diff options
Diffstat (limited to 'generic/tclEncoding.c')
-rw-r--r-- | generic/tclEncoding.c | 640 |
1 files changed, 352 insertions, 288 deletions
diff --git a/generic/tclEncoding.c b/generic/tclEncoding.c index 1842fb6..179ca17 100644 --- a/generic/tclEncoding.c +++ b/generic/tclEncoding.c @@ -92,7 +92,7 @@ typedef struct TableEncodingData { */ typedef struct EscapeSubTable { - unsigned int sequenceLen; /* Length of following string. */ + unsigned sequenceLen; /* Length of following string. */ char sequence[16]; /* Escape code that marks this encoding. */ char name[32]; /* Name for encoding. */ Encoding *encodingPtr; /* Encoding loaded using above name, or NULL @@ -104,10 +104,10 @@ typedef struct EscapeEncodingData { int fallback; /* Character (in this encoding) to substitute * when this encoding cannot represent a UTF-8 * character. */ - unsigned int initLen; /* Length of following string. */ + unsigned initLen; /* Length of following string. */ char init[16]; /* String to emit or expect before first char * in conversion. */ - unsigned int finalLen; /* Length of following string. */ + unsigned finalLen; /* Length of following string. */ char final[16]; /* String to emit or expect after last char in * conversion. */ char prefixBytes[256]; /* If a byte in the input stream is the first @@ -123,7 +123,7 @@ typedef struct EscapeEncodingData { } EscapeEncodingData; /* - * constants used when loading an encoding file to identify the type of the + * Constants used when loading an encoding file to identify the type of the * file. */ @@ -180,9 +180,9 @@ TCL_DECLARE_MUTEX(encodingMutex) * the system encoding will be used to perform the conversion. */ -static Tcl_Encoding defaultEncoding; -static Tcl_Encoding systemEncoding; -Tcl_Encoding tclIdentityEncoding; +static Tcl_Encoding defaultEncoding = NULL; +static Tcl_Encoding systemEncoding = NULL; +Tcl_Encoding tclIdentityEncoding = NULL; /* * The following variable is used in the sparse matrix code for a @@ -276,7 +276,7 @@ static int Iso88591ToUtfProc(ClientData clientData, * See concerns raised in [Bug 1077262]. */ -static Tcl_ObjType encodingType = { +static const Tcl_ObjType encodingType = { "encoding", FreeEncodingIntRep, DupEncodingIntRep, NULL, NULL }; @@ -294,7 +294,7 @@ static Tcl_ObjType encodingType = { * Standard Tcl return code. * * Side effects: - * Caches the Tcl_Encoding value as the internal rep of (*objPtr). + * Caches the Tcl_Encoding value as the internal rep of (*objPtr). * *---------------------------------------------------------------------- */ @@ -306,6 +306,7 @@ Tcl_GetEncodingFromObj( Tcl_Encoding *encodingPtr) { const char *name = Tcl_GetString(objPtr); + if (objPtr->typePtr != &encodingType) { Tcl_Encoding encoding = Tcl_GetEncoding(interp, name); @@ -313,7 +314,7 @@ Tcl_GetEncodingFromObj( return TCL_ERROR; } TclFreeIntRep(objPtr); - objPtr->internalRep.twoPtrValue.ptr1 = (VOID *) encoding; + objPtr->internalRep.twoPtrValue.ptr1 = encoding; objPtr->typePtr = &encodingType; } *encodingPtr = Tcl_GetEncoding(NULL, name); @@ -334,7 +335,7 @@ static void FreeEncodingIntRep( Tcl_Obj *objPtr) { - Tcl_FreeEncoding((Tcl_Encoding) objPtr->internalRep.twoPtrValue.ptr1); + Tcl_FreeEncoding(objPtr->internalRep.twoPtrValue.ptr1); objPtr->typePtr = NULL; } @@ -353,8 +354,7 @@ DupEncodingIntRep( Tcl_Obj *srcPtr, Tcl_Obj *dupPtr) { - dupPtr->internalRep.twoPtrValue.ptr1 = (VOID *) - Tcl_GetEncoding(NULL, srcPtr->bytes); + dupPtr->internalRep.twoPtrValue.ptr1 = Tcl_GetEncoding(NULL, srcPtr->bytes); } /* @@ -453,8 +453,8 @@ TclSetLibraryPath( * * FillEncodingFileMap -- * - * Called to bring the encoding file map in sync with the current value - * of the encoding search path. + * Called to bring the encoding