* faq-using.xml: Rework UTF FAQ to accommodate latest setlocale

change in newlib.
This commit is contained in:
Corinna Vinschen 2009-03-03 10:36:59 +00:00
parent d6cd9169dc
commit 92fa4552fd
2 changed files with 19 additions and 5 deletions

View File

@ -1,3 +1,8 @@
2009-03-03 Corinna Vinschen <corinna@vinschen.de>
* faq-using.xml: Rework UTF FAQ to accommodate latest setlocale
change in newlib.
2009-03-03 Corinna Vinschen <corinna@vinschen.de>
* pathnames.sgml: Remove reference to managed mountpoints in mount

View File

@ -368,11 +368,20 @@ formfeed character to your file.
<para>Internationalization is a complex issue. The short answer is that
Cygwin relies on the setting of the CYGWIN environment variable as well
as on the setting of LANG environment variable. The underlying C library,
newlib, only supports a small subset of LANG settings. The default is "C".
To get UTF-8 support you must set LANG to "C-UTF-8" and CYGWIN so that
it contains "codepage:utf8".
</para>
as on the setting of LANG/LC_xxx environment variables.</para>
<para>To get UTF-8 support you must set the environment variable CYGWIN
so that it contains the substring "codepage:utf8". This is required in
Cygwin so far to get correct translation from Windows wide character
filenames to their UTF-8 counterpart. Applications on the other hand
require the setting of the LANG, LC_ALL, or LC_CTYPE environment variables.
To get UTF-8 support you can set, for instance, $LANG to "en_US.UTF-8".
This will give you support for the UTF-8 character set. Note that the
language part has to contain a valid language specifier, but is otherwise
so far ignored by newlib, the underlying C library. There's no support
for correct language-specific collation, monetary or date/time-related
string handling. This is planned for a later release, though.</para>
<para>To type international characters (&pound;&auml;&ouml;) in
<literal>bash</literal>, add the following lines to your
<literal>~/.inputrc</literal> file and restart <literal>bash</literal>: