</b> owned by the APE::File and should not be
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an ID3v1 tag.  Use hasID3v1Tag() to check if the file 
+       * on disk actually has an ID3v1 tag.
+       *
+       * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasID3v1Tag()
        */
       ID3v1::Tag *ID3v1Tag(bool create = false);
 
       /*!
        * Returns a pointer to the APE tag of the file.
        *
-       * If \a create is false (the default) this will return a null pointer
+       * If \a create is false (the default) this may return a null pointer
        * if there is no valid APE tag.  If \a create is true it will create
-       * a APE tag if one does not exist.
+       * an APE tag if one does not exist and returns a valid pointer.
        *
-       * \note The Tag <b>is still</b> owned by the APE::File and should not be
+       * \note This may return a valid pointer regardless of whether or not the 
+       * file on disk has an APE tag.  Use hasAPETag() to check if the file 
+       * on disk actually has an APE tag.
+       *
+       * \note The Tag <b>is still</b> owned by the MPEG::File and should not be
        * deleted by the user.  It will be deleted when the file (object) is
        * destroyed.
+       *
+       * \see hasAPETag()
        */
       APE::Tag *APETag(bool create = false);
 
@@ -170,7 +183,21 @@ namespace TagLib {
        * \note In order to make the removal permanent save() still needs to be called
        */
       void strip(int tags = AllTags);
+      
+      /*!
+       * Returns whether or not the file on disk actually has an ID3v1 tag.
+       *
+       * \see ID3v1Tag()
+       */
+      bool hasID3v1Tag() const;
 
+      /*!
+       * Returns whether or not the file on disk actually has an APE tag.
+       *
+       * \see APETag()
+       */
+      bool hasAPETag() const;
+    
     private:
       File(const File &);
       File &operator=(const File &);
diff --git a/3rdparty/taglib/wavpack/wavpackproperties.cpp b/3rdparty/taglib/wavpack/wavpackproperties.cpp
index 1715f425d..3f7915443 100644
--- a/3rdparty/taglib/wavpack/wavpackproperties.cpp
+++ b/3rdparty/taglib/wavpack/wavpackproperties.cpp
@@ -148,17 +148,17 @@ void WavPack::Properties::read()
   if(!d->data.startsWith("wvpk"))
     return;
 
-  d->version = d->data.mid(8, 2).toShort(false);
+  d->version = d->data.toShort(8, false);
   if(d->version < MIN_STREAM_VERS || d->version > MAX_STREAM_VERS)
     return;
 
-  unsigned int flags = d->data.mid(24, 4).toUInt(false);
+  const unsigned int flags = d->data.toUInt(24, false);
   d->bitsPerSample = ((flags & BYTES_STORED) + 1) * 8 -
     ((flags & SHIFT_MASK) >> SHIFT_LSB);
   d->sampleRate = sample_rates[(flags & SRATE_MASK) >> SRATE_LSB];
   d->channels = (flags & MONO_FLAG) ? 1 : 2;
 
-  unsigned int samples = d->data.mid(12, 4).toUInt(false);
+  unsigned int samples = d->data.toUInt(12, false);
   if(samples == ~0u) {
     if(d->file && d->style != Fast) {
       samples = seekFinalIndex();
@@ -186,14 +186,14 @@ unsigned int WavPack::Properties::seekFinalIndex()
     ByteVector data = d->file->readBlock(32);
     if(data.size() != 32)
       return 0;
-    int version = data.mid(8, 2).toShort(false);
+    const int version = data.toShort(8, false);
     if(version < MIN_STREAM_VERS || version > MAX_STREAM_VERS)
       continue;
-    unsigned int flags = data.mid(24, 4).toUInt(false);
+    const unsigned int flags = data.toUInt(24, false);
     if(!(flags & FINAL_BLOCK))
       return 0;
-    unsigned int blockIndex = data.mid(16, 4).toUInt(false);
-    unsigned int blockSamples = data.mid(20, 4).toUInt(false);
+    const unsigned int blockIndex   = data.toUInt(16, false);
+    const unsigned int blockSamples = data.toUInt(20, false);
     return blockIndex + blockSamples;
   }
 
diff --git a/3rdparty/taglib/xm/xmfile.cpp b/3rdparty/taglib/xm/xmfile.cpp
index 272e5fe0b..d8373a781 100644
--- a/3rdparty/taglib/xm/xmfile.cpp
+++ b/3rdparty/taglib/xm/xmfile.cpp
@@ -359,7 +359,8 @@ XM::File::File(FileName file, bool readProperties,
   Mod::FileBase(file),
   d(new FilePrivate(propertiesStyle))
 {
-  read(readProperties);
+  if(isOpen())
+    read(readProperties);
 }
 