file map in sync with the current value + * of the encoding search path. * * Scan the directories on the encoding search path, find the *.enc * files, and store the found pathnames in a map associated with the @@ -507,12 +507,12 @@ FillEncodingFileMap(void) Tcl_ListObjGetElements(NULL, matchFileList, &numFiles, &filev); for (j=0; j<numFiles; j++) { - Tcl_Obj *encodingName, *file; + Tcl_Obj *encodingName, *fileObj; - file = TclPathPart(NULL, filev[j], TCL_PATH_TAIL); - encodingName = TclPathPart(NULL, file, TCL_PATH_ROOT); + fileObj = TclPathPart(NULL, filev[j], TCL_PATH_TAIL); + encodingName = TclPathPart(NULL, fileObj, TCL_PATH_ROOT); Tcl_DictObjPut(NULL, map, encodingName, directory); - Tcl_DecrRefCount(file); + Tcl_DecrRefCount(fileObj); Tcl_DecrRefCount(encodingName); } Tcl_DecrRefCount(matchFileList); @@ -544,6 +544,9 @@ void TclInitEncodingSubsystem(void) { Tcl_EncodingType type; + TableEncodingData *dataPtr; + unsigned size; + unsigned short i; if (encodingsInitialized) { return; @@ -565,10 +568,7 @@ TclInitEncodingSubsystem(void) type.freeProc = NULL; type.nullSize = 1; type.clientData = NULL; - - defaultEncoding = Tcl_CreateEncoding(&type); - tclIdentityEncoding = Tcl_GetEncoding(NULL, type.encodingName); - systemEncoding = Tcl_GetEncoding(NULL, type.encodingName); + tclIdentityEncoding = Tcl_CreateEncoding(&type); type.encodingName = "utf-8"; type.toUtfProc = UtfExtToUtfIntProc; @@ -593,43 +593,37 @@ TclInitEncodingSubsystem(void) * code to duplicate the structure of a table encoding here. */ - { - TableEncodingData *dataPtr = (TableEncodingData *) - ckalloc(sizeof(TableEncodingData)); - unsigned size; - unsigned short i; - - memset(dataPtr, 0, sizeof(TableEncodingData)); - dataPtr->fallback = '?'; - - size = 256*(sizeof(unsigned short *) + sizeof(unsigned short)); - dataPtr->toUnicode = (unsigned short **) ckalloc(size); - memset(dataPtr->toUnicode, 0, size); - dataPtr->fromUnicode = (unsigned short **) ckalloc(size); - memset(dataPtr->fromUnicode, 0, size); - - dataPtr->toUnicode[0] = (unsigned short *) (dataPtr->toUnicode + 256); - dataPtr->fromUnicode[0] = (unsigned short *) - (dataPtr->fromUnicode + 256); - for (i=1 ; i<256 ; i++) { - dataPtr->toUnicode[i] = emptyPage; - dataPtr->fromUnicode[i] = emptyPage; - } + dataPtr = ckalloc(sizeof(TableEncodingData)); + memset(dataPtr, 0, sizeof(TableEncodingData)); + dataPtr->fallback = '?'; - for (i=0 ; i<256 ; i++) { - dataPtr->toUnicode[0][i] = i; - dataPtr->fromUnicode[0][i] = i; - } + size = 256*(sizeof(unsigned short *) + sizeof(unsigned short)); + dataPtr->toUnicode = ckalloc(size); + memset(dataPtr->toUnicode, 0, size); + dataPtr->fromUnicode = ckalloc(size); + memset(dataPtr->fromUnicode, 0, size); - type.encodingName = "iso8859-1"; - type.toUtfProc = Iso88591ToUtfProc; - type.fromUtfProc = Iso88591FromUtfProc; - type.freeProc = TableFreeProc; - type.nullSize = 1; - type.clientData = dataPtr; - Tcl_CreateEncoding(&type); + dataPtr->toUnicode[0] = (unsigned short *) (dataPtr->toUnicode + 256); + dataPtr->fromUnicode[0] = (unsigned short *) (dataPtr->fromUnicode + 256); + for (i=1 ; i<256 ; i++) { + dataPtr->toUnicode[i] = emptyPage; + dataPtr->fromUnicode[i] = emptyPage; } + for (i=0 ; i<256 ; i++) { + dataPtr->toUnicode[0][i] = i; + dataPtr->fromUnicode[0][i] = i; + } + + type.encodingName = "iso8859-1"; + type.toUtfProc = Iso88591ToUtfProc; + type.fromUtfProc = Iso88591FromUtfProc; + type.freeProc = TableFreeProc; + type.nullSize = 1; + type.clientData = dataPtr; + defaultEncoding = Tcl_CreateEncoding(&type); + systemEncoding = Tcl_GetEncoding(NULL, type.encodingName); + encodingsInitialized = 1; } @@ -658,7 +652,10 @@ TclFinalizeEncodingSubsystem(void) Tcl_MutexLock(&encodingMutex); encodingsInitialized = 0; FreeEncoding(systemEncoding); + systemEncoding = NULL; + defaultEncoding = NULL; FreeEncoding(tclIdentityEncoding); + tclIdentityEncoding = NULL; hPtr = Tcl_FirstHashEntry(&encodingTable, &search); while (hPtr != NULL) { @@ -669,7 +666,7 @@ TclFinalizeEncodingSubsystem(void) * cleaned up. */ - FreeEncoding((Tcl_Encoding) Tcl_GetHashValue(hPtr)); + FreeEncoding(Tcl_GetHashValue(hPtr)); hPtr = Tcl_FirstHashEntry(&encodingTable, &search); } @@ -682,15 +679,15 @@ TclFinalizeEncodingSubsystem(void) * * Tcl_GetDefaultEncodingDir -- * - * Legacy public interface to retrieve first directory in the encoding - * searchPath. + * Legacy public interface to retrieve first directory in the encoding + * searchPath. * * Results: * The directory pathname, as a string, or NULL for an empty encoding * search path. * * Side effects: - * None. + * None. * *------------------------------------------------------------------------- */ @@ -715,14 +712,14 @@ Tcl_GetDefaultEncodingDir(void) * * Tcl_SetDefaultEncodingDir -- * - * Legacy public interface to set the first directory in the encoding - * search path. + * Legacy public interface to set the first directory in the encoding + * search path. * * Results: - * None. + * None. * * Side effects: - * Modifies the encoding search path. + * Modifies the encoding search path. * *------------------------------------------------------------------------- */ @@ -782,7 +779,7 @@ Tcl_GetEncoding( hPtr = Tcl_FindHashEntry(&encodingTable, name); if (hPtr != NULL) { - encodingPtr = (Encoding *) Tcl_GetHashValue(hPtr); + encodingPtr = Tcl_GetHashValue(hPtr); encodingPtr->refCount++; Tcl_MutexUnlock(&encodingMutex); return (Tcl_Encoding) encodingPtr; @@ -841,9 +838,8 @@ static void FreeEncoding( Tcl_Encoding encoding) { - Encoding *encodingPtr; + Encoding *encodingPtr = (Encoding *) encoding; - encodingPtr = (Encoding *) encoding; if (encodingPtr == NULL) { return; } @@ -853,13 +849,13 @@ FreeEncoding( encodingPtr->refCount--; if (encodingPtr->refCount == 0) { if (encodingPtr->freeProc != NULL) { - (*encodingPtr->freeProc)(encodingPtr->clientData); + encodingPtr->freeProc(encodingPtr->clientData); } if (encodingPtr->hPtr != NULL) { Tcl_DeleteHashEntry(encodingPtr->hPtr); } - ckfree((char *) encodingPtr->name); - ckfree((char *) encodingPtr); + ckfree(encodingPtr->name); + ckfree(encodingPtr); } } @@ -929,9 +925,10 @@ Tcl_GetEncodingNames( Tcl_MutexLock(&encodingMutex); for (hPtr = Tcl_FirstHashEntry(&encodingTable, &search); hPtr != NULL; hPtr = Tcl_NextHashEntry(&search)) { - Encoding *encodingPtr = (Encoding *) Tcl_GetHashValue(hPtr); + Encoding *encodingPtr = Tcl_GetHashValue(hPtr); + Tcl_CreateHashEntry(&table, - (char *) Tcl_NewStringObj(encodingPtr->name, -1), &dummy); + Tcl_NewStringObj(encodingPtr->name, -1), &dummy); } Tcl_MutexUnlock(&encodingMutex); @@ -944,7 +941,7 @@ Tcl_GetEncodingNames( Tcl_DictObjFirst(NULL, map, &mapSearch, &name, NULL, &done); for (; !done; Tcl_DictObjNext(&mapSearch, &name, NULL, &done)) { - Tcl_CreateHashEntry(&table, (char *) name, &dummy); + Tcl_CreateHashEntry(&table, name, &dummy); } /* @@ -1058,13 +1055,13 @@ Tcl_CreateEncoding( * reference goes away. */ - encodingPtr = (Encoding *) Tcl_GetHashValue(hPtr); + encodingPtr = Tcl_GetHashValue(hPtr); encodingPtr->hPtr = NULL; } - name = ckalloc((unsigned) strlen(typePtr->encodingName) + 1); + name = ckalloc(strlen(typePtr->encodingName) + 1); - encodingPtr = (Encoding *) ckalloc(sizeof(Encoding)); + encodingPtr = ckalloc(sizeof(Encoding)); encodingPtr->name = strcpy(name, typePtr->encodingName); encodingPtr->toUtfProc = typePtr->toUtfProc; encodingPtr->fromUtfProc = typePtr->fromUtfProc; @@ -1118,7 +1115,7 @@ Tcl_ExternalToUtfDString( { char *dst; Tcl_EