diff options
Diffstat (limited to 'generic/tclUtf.c')
-rw-r--r-- | generic/tclUtf.c | 543 |
1 files changed, 417 insertions, 126 deletions
diff --git a/generic/tclUtf.c b/generic/tclUtf.c index 4b70f96..1387ee7 100644 --- a/generic/tclUtf.c +++ b/generic/tclUtf.c @@ -67,25 +67,13 @@ static const unsigned char totalBytes[256] = { 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, - 3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3, -#if TCL_UTF_MAX > 3 - 4,4,4,4,4, -#else - 1,1,1,1,1, -#endif - 1,1,1,1,1,1,1,1,1,1,1 + 3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,4,4,4,4,4,1,1,1,1,1,1,1,1,1,1,1 }; - -/* - * Functions used only in this module. - */ - -static int UtfCount(int ch); /* *--------------------------------------------------------------------------- * - * UtfCount -- + * TclUtfCount -- * * Find the number of bytes in the Utf character "ch". * @@ -98,8 +86,8 @@ static int UtfCount(int ch); *--------------------------------------------------------------------------- */ -static inline int -UtfCount( +int +TclUtfCount( int ch) /* The Unicode character whose size is returned. */ { if ((unsigned)(ch - 1) < (UNICODE_SELF - 1)) { @@ -108,11 +96,9 @@ UtfCount( if (ch <= 0x7FF) { return 2; } -#if TCL_UTF_MAX > 3 if (((unsigned)(ch - 0x10000) <= 0xFFFFF)) { return 4; } -#endif return 3; } @@ -124,6 +110,19 @@ UtfCount( * Store the given Tcl_UniChar as a sequence of UTF-8 bytes in the * provided buffer. Equivalent to Plan 9 runetochar(). * + * Special handling of Surrogate pairs is handled as follows: + * When this function is called for ch being a high surrogate, + * the first byte of the 4-byte UTF-8 sequence is produced and + * the function returns 1. Calling the function again with a + * low surrogate, the remaining 3 bytes of the 4-byte UTF-8 + * sequence is produced, and the function returns 3. The buffer + * is used to remember the high surrogate between the two calls. + * + * If no low surrogate follows the high surrogate (which is actually + * illegal), this can be handled reasonably by calling Tcl_UniCharToUtf + * again with ch = -1. This will produce a 3-byte UTF-8 sequence + * representing the high surrogate. + * * Results: * The return values is the number of bytes in the buffer that were * consumed. @@ -141,7 +140,7 @@ Tcl_UniCharToUtf( char *buf) /* Buffer in which the UTF-8 representation of * the Tcl_UniChar is stored. Buffer must be * large enough to hold the UTF-8 character - * (at most TCL_UTF_MAX bytes). */ + * (at most 4 bytes). */ { if ((unsigned)(ch - 1) < (UNICODE_SELF - 1)) { buf[0] = (char) ch; @@ -154,7 +153,6 @@ Tcl_UniCharToUtf( return 2; } if (ch <= 0xFFFF) { -#if TCL_UTF_MAX > 3 if ((ch & 0xF800) == 0xD800) { if (ch & 0x0400) { /* Low surrogate */ @@ -176,11 +174,8 @@ Tcl_UniCharToUtf( return 1; } } -#endif goto three; } - -#if TCL_UTF_MAX > 3 if (ch <= 0x10FFFF) { buf[3] = (char) ((ch | 0x80) & 0xBF); buf[2] = (char) (((ch >> 6) | 0x80) & 0xBF); @@ -198,7 +193,6 @@ Tcl_UniCharToUtf( buf[-1] = (char) ((ch >> 12) | 0xE0); return 2; } -#endif } ch = 0xFFFD; @@ -237,28 +231,83 @@ Tcl_UniCharToUtfDString( { const Tcl_UniChar *w, *wEnd; char *p, *string; - int oldLength; + int oldLength, len = 1; /* - * UTF-8 string length in bytes will be <= Unicode string length * - * TCL_UTF_MAX. + * UTF-8 string length in bytes will be <= Unicode string length * 4. */ oldLength = Tcl_DStringLength(dsPtr); - Tcl_DStringSetLength(dsPtr, oldLength + (uniLength + 