diff options
Diffstat (limited to 'generic/tclUtf.c')
| -rw-r--r-- | generic/tclUtf.c | 94 |
1 files changed, 48 insertions, 46 deletions
diff --git a/generic/tclUtf.c b/generic/tclUtf.c index 1c7e1a9..6e80bef 100644 --- a/generic/tclUtf.c +++ b/generic/tclUtf.c @@ -103,7 +103,7 @@ TclUtfCount( return 2; } #if TCL_UTF_MAX > 3 - if (((unsigned)(ch - 0x10000) <= 0xfffff)) { + if (((unsigned)(ch - 0x10000) <= 0xFFFFF)) { return 4; } #endif @@ -208,14 +208,14 @@ three: char * Tcl_UniCharToUtfDString( const Tcl_UniChar *uniStr, /* Unicode string to convert to UTF-8. */ - int uniLength, /* Length of Unicode string in Tcl_UniChars + size_t uniLength, /* Length of Unicode string in Tcl_UniChars * (must be >= 0). */ Tcl_DString *dsPtr) /* UTF-8 representation of string is appended * to this previously initialized DString. */ { const Tcl_UniChar *w, *wEnd; char *p, *string; - int oldLength; + size_t oldLength; /* * UTF-8 string length in bytes will be <= Unicode string length * @@ -292,7 +292,9 @@ Tcl_UtfToUniChar( */ *chPtr = (Tcl_UniChar) (((byte & 0x1F) << 6) | (src[1] & 0x3F)); - return 2; + if ((unsigned)(*chPtr - 1) >= (UNICODE_SELF - 1)) { + return 2; + } } /* @@ -307,7 +309,9 @@ Tcl_UtfToUniChar( *chPtr = (Tcl_UniChar) (((byte & 0x0F) << 12) | ((src[1] & 0x3F) << 6) | (src[2] & 0x3F)); - return 3; + if (*chPtr > 0x7FF) { + return 3; + } } /* @@ -322,13 +326,15 @@ Tcl_UtfToUniChar( * Four-byte-character lead byte followed by three trail bytes. */ - *chPtr = (Tcl_UniChar) (((byte & 0x0E) << 18) | ((src[1] & 0x3F) << 12) + *chPtr = (Tcl_UniChar) (((byte & 0x07) << 18) | ((src[1] & 0x3F) << 12) | ((src[2] & 0x3F) << 6) | (src[3] & 0x3F)); - return 4; + if ((unsigned)(*chPtr - 0x10000) <= 0xFFFFF) { + return 4; + } } /* - * A three-byte-character lead-byte not followed by two trail-bytes + * A four-byte-character lead-byte not followed by two trail-bytes * represents itself. */ } @@ -359,7 +365,7 @@ Tcl_UtfToUniChar( Tcl_UniChar * Tcl_UtfToUniCharDString( const char *src, /* UTF-8 string to convert to Unicode. */ - int length, /* Length of UTF-8 string in bytes, or -1 for + size_t length, /* Length of UTF-8 string in bytes, or -1 for * strlen(). */ Tcl_DString *dsPtr) /* Unicode representation of string is * appended to this previously initialized @@ -367,9 +373,9 @@ Tcl_UtfToUniCharDString( { Tcl_UniChar *w, *wString; const char *p, *end; - int oldLength; + size_t oldLength; - if (length < 0) { + if (length == (size_t)-1) { length = strlen(src); } @@ -420,12 +426,9 @@ int Tcl_UtfCharComplete( const char *src, /* String to check if first few bytes contain * a complete UTF-8 character. */ - int length) /* Length of above string in bytes. */ + size_t length) /* Length of above string in bytes. */ { - int ch; - - ch = *((unsigned char *) src); - return length >= totalBytes[ch]; + return length >= totalBytes[(unsigned char) *src]; } /* @@ -446,15 +449,14 @@ Tcl_UtfCharComplete( *--------------------------------------------------------------------------- */ -int +size_t Tcl_NumUtfChars( register const char *src, /* The UTF-8 string to measure. */ - int length) /* The length of the string in bytes, or -1 + size_t length) /* The length of the string in bytes, or -1 * for strlen(string). */ { Tcl_UniChar ch; - register Tcl_UniChar *chPtr = &ch; - register int i; + register size_t i = 0; /* * The separate implementations are faster. @@ -463,26 +465,26 @@ Tcl_NumUtfChars( * single-byte char case specially. */ - i = 0; - if (length < 0) { + if (length == (size_t)-1) { while (*src != '\0') { - src += TclUtfToUniChar(src, chPtr); + src += TclUtfToUniChar(src, &ch); i++; } } else { - register int n; - - while (length > 0) { - if (UCHAR(*src) < 0xC0) { - length--; - src++; - } else { - n = Tcl_UtfToUniChar(src, chPtr); - length -= n; - src += n; - } + register const char *endPtr = src + length - TCL_UTF_MAX; + + while (src < endPtr) { + src += TclUtfToUniChar(src, &ch); i++; } + endPtr += TCL_UTF_MAX; + while ((src < endPtr) && Tcl_UtfCharComplete(src, endPtr - src)) { + src += TclUtfToUniChar(src, &ch); + i++; + } + if (src < endPtr) { + i += endPtr - src; + } } return i; } @@ -1004,7 +1006,7 @@ Tcl_UtfNcmp( /* * Cannot use 'memcmp(cs, ct, n);' as byte representation of \u0000 (the - * pair of bytes 0xc0,0x80) is larger than byte representation of \u0001 + * pair of bytes 0xC0,0x80) is larger than byte representation of \u0001 * (the byte 0x01.) */ @@ -1388,11 +1390,11 @@ Tcl_UniCharIsControl( { #if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { - ch &= 0x1fffff; - if ((ch == 0xe0001) || ((ch >= 0xe0020) && (ch <= 0xe007f))) { + ch &= 0x1FFFFF; + if ((ch == 0xE0001) || ((ch >= 0xE0020) && (ch <= 0xE007f))) { return 1; } - if ((ch >= 0xf0000) && ((ch & 0xffff) <= 0xfffd)) { + if ((ch >= 0xF0000) && ((ch & 0xFFFF) <= 0xFFFD)) { return 1; } return 0; @@ -1451,8 +1453,8 @@ Tcl_UniCharIsGraph( { #if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { - ch &= 0x1fffff; - return (ch >= 0xe0100) && (ch <= 0xe01ef); + ch &= 0x1FFFFF; + return (ch >= 0xE0100) && (ch <= 0xE01EF); } #endif return ((GRAPH_BITS >> GetCategory(ch)) & 1); @@ -1508,8 +1510,8 @@ Tcl_UniCharIsPrint( { #if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { - ch &= 0x1fffff; - return (ch >= 0xe0100) && (ch <= 0xe01ef); + ch &= 0x1FFFFF; + return (ch >= 0xE0100) && (ch <= 0xE01EF); } #endif return (((GRAPH_BITS|SPACE_BITS) >> GetCategory(ch)) & 1); @@ -1565,10 +1567,10 @@ Tcl_UniCharIsSpace( { #if TCL_UTF_MAX > 3 /* Ignore upper 11 bits. */ - ch &= 0x1fffff; + ch &= 0x1FFFFF; #else /* Ignore upper 16 bits. */ - ch &= 0xffff; + ch &= 0xFFFF; #endif /* @@ -1582,8 +1584,8 @@ Tcl_UniCharIsSpace( } else if (UNICODE_OUT_OF_RANGE(ch)) { return 0; #endif - } else if (ch == 0x0085 || ch == 0x180e || ch == 0x200b - || ch == 0x202f || ch == 0x2060 || ch == 0xfeff) { + } else if (ch == 0x0085 || ch == 0x180E || ch == 0x200B + || ch == 0x202F || ch == 0x2060 || ch == 0xFEFF) { return 1; } else { return ((SPACE_BITS >> GetCategory(ch)) & 1); |