From 367ed4e13d697ceb415183d8a7acddf5f707667c Mon Sep 17 00:00:00 2001 From: "Andrey A. Chernov" Date: Sat, 13 Oct 2007 16:28:22 +0000 Subject: The problem is: currently our single byte ctype(3) functions are broken for wide characters locales in the argument range >= 0x80 - they may return false positives. Example 1: for UTF-8 locale we currently have: iswspace(0xA0)==1 and isspace(0xA0)==1 (because iswspace() and isspace() are the same code) but must have iswspace(0xA0)==1 and isspace(0xA0)==0 (because there is no such character and all others in the range 0x80..0xff for the UTF-8 locale, it keeps ASCII only in the single byte range because our internal wchar_t representation for UTF-8 is UCS-4). Example 2: for all wide character locales isalpha(arg) when arg > 0xFF may return false positives (must be 0). (because iswalpha() and isalpha() are the same code) This change address this issue separating single byte and wide ctype and also fix iswascii() (currently iswascii() is broken for arguments > 0xFF). This change is 100% binary compatible with old binaries. Reviewied by: i18n@ --- include/_ctype.h | 35 ++++++++++++++++++++++++++++++++++- include/ctype.h | 42 +++++++++++++++++++++--------------------- include/wctype.h | 2 +- 3 files changed, 56 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/include/_ctype.h b/include/_ctype.h index 1a42952a09dc..e27ab24f6c19 100644 --- a/include/_ctype.h +++ b/include/_ctype.h @@ -87,6 +87,8 @@ __END_DECLS #define __inline #endif +extern int __mb_sb_limit; + /* * Use inline functions if we are allowed to and the compiler supports them. */ @@ -102,16 +104,29 @@ __maskrune(__ct_rune_t _c, unsigned long _f) _CurrentRuneLocale->__runetype[_c]) & _f; } +static __inline int +__sbmaskrune(__ct_rune_t _c, unsigned long _f) +{ + return (_c < 0 || _c >= __mb_sb_limit) ? 0 : + _CurrentRuneLocale->__runetype[_c] & _f; +} + static __inline int __istype(__ct_rune_t _c, unsigned long _f) { return (!!__maskrune(_c, _f)); } +static __inline int +__sbistype(__ct_rune_t _c, unsigned long _f) +{ + return (!!__sbmaskrune(_c, _f)); +} + static __inline int __isctype(__ct_rune_t _c, unsigned long _f) { - return (_c < 0 || _c >= _CACHED_RUNES) ? 0 : + return (_c < 0 || _c >= __mb_sb_limit) ? 0 : !!(_DefaultRuneLocale.__runetype[_c] & _f); } @@ -122,6 +137,13 @@ __toupper(__ct_rune_t _c) _CurrentRuneLocale->__mapupper[_c]; } +static __inline __ct_rune_t +__sbtoupper(__ct_rune_t _c) +{ + return (_c < 0 || _c >= __mb_sb_limit) ? _c : + _CurrentRuneLocale->__mapupper[_c]; +} + static __inline __ct_rune_t __tolower(__ct_rune_t _c) { @@ -129,6 +151,13 @@ __tolower(__ct_rune_t _c) _CurrentRuneLocale->__maplower[_c]; } +static __inline __ct_rune_t +__sbtolower(__ct_rune_t _c) +{ + return (_c < 0 || _c >= __mb_sb_limit) ? _c : + _CurrentRuneLocale->__maplower[_c]; +} + static __inline int __wcwidth(__ct_rune_t _c) { @@ -146,10 +175,14 @@ __wcwidth(__ct_rune_t _c) __BEGIN_DECLS int __maskrune(__ct_rune_t, unsigned long); +int __sbmaskrune(__ct_rune_t, unsigned long); int __istype(__ct_rune_t, unsigned long); +int __sbistype(__ct_rune_t, unsigned long); int __isctype(__ct_rune_t, unsigned long); __ct_rune_t __toupper(__ct_rune_t); +__ct_rune_t __sbtoupper(__ct_rune_t); __ct_rune_t __tolower(__ct_rune_t); +__ct_rune_t __sbtolower(__ct_rune_t); int __wcwidth(__ct_rune_t); __END_DECLS #endif /* using inlines */ diff --git a/include/ctype.h b/include/ctype.h index 0825ff652c28..dfc89ccd5951 100644 --- a/include/ctype.h +++ b/include/ctype.h @@ -86,19 +86,19 @@ int isspecial(int); #endif __END_DECLS -#define isalnum(c) __istype((c), _CTYPE_A|_CTYPE_D) -#define isalpha(c) __istype((c), _CTYPE_A) -#define iscntrl(c) __istype((c), _CTYPE_C) +#define isalnum(c) __sbistype((c), _CTYPE_A|_CTYPE_D) +#define isalpha(c) __sbistype((c), _CTYPE_A) +#define iscntrl(c) __sbistype((c), _CTYPE_C) #define isdigit(c) __isctype((c), _CTYPE_D) /* ANSI -- locale independent */ -#define isgraph(c) __istype((c), _CTYPE_G) -#define islower(c) __istype((c), _CTYPE_L) -#define isprint(c) __istype((c), _CTYPE_R) -#define ispunct(c) __istype((c), _CTYPE_P) -#define isspace(c) __istype((c), _CTYPE_S) -#define isupper(c) __istype((c), _CTYPE_U) +#define isgraph(c) __sbistype((c), _CTYPE_G) +#define islower(c) __sbistype((c), _CTYPE_L) +#define isprint(c) __sbistype((c), _CTYPE_R) +#define ispunct(c) __sbistype((c), _CTYPE_P) +#define isspace(c) __sbistype((c), _CTYPE_S) +#define isupper(c) __sbistype((c), _CTYPE_U) #define isxdigit(c) __isctype((c), _CTYPE_X) /* ANSI -- locale independent */ -#define tolower(c) __tolower(c) -#define toupper(c) __toupper(c) +#define tolower(c) __sbtolower(c) +#define toupper(c) __sbtoupper(c) #if __XSI_VISIBLE /* @@ -112,24 +112,24 @@ __END_DECLS * * XXX isascii() and toascii() should similarly be undocumented. */ -#define _tolower(c) __tolower(c) -#define _toupper(c) __toupper(c) +#define _tolower(c) __sbtolower(c) +#define _toupper(c) __sbtoupper(c) #define isascii(c) (((c) & ~0x7F) == 0) #define toascii(c) ((c) & 0x7F) #endif #if __ISO_C_VISIBLE >= 1999 -#define isblank(c) __istype((c), _CTYPE_B) +#define isblank(c) __sbistype((c), _CTYPE_B) #endif #if __BSD_VISIBLE -#define digittoint(c) __maskrune((c), 0xFF) -#define ishexnumber(c) __istype((c), _CTYPE_X) -#define isideogram(c) __istype((c), _CTYPE_I) -#define isnumber(c) __istype((c), _CTYPE_D) -#define isphonogram(c) __istype((c), _CTYPE_Q) -#define isrune(c) __istype((c), 0xFFFFFF00L) -#define isspecial(c) __istype((c), _CTYPE_T) +#define digittoint(c) __sbmaskrune((c), 0xFF) +#define ishexnumber(c) __sbistype((c), _CTYPE_X) +#define isideogram(c) __sbistype((c), _CTYPE_I) +#define isnumber(c) __sbistype((c), _CTYPE_D) +#define isphonogram(c) __sbistype((c), _CTYPE_Q) +#define isrune(c) __sbistype((c), 0xFFFFFF00L) +#define isspecial(c) __sbistype((c), _CTYPE_T) #endif #endif /* !_CTYPE_H_ */ diff --git a/include/wctype.h b/include/wctype.h index 098045fa11a5..31f401ff76e6 100644 --- a/include/wctype.h +++ b/include/wctype.h @@ -106,7 +106,7 @@ __END_DECLS #define towupper(wc) __toupper(wc) #if __BSD_VISIBLE -#define iswascii(wc) (((wc) & ~0x7F) == 0) +#define iswascii(wc) ((wc) < 0x80) #define iswhexnumber(wc) __istype((wc), _CTYPE_X) #define iswideogram(wc) __istype((wc), _CTYPE_I) #define iswnumber(wc) __istype((wc), _CTYPE_D) -- cgit v1.3