1) * TCL_UTF_MAX); + Tcl_DStringSetLength(dsPtr, oldLength + (uniLength + 1) * 4); string = Tcl_DStringValue(dsPtr) + oldLength; p = string; wEnd = uniStr + uniLength; for (w = uniStr; w < wEnd; ) { - p += Tcl_UniCharToUtf(*w, p); + if (!len && ((*w & 0xFC00) != 0xDC00)) { + /* Special case for handling high surrogates. */ + p += Tcl_UniCharToUtf(-1, p); + } + len = Tcl_UniCharToUtf(*w, p); + p += len; + if ((*w >= 0xD800) && (len < 3)) { + len = 0; /* Indication that high surrogate was found */ + } w++; } + if (!len) { + /* Special case for handling high surrogates. */ + p += Tcl_UniCharToUtf(-1, p); + } Tcl_DStringSetLength(dsPtr, oldLength + (p - string)); return string; } +#if (TCL_UTF_MAX > 4) && (defined(__CYGWIN__) || defined(_WIN32)) +char * +TclWCharToUtfDString( + const WCHAR *uniStr, /* WCHAR string to convert to UTF-8. */ + int uniLength, /* Length of WCHAR string in Tcl_UniChars + * (must be >= 0). */ + Tcl_DString *dsPtr) /* UTF-8 representation of string is appended + * to this previously initialized DString. */ +{ + const WCHAR *w, *wEnd; + char *p, *string; + int oldLength, len = 1; + + /* + * UTF-8 string length in bytes will be <= Unicode string length * 4. + */ + + oldLength = Tcl_DStringLength(dsPtr); + Tcl_DStringSetLength(dsPtr, oldLength + (uniLength + 1) * 4); + string = Tcl_DStringValue(dsPtr) + oldLength; + + p = string; + wEnd = uniStr + uniLength; + for (w = uniStr; w < wEnd; ) { + if (!len && ((*w & 0xFC00) != 0xDC00)) { + /* Special case for handling high surrogates. */ + p += Tcl_UniCharToUtf(-1, p); + } + len = Tcl_UniCharToUtf(*w, p); + p += len; + if ((*w >= 0xD800) && (len < 3)) { + len = 0; /* Indication that high surrogate was found */ + } + w++; + } + if (!len) { + /* Special case for handling high surrogates. */ + p += Tcl_UniCharToUtf(-1, p); + } + Tcl_DStringSetLength(dsPtr, oldLength + (p - string)); + + return string; +} +#endif /* *--------------------------------------------------------------------------- * @@ -275,11 +324,11 @@ Tcl_UniCharToUtfDString( * Tcl_UtfCharComplete() before calling this routine to ensure that * enough bytes remain in the string. * - * If TCL_UTF_MAX == 4, special handling of Surrogate pairs is done: + * Special handling of Surrogate pairs is handled as follows: * For any UTF-8 string containing a character outside of the BMP, the * first call to this function will fill *chPtr with the high surrogate - * and generate a return value of 0. Calling Tcl_UtfToUniChar again - * will produce the low surrogate and a return value of 4. Because *chPtr + * and generate a return value of 1. Calling Tcl_UtfToUniChar again + * will produce the low surrogate and a return value of 3. Because *chPtr * is used to remember whether the high surrogate is already produced, it * is recommended to initialize the variable it points to as 0 before * the first call to Tcl_UtfToUniChar is done. @@ -294,6 +343,13 @@ Tcl_UniCharToUtfDString( *--------------------------------------------------------------------------- */ +static const unsigned short cp1252[32] = { + 0x20ac, 0x81, 0x201A, 0x0192, 0x201E, 0x2026, 0x2020, 0x2021, + 0x02C6, 0x2030, 0x0160, 0x2039, 0x0152, 0x8D, 0x017D, 0x8F, + 0x90, 0x2018, 0x2019, 0x201C, 0x201D, 0x2022, 0x2013, 0x2014, + 0x2DC, 0x2122, 0x0161, 0x203A, 0x0153, 0x9D, 0x017E, 0x0178 +}; + int Tcl_UtfToUniChar( register const char *src, /* The UTF-8 string. */ @@ -303,18 +359,20 @@ Tcl_UtfToUniChar( Tcl_UniChar byte; /* - * Unroll 1 to 3 (or 4) byte UTF-8 sequences. + * Unroll 