JehanneOS/newlib
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

* strfuncs.cc (sys_cp_wcstombs): Implement reverse functionality

Browse Source 
of the change to sys_cp_mbstowcs from 2009-05-30.
	(sys_cp_mbstowcs): Slightly reformat.  Fix comment to accommodate
	change to sys_cp_wcstombs.  Don't write to *ptr if dst is NULL.
This commit is contained in:
Corinna Vinschen
2009-06-03 17:23:39 +00:00
parent c48a1e46bf
commit 397775c6f6
2 changed files with 29 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
winsup/cygwin/ChangeLog
View File

@@ -1,3 +1,10 @@
2009-06-03 Corinna Vinschen <corinna@vinschen.de>
* strfuncs.cc (sys_cp_wcstombs): Implement reverse functionality
of the change to sys_cp_mbstowcs from 2009-05-30.
(sys_cp_mbstowcs): Slightly reformat. Fix comment to accommodate
change to sys_cp_wcstombs. Don't write to *ptr if dst is NULL.
2009-06-03 Corinna Vinschen <corinna@vinschen.de> 2009-06-03 Corinna Vinschen <corinna@vinschen.de>
* fhandler_console.cc (fhandler_console::read): Convert Alt-Backspace * fhandler_console.cc (fhandler_console::read): Convert Alt-Backspace

30
winsup/cygwin/strfuncs.cc
View File

@@ -427,10 +427,19 @@ sys_cp_wcstombs (wctomb_p f_wctomb, char *charset, char *dst, size_t len,
if ((pw & 0xff00) == 0xf000) if ((pw & 0xff00) == 0xf000)
pw &= 0xff; pw &= 0xff;
int bytes = f_wctomb (_REENT, buf, pw, charset, &ps); int bytes = f_wctomb (_REENT, buf, pw, charset, &ps);
if (bytes == -1 && (pw & 0xff00) == 0xdc00)
{
/* Reverse functionality of the single invalid second half of a
surrogate pair in the 0xDCxx range specifying an invalid byte
value when converting from MB to WC.
The comment in sys_cp_mbstowcs below explains it. */
buf[0] = (char) (pw & 0xff);
bytes = 1;
}
else if (bytes == -1 && *charset != 'U'/*TF-8*/)
{
/* Convert chars invalid in the current codepage to a sequence /* Convert chars invalid in the current codepage to a sequence
ASCII SO; UTF-8 representation of invalid char. */ ASCII SO; UTF-8 representation of invalid char. */
if (bytes == -1 && *charset != 'U'/*TF-8*/)
{
buf[0] = 0x0e; /* ASCII SO */ buf[0] = 0x0e; /* ASCII SO */
bytes = __utf8_wctomb (_REENT, buf + 1, pw, charset, &ps); bytes = __utf8_wctomb (_REENT, buf + 1, pw, charset, &ps);
if (bytes == -1) if (bytes == -1)
@@ -561,7 +570,8 @@ sys_cp_mbstowcs (mbtowc_p f_mbtowc, char *charset, wchar_t *dst, size_t dlen,
if (ps.__count == 4) /* First half of a surrogate. */ if (ps.__count == 4) /* First half of a surrogate. */
{ {
wchar_t *ptr2 = dst ? ptr + 1 : NULL; wchar_t *ptr2 = dst ? ptr + 1 : NULL;
int bytes2 = __utf8_mbtowc (_REENT, ptr2, (const char *) pmbs + bytes, int bytes2 = __utf8_mbtowc (_REENT, ptr2,
(const char *) pmbs + bytes,
nms - bytes, charset, &ps); nms - bytes, charset, &ps);
if (bytes2 < 0) if (bytes2 < 0)
break; break;
@@ -572,7 +582,9 @@ sys_cp_mbstowcs (mbtowc_p f_mbtowc, char *charset, wchar_t *dst, size_t dlen,
--len; --len;
} }
} }
else if ((bytes = f_mbtowc (_REENT, ptr, (const char *) pmbs, nms, charset, &ps)) < 0 && *pmbs > '\x80') else if ((bytes = f_mbtowc (_REENT, ptr, (const char *) pmbs, nms,
charset, &ps)) < 0
&& *pmbs > '\x80')
{ {
/* This should probably be handled in f_mbtowc which can operate /* This should probably be handled in f_mbtowc which can operate
on sequences rather than individual characters. on sequences rather than individual characters.
@@ -581,12 +593,10 @@ sys_cp_mbstowcs (mbtowc_p f_mbtowc, char *charset, wchar_t *dst, size_t dlen,
http://www.mail-archive.com/linux-utf8@nl.linux.org/msg00080.html http://www.mail-archive.com/linux-utf8@nl.linux.org/msg00080.html
This is hardly perfect. Windows doesn't do anything sensical with This is hardly perfect. Windows doesn't do anything sensical with
characters converted to this format and (currently) we don't convert characters converted to this format. It does allow processing of
them back into their original single byte form. It does allow src to continue, however, which, since there is no way to signal
processing of src to continue, however, which, since there is no decoding errors, seems like the best we can do. */
way to signal decoding errors, seems like the best we can do. if (dst)
*/
*ptr = L'\xdc80' | *pmbs; *ptr = L'\xdc80' | *pmbs;
bytes = 1; bytes = 1;
} }