 XM::File::File(IOStream *stream, bool readProperties,
@@ -367,7 +368,8 @@ XM::File::File(IOStream *stream, bool readProperties,
   Mod::FileBase(stream),
   d(new FilePrivate(propertiesStyle))
 {
-  read(readProperties);
+  if(isOpen())
+    read(readProperties);
 }
 
 XM::File::~File()
@@ -443,7 +445,7 @@ bool XM::File::save()
       return false;
 
     uint len = std::min(22UL, instrumentHeaderSize - 4U);
-    if(i > lines.size())
+    if(i >= lines.size())
       writeString(String::null, len);
     else
       writeString(lines[i], len);
diff --git a/3rdparty/taglib/xm/xmfile.h b/3rdparty/taglib/xm/xmfile.h
index 9d1bb71c6..ae48a511d 100644
--- a/3rdparty/taglib/xm/xmfile.h
+++ b/3rdparty/taglib/xm/xmfile.h
@@ -36,18 +36,22 @@ namespace TagLib {
     class TAGLIB_EXPORT File : public Mod::FileBase {
       public:
         /*!
-         * Contructs a Extended Module file from \a file. If \a readProperties
-         * is true the file's audio properties will also be read using
-         * \a propertiesStyle. If false, \a propertiesStyle is ignored.
+         * Constructs an Extended Module file from \a file.
+         *
+         * \note In the current implementation, both \a readProperties and
+         * \a propertiesStyle are ignored.  The audio properties are always
+         * read.
          */
         File(FileName file, bool readProperties = true,
              AudioProperties::ReadStyle propertiesStyle =
              AudioProperties::Average);
 
         /*!
-         * Contructs a Extended Module file from \a stream. If \a readProperties
-         * is true the file's audio properties will also be read using
-         * \a propertiesStyle. If false, \a propertiesStyle is ignored.
+         * Constructs an Extended Module file from \a stream.
+         *
+         * \note In the current implementation, both \a readProperties and
+         * \a propertiesStyle are ignored.  The audio properties are always
+         * read.
          *
          * \note TagLib will *not* take ownership of the stream, the caller is
          * responsible for deleting it after the File object.
diff --git a/CMakeLists.txt b/CMakeLists.txt
index d1d738ab4..7bf220a9f 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -105,16 +105,17 @@ if (USE_BUILTIN_TAGLIB AND
   set(TAGLIB_INCLUDE_DIRS "${CMAKE_BINARY_DIR}/3rdparty/taglib/headers/taglib/;${CMAKE_BINARY_DIR}/3rdparty/taglib/headers/")
   set(TAGLIB_LIBRARY_DIRS "")
   set(TAGLIB_LIBRARIES tag)
+  set(TAGLIB_HAS_OPUS ON)
   add_subdirectory(3rdparty/taglib)
+else()
+  set(CMAKE_REQUIRED_INCLUDES "${TAGLIB_INCLUDE_DIRS}")
+  set(CMAKE_REQUIRED_LIBRARIES "${TAGLIB_LIBRARIES}")
+  check_cxx_source_compiles("#include <opusfile.h>
+      int main() { char *s; TagLib::Ogg::Opus::File opusfile(s); return 0;}" TAGLIB_HAS_OPUS)
+  set(CMAKE_REQUIRED_INCLUDES)
+  set(CMAKE_REQUIRED_LIBRARIES)
 endif()
 
-set(CMAKE_REQUIRED_INCLUDES "${TAGLIB_INCLUDE_DIRS}")
-set(CMAKE_REQUIRED_LIBRARIES "${TAGLIB_LIBRARIES}")
-check_cxx_source_compiles("#include <opusfile.h>
-    int main() { char *s; TagLib::Ogg::Opus::File opusfile(s); return 0;}" TAGLIB_HAS_OPUS)
-set(CMAKE_REQUIRED_INCLUDES)
-set(CMAKE_REQUIRED_LIBRARIES)
-
 if(LASTFM_INCLUDE_DIRS AND LASTFM1_INCLUDE_DIRS)
   set(HAVE_LIBLASTFM1 ON)
 endif()