1 to 4 byte UTF-8 sequences. */ byte = *((unsigned char *) src); if (byte < 0xC0) { /* * Handles properly formed UTF-8 characters between 0x01 and 0x7F. - * Also treats \0 and naked trail bytes 0x80 to 0xBF as valid + * Treats naked trail bytes 0x80 to 0x9F as valid characters from + * the cp1252 table. See: <https://en.wikipedia.org/wiki/UTF-8> + * Also treats \0 and other naked trail bytes 0xA0 to 0xBF as valid * characters representing themselves. */ -#if TCL_UTF_MAX == 4 +#if TCL_UTF_MAX <= 4 /* If *chPtr contains a high surrogate (produced by a previous * Tcl_UtfToUniChar() call) and the next 3 bytes are UTF-8 continuation * bytes, then we must produce a follow-up low surrogate. We only @@ -328,7 +386,11 @@ Tcl_UtfToUniChar( return 3; } #endif - *chPtr = byte; + if ((unsigned)(byte-0x80) < (unsigned)0x20) { + *chPtr = cp1252[byte-0x80]; + } else { + *chPtr = byte; + } return 1; } else if (byte < 0xE0) { if ((src[1] & 0xC0) == 0x80) { @@ -364,13 +426,12 @@ Tcl_UtfToUniChar( * represents itself. */ } -#if TCL_UTF_MAX > 3 else if (byte < 0xF8) { if (((src[1] & 0xC0) == 0x80) && ((src[2] & 0xC0) == 0x80) && ((src[3] & 0xC0) == 0x80)) { /* * Four-byte-character lead byte followed by three trail bytes. */ -#if TCL_UTF_MAX == 4 +#if TCL_UTF_MAX <= 4 Tcl_UniChar high = (((byte & 0x07) << 8) | ((src[1] & 0x3F) << 2) | ((src[2] & 0x3F) >> 4)) - 0x40; if (high >= 0x400) { @@ -394,12 +455,113 @@ Tcl_UtfToUniChar( * represents itself. */ } -#endif *chPtr = byte; return 1; } - + +#if (TCL_UTF_MAX > 4) && (defined(__CYGWIN__) || defined(_WIN32)) +int +TclUtfToWChar( + const char *src, /* The UTF-8 string. */ + WCHAR *chPtr)/* Filled with the WCHAR represented by + * the UTF-8 string. */ +{ + WCHAR byte; + + /* + * Unroll 1 to 4 byte UTF-8 sequences. + */ + + byte = *((unsigned char *) src); + if (byte < 0xC0) { + /* + * Handles properly formed UTF-8 characters between 0x01 and 0x7F. + * Treats naked trail bytes 0x80 to 0x9F as valid characters from + * the cp1252 table. See: <https://en.wikipedia.org/wiki/UTF-8> + * Also treats \0 and other naked trail bytes 0xA0 to 0xBF as valid + * characters representing themselves. + */ + + /* If *chPtr contains a high surrogate (produced by a previous + * Tcl_UtfToUniChar() call) and the next 3 bytes are UTF-8 continuation + * bytes, then we must produce a follow-up low surrogate. We only + * do that if the high surrogate matches the bits we encounter. + */ + if ((byte >= 0x80) + && (((((byte - 0x10) << 2) & 0xFC) | 0xD800) == (*chPtr & 0xFCFC)) + && ((src[1] & 0xF0) == (((*chPtr << 4) & 0x30) | 0x80)) + && ((src[2] & 0xC0) == 0x80)) { + *chPtr = ((src[1] & 0x0F) << 6) + (src[2] & 0x3F) + 0xDC00; + return 3; + } + if ((unsigned)(byte-0x80) < (unsigned)0x20) { + *chPtr = cp1252[byte-0x80]; + } else { + *chPtr = byte; + } + return 1; + } else if (byte < 0xE0) { + if ((src[1] & 0xC0) == 0x80) { + /* + * Two-byte-character lead-byte followed by a trail-byte. + */ + + *chPtr = (((byte & 0x1F) << 6) | (src[1] & 0x3F)); + if ((unsigned)(*chPtr - 1) >= (UNICODE_SELF - 1)) { + return 2; + } + } + + /* + * A two-byte-character lead-byte not followed by trail-byte + * represents itself. + */ + } else if (byte < 0xF0) { + if (((src[1] & 0xC0) == 0x80) && ((src[2] & 0xC0) == 0x80)) { + /* + * Three-byte-character lead byte followed by two trail bytes. + */ + + *chPtr = (((byte & 0x0F) << 12) + | ((src[1] & 0x3F) << 6) | (src[2] & 0x3F)); + if (*chPtr > 0x7FF) { + return 3; + } + } + + /* + * A three-byte-character lead-byte not followed by two trail-bytes + * represents itself. + */ + } + else if (byte < 0xF8) { + if (((src[1] & 0xC0) == 0x80) && ((src[2] & 0xC0) == 0x80) && ((src[3] & 0xC0) == 0x80)) { + /* + * Four-byte-character lead byte followed by three trail bytes. + */ + WCHAR high = (((byte & 0x07) << 8) | ((src[1] & 0x3F) << 2) + | ((src[2] & 0x3F) >> 4)) - 0x40; + if (high >= 0x400) { + /* out of range, < 0x10000 or > 0x10ffff */ + } else { + /* produce high surrogate, advance source pointer */ + *chPtr = 0xD800 + high; + return 1; + } + } + + /* + * A four-byte-character lead-byte not followed by three trail-bytes + * represents itself. + */ + } + + *chPtr = byte; + return 1; +} +#endif + /* *--------------------------------------------------------------------------- * @@ -448,12 +610,12 @@ Tcl_UtfToUniCharDString( w = wString; p = src; - end = src + length - TCL_UTF_MAX; + end = src + length - 4; while (p < end) { p += TclUtfToUniChar(p, &ch); *w++ = ch; } - end += TCL_UTF_MAX; + end += 4; while (p < end) { if (Tcl_UtfCharComplete(p, end-p)) { p += TclUtfToUniChar(p, &ch); @@ -464,11 +626,63 @@ Tcl_UtfToUniCharDString( } *w = '\0'; Tcl_DStringSetLength(dsPtr, - (oldLength + ((char *) w - (char *) wString))); + oldLength + ((char *) w - (char *) wString)); return wString; } - + +#if (TCL_UTF_MAX > 4) && (defined(__CYGWIN__) || defined(_WIN32)) +WCHAR * +TclUtfToWCharDString( + const char *src, /* UTF-8 string to convert to Unicode. */ + int 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 + * DString. */ +{ + WCHAR ch = 0, *w, *wString; + const char *p, *end; + int oldLength; + + if (length < 0) { + length = strlen(src); + } + + /* + * Unicode string length in Tcl_UniChars will be <= UTF-8 string length in + * bytes. + */ + + oldLength = Tcl_DStringLength(dsPtr); + + Tcl_DStringSetLength(dsPtr, + oldLength + (int) ((length + 1) * sizeof(WCHAR))); + wString = (WCHAR *) (Tcl_DStringValue(dsPtr) + oldLength); + + w = wString; + p = src; + end = src + length - 4; + while (p < end) { + p += TclUtfToWChar(p, &ch); + *w++ = ch; + } + end += 4; + while (p < end) { + if (Tcl_UtfCharComplete(p, end-p)) { + p += TclUtfToWChar(p, &ch); + } else { + ch = UCHAR(*p++); + } + *w++ = ch; + } + *w = '\0'; + Tcl_DStringSetLength(dsPtr, + oldLength + ((char *) w - (char *) wString)); + + return wString; +} +#endif /* *--------------------------------------------------------------------------- * @@ -538,13 +752,13 @@ Tcl_NumUtfChars( } if (i < 0) i = INT_MAX; /* Bug [2738427] */ } else { - register const char *endPtr = src + length - TCL_UTF_MAX; + register const char *endPtr = src + length - 4; while (src < endPtr) { src += TclUtfToUniChar(src, &ch); i++; } - endPtr += TCL_UTF_MAX; + endPtr += 4; while ((src < endPtr) && Tcl_UtfCharComplete(src, endPtr - src)) { src += TclUtfToUniChar(src, &ch); i++; @@ -586,7 +800,7 @@ Tcl_UtfFindFirst( while (1) { len = TclUtfToUniChar(src, &find); fullchar = find; -#if TCL_UTF_MAX == 4 +#if TCL_UTF_MAX <= 4 if ((ch >= 0xD800) && (len < 3)) { len += TclUtfToUniChar(src + len, &find); fullchar = (((fullchar & 0x3ff) << 10) | (find & 0x3ff)) + 0x10000; @@ -634,7 +848,7 @@ Tcl_UtfFindLast( while (1) { len = TclUtfToUniChar(src, &find); fullchar = find; -#if TCL_UTF_MAX == 4 +#if TCL_UTF_MAX <= 4 if ((ch >= 0xD800) && (len < 3)) { len += TclUtfToUniChar(src + len, &find); fullchar = (((fullchar & 0x3ff) << 10) | (find & 0x3ff)) + 0x10000; @@ -677,7 +891,7 @@ Tcl_UtfNext( Tcl_UniChar ch = 0; int len = TclUtfToUniChar(src, &ch); -#if TCL_UTF_MAX == 4 +#if TCL_UTF_MAX <= 4 if ((ch >= 0xD800) && (len < 3)) { len += TclUtfToUniChar(src + len, &ch); } @@ -716,7 +930,7 @@ Tcl_UtfPrev( int i, byte; look = --src; - for (i = 0; i < TCL_UTF_MAX; i++) { + for (i = 0; i < 4; i++) { if (look < start) { if (src < start) { src = start; @@ -752,17 +966,33 @@ Tcl_UtfPrev( *--------------------------------------------------------------------------- */ -Tcl_UniChar +int Tcl_UniCharAtIndex( register const char *src, /* The UTF-8 string to dereference. */ register int index) /* The position of the desired character. */ { Tcl_UniChar ch = 0; + int fullchar = 0; +#if TCL_UTF_MAX <= 4 + int len = 0; +#endif while (index-- >= 0) { +#if TCL_UTF_MAX <= 4 + src += (len = TclUtfToUniChar(src, &ch)); +#else src += TclUtfToUniChar(src, &ch); +#endif + } + fullchar = ch; +#if TCL_UTF_MAX <= 4 + if ((ch >= 0xD800) && (len < 3)) { + /* If last Tcl_UniChar was a high surrogate, combine with low surrogate */ + (void)TclUtfToUniChar(src, &ch); + fullchar = (((fullchar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000; } - return ch; +#endif + return fullchar; } /* @@ -771,7 +1001,9 @@ Tcl_UniCharAtIndex( * Tcl_UtfAtIndex -- * * Returns a pointer to the specified character (not byte) position in - * the UTF-8 string. + * the UTF-8 string. If TCL_UTF_MAX <= 4, characters > U+FFFF count as + * 2 positions, but then the pointer should never be placed between + * the two positions. * * Results: * As above. @@ -794,7 +1026,7 @@ Tcl_UtfAtIndex( len = TclUtfToUniChar(src, &ch); src += len; } -#if TCL_UTF_MAX == 4 +#if TCL_UTF_MAX <= 4 if ((ch >= 0xD800) && (len < 3)) { /* Index points at character following high Surrogate */ src += TclUtfToUniChar(src, &ch); @@ -812,7 +1044,7 @@ Tcl_UtfAtIndex( * * Results: * Stores the bytes represented by the backslash sequence in dst and - * returns the number of bytes written to dst. At most TCL_UTF_MAX bytes + * returns the number of bytes written to dst. At most 4 bytes * are written to dst; dst must have been large enough to accept those * bytes. If readPtr isn't NULL then it is filled in with a count of the * number of bytes in the backslash sequence. @@ -878,7 +1110,8 @@ int Tcl_UtfToUpper( char *str) /* String to convert in place. */ { - Tcl_UniChar ch = 0, upChar; + Tcl_UniChar ch = 0; + int upChar; char *src, *dst; int len; @@ -889,7 +1122,15 @@ Tcl_UtfToUpper( src = dst = str; while (*src) { len = TclUtfToUniChar(src, &ch); - upChar = Tcl_UniCharToUpper(ch); + upChar = ch; +#if TCL_UTF_MAX <= 4 + if ((ch >= 0xD800) && (len < 3)) { + len += TclUtfToUniChar(src + len, &ch); + /* Combine surrogates */ + upChar = (((upChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000; + } +#endif + upChar = Tcl_UniCharToUpper(upChar); /* * To keep badly formed Utf strings from getting inflated by the @@ -897,7 +1138,7 @@ Tcl_UtfToUpper( * char to dst if its size is <= the original char. */ - if (len < UtfCount(upChar)) { + if ((len < TclUtfCount(upChar)) || ((upChar & 0xF800) == 0xD800)) { memmove(dst, src, len); dst += len; } else { @@ -931,7 +1172,8 @@ int Tcl_UtfToLower( char *str) /* String to convert in place. */ { - Tcl_UniChar ch = 0, lowChar; + Tcl_UniChar ch = 0; + int lowChar; char *src, *dst; int len; @@ -942,7 +1184,15 @@ Tcl_UtfToLower( src = dst = str; while (*src) { len = TclUtfToUniChar(src, &ch); - lowChar = Tcl_UniCharToLower(ch); + lowChar = ch; +#if TCL_UTF_MAX <= 4 + if ((ch >= 0xD800) && (len < 3)) { + len += TclUtfToUniChar(src + len, &ch); + /* Combine surrogates */ + lowChar = (((lowChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000; + } +#endif + lowChar = Tcl_UniCharToLower(lowChar); /* * To keep badly formed Utf strings from getting inflated by the @@ -950,7 +1200,7 @@ Tcl_UtfToLower( * char to dst if its size is <= the original char. */ - if (len < UtfCount(lowChar)) { + if ((len < TclUtfCount(lowChar)) || ((lowChar & 0xF800) == 0xD800)) { memmove(dst, src, len); dst += len; } else { @@ -985,7 +1235,8 @@ int Tcl_UtfToTitle( char *str) /* String to convert in place. */ { - Tcl_UniChar ch = 0, titleChar, lowChar; + Tcl_UniChar ch = 0; + int titleChar, lowChar; char *src, *dst; int len; @@ -998,9 +1249,17 @@ Tcl_UtfToTitle( if (*src) { len = TclUtfToUniChar(src, &ch); - titleChar = Tcl_UniCharToTitle(ch); + titleChar = ch; +#if TCL_UTF_MAX <= 4 + if ((ch >= 0xD800) && (len < 3)) { + len += TclUtfToUniChar(src + len, &ch); + /* Combine surrogates */ + titleChar = (((titleChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000; + } +#endif + titleChar = Tcl_UniCharToTitle(titleChar); - if (len < UtfCount(titleChar)) { + if ((len < TclUtfCount(titleChar)) || ((titleChar & 0xF800) == 0xD800)) { memmove(dst, src, len); dst += len; } else { @@ -1011,12 +1270,19 @@ Tcl_UtfToTitle( while (*src) { len = TclUtfToUniChar(src, &ch); lowChar = ch; +#if TCL_UTF_MAX <= 4 + if ((ch >= 0xD800) && (len < 3)) { + len += TclUtfToUniChar(src + len, &ch); + /* Combine surrogates */ + lowChar = (((lowChar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000; + } +#endif /* Special exception for Georgian Asomtavruli chars, no titlecase. */ if ((unsigned)(lowChar - 0x1C90) >= 0x30) { lowChar = Tcl_UniCharToLower(lowChar); } - if (len < UtfCount(lowChar)) { + if ((len < TclUtfCount(lowChar)) || ((lowChar & 0xF800) == 0xD800)) { memmove(dst, src, len); dst += len; } else { @@ -1116,7 +1382,7 @@ Tcl_UtfNcmp( cs += TclUtfToUniChar(cs, &ch1); ct += TclUtfToUniChar(ct, &ch2); if (ch1 != ch2) { -#if TCL_UTF_MAX == 4 +#if TCL_UTF_MAX <= 4 /* Surrogates always report higher than non-surrogates */ if (((ch1 & 0xFC00) == 0xD800)) { if ((ch2 & 0xFC00) != 0xD800) { @@ -1167,7 +1433,7 @@ Tcl_UtfNcasecmp( cs += TclUtfToUniChar(cs, &ch1); ct += TclUtfToUniChar(ct, &ch2); if (ch1 != ch2) { -#if TCL_UTF_MAX == 4 +#if TCL_UTF_MAX <= 4 /* Surrogates always report higher than non-surrogates */ if (((ch1 & 0xFC00) == 0xD800)) { if ((ch2 & 0xFC00) != 0xD800) { @@ -1186,6 +1452,52 @@ Tcl_UtfNcasecmp( } return 0; } + +/* + *---------------------------------------------------------------------- + * + * Tcl_UtfCmp -- + * + * Compare UTF chars of string cs to string ct case sensitively. + * Replacement for strcmp in Tcl core, in places where UTF-8 should + * be handled. + * + * Results: + * Return <0 if cs < ct, 0 if cs == ct, or >0 if cs > ct. + * + * Side effects: + * None. + * + *---------------------------------------------------------------------- + */ + +int +TclUtfCmp( + const char *cs, /* UTF string to compare to ct. */ + const char *ct) /* UTF string cs is compared to. */ +{ + Tcl_UniChar ch1 = 0, ch2 = 0; + + while (*cs && *ct) { + cs += TclUtfToUniChar(cs, &ch1); + ct += TclUtfToUniChar(ct, &ch2); + if (ch1 != ch2) { +#if TCL_UTF_MAX <= 4 + /* Surrogates always report higher than non-surrogates */ + if (((ch1 & 0xFC00) == 0xD800)) { + if ((ch2 & 0xFC00) != 0xD800) { + return ch1; + } + } else if ((ch2 & 0xFC00) == 0xD800) { + return -ch2; + } +#endif + return ch1 - ch2; + } + } + return UCHAR(*cs) - UCHAR(*ct); +} + /* *---------------------------------------------------------------------- @@ -1216,7 +1528,7 @@ TclUtfCasecmp( cs += TclUtfToUniChar(cs, &ch1); ct += TclUtfToUniChar(ct, &ch2); if (ch1 != ch2) { -#if TCL_UTF_MAX == 4 +#if TCL_UTF_MAX <= 4 /* Surrogates always report higher than non-surrogates */ if (((ch1 & 0xFC00) == 0xD800)) { if ((ch2 & 0xFC00) != 0xD800) { @@ -1253,16 +1565,18 @@ TclUtfCasecmp( *---------------------------------------------------------------------- */ -Tcl_UniChar +int Tcl_UniCharToUpper( int ch) /* Unicode character to convert. */ { - int info = GetUniCharInfo(ch); + if (!UNICODE_OUT_OF_RANGE(ch)) { + int info = GetUniCharInfo(ch); - if (GetCaseType(info) & 0x04) { - ch -= GetDelta(info); + if (GetCaseType(info) & 0x04) { + ch -= GetDelta(info); + } } - return (Tcl_UniChar) ch; + return ch & 0x1FFFFF; } /* @@ -1281,17 +1595,19 @@ Tcl_UniCharToUpper( *---------------------------------------------------------------------- */ -Tcl_UniChar +int Tcl_UniCharToLower( int ch) /* Unicode character to convert. */ { - int info = GetUniCharInfo(ch); - int mode = GetCaseType(info); + if (!UNICODE_OUT_OF_RANGE(ch)) { + int info = GetUniCharInfo(ch); + int mode = GetCaseType(info); - if ((mode & 0x02) && (mode != 0x7)) { - ch += GetDelta(info); + if ((mode & 0x02) && (mode != 0x7)) { + ch += GetDelta(info); + } } - return (Tcl_UniChar) ch; + return ch & 0x1FFFFF; } /* @@ -1310,25 +1626,27 @@ Tcl_UniCharToLower( *---------------------------------------------------------------------- */ -Tcl_UniChar +int Tcl_UniCharToTitle( int ch) /* Unicode character to convert. */ { - int info = GetUniCharInfo(ch); - int mode = GetCaseType(info); + if (!UNICODE_OUT_OF_RANGE(ch)) { + int info = GetUniCharInfo(ch); + int mode = GetCaseType(info); - if (mode & 0x1) { - /* - * Subtract or add one depending on the original case. - */ + if (mode & 0x1) { + /* + * Subtract or add one depending on the original case. + */ - if (mode != 0x7) { - ch += ((mode & 0x4) ? -1 : 1); + if (mode != 0x7) { + ch += ((mode & 0x4) ? -1 : 1); + } + } else if (mode == 0x4) { + ch -= GetDelta(info); } - } else if (mode == 0x4) { - ch -= GetDelta(info); } - return (Tcl_UniChar) ch; + return ch & 0x1FFFFF; } /* @@ -1462,11 +1780,9 @@ int Tcl_UniCharIsAlnum( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { return 0; } -#endif return (((ALPHA_BITS | DIGIT_BITS) >> GetCategory(ch)) & 1); } @@ -1490,11 +1806,9 @@ int Tcl_UniCharIsAlpha( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { return 0; } -#endif return ((ALPHA_BITS >> GetCategory(ch)) & 1); } @@ -1518,7 +1832,6 @@ int Tcl_UniCharIsControl( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { ch &= 0x1FFFFF; if ((ch == 0xE0001) || ((ch >= 0xE0020) && (ch <= 0xE007f))) { @@ -1529,7 +1842,6 @@ Tcl_UniCharIsControl( } return 0; } -#endif return ((CONTROL_BITS >> GetCategory(ch)) & 1); } @@ -1553,11 +1865,9 @@ int Tcl_UniCharIsDigit( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { return 0; } -#endif return (GetCategory(ch) == DECIMAL_DIGIT_NUMBER); } @@ -1581,12 +1891,10 @@ int Tcl_UniCharIsGraph( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { ch &= 0x1FFFFF; return (ch >= 0xE0100) && (ch <= 0xE01EF); } -#endif return ((GRAPH_BITS >> GetCategory(ch)) & 1); } @@ -1610,11 +1918,9 @@ int Tcl_UniCharIsLower( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { return 0; } -#endif return (GetCategory(ch) == LOWERCASE_LETTER); } @@ -1638,12 +1944,10 @@ int Tcl_UniCharIsPrint( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { ch &= 0x1FFFFF; return (ch >= 0xE0100) && (ch <= 0xE01EF); } -#endif return (((GRAPH_BITS|SPACE_BITS) >> GetCategory(ch)) & 1); } @@ -1667,11 +1971,9 @@ int Tcl_UniCharIsPunct( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { return 0; } -#endif return ((PUNCT_BITS >> GetCategory(ch)) & 1); } @@ -1695,13 +1997,8 @@ int Tcl_UniCharIsSpace( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 /* Ignore upper 11 bits. */ ch &= 0x1FFFFF; -#else - /* Ignore upper 16 bits. */ - ch &= 0xFFFF; -#endif /* * If the character is within the first 127 characters, just use the @@ -1710,10 +2007,8 @@ Tcl_UniCharIsSpace( if (ch < 0x80) { return TclIsSpaceProc((char) ch); -#if TCL_UTF_MAX > 3 } else if (UNICODE_OUT_OF_RANGE(ch)) { return 0; -#endif } else if (ch == 0x0085 || ch == 0x180E || ch == 0x200B || ch == 0x202F || ch == 0x2060 || ch == 0xFEFF) { return 1; @@ -1742,11 +2037,9 @@ int Tcl_UniCharIsUpper( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { return 0; } -#endif return (GetCategory(ch) == UPPERCASE_LETTER); } @@ -1770,11 +2063,9 @@ int Tcl_UniCharIsWordChar( int ch) /* Unicode character to test. */ { -#if TCL_UTF_MAX > 3 if (UNICODE_OUT_OF_RANGE(ch)) { return 0; } -#endif return ((WORD_BITS >> GetCategory(ch)) & 1); } @@ -1860,7 +2151,7 @@ Tcl_UniCharCaseMatch( if ((p != '[') && (p != '?') && (p != '\\')) { if (nocase) { while (*uniStr && (p != *uniStr) - && (p != Tcl_UniCharToLower(*uniStr))) { + && (p != (Tcl_UniChar)Tcl_UniCharToLower(*uniStr))) { uniStr++; } } else { @@ -1900,13 +2191,13 @@ Tcl_UniCharCaseMatch( Tcl_UniChar startChar, endChar; uniPattern++; - ch1 = (nocase ? Tcl_UniCharToLower(*uniStr) : *uniStr); + ch1 = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*uniStr) : *uniStr); uniStr++; while (1) { if ((*uniPattern == ']') || (*uniPattern == 0)) { return 0; } - startChar = (nocase ? Tcl_UniCharToLower(*uniPattern) + startChar = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*uniPattern) : *uniPattern); uniPattern++; if (*uniPattern == '-') { @@ -1914,7 +2205,7 @@ Tcl_UniCharCaseMatch( if (*uniPattern == 0) { return 0; } - endChar = (nocase ? Tcl_UniCharToLower(*uniPattern) + endChar = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*uniPattern) : *uniPattern); uniPattern++; if (((startChar <= ch1) && (ch1 <= endChar)) @@ -2052,7 +2343,7 @@ TclUniCharMatch( if ((p != '[') && (p != '?') && (p != '\\')) { if (nocase) { while ((string < stringEnd) && (p != *string) - && (p != Tcl_UniCharToLower(*string))) { + && (p != (Tcl_UniChar)Tcl_UniCharToLower(*string))) { string++; } } else { @@ -2093,20 +2384,20 @@ TclUniCharMatch( Tcl_UniChar ch1, startChar, endChar; pattern++; - ch1 = (nocase ? Tcl_UniCharToLower(*string) : *string); + ch1 = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*string) : *string); string++; while (1) { if ((*pattern == ']') || (pattern == patternEnd)) { return 0; } - startChar = (nocase ? Tcl_UniCharToLower(*pattern) : *pattern); + startChar = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*pattern) : *pattern); pattern++; if (*pattern == '-') { pattern++; if (pattern == patternEnd) { return 0; } - endChar = (nocase ? Tcl_UniCharToLower(*pattern) + endChar = (nocase ? (Tcl_UniChar)Tcl_UniCharToLower(*pattern) : *pattern); pattern++; if (((startChar <= ch1) && (ch1 <= endChar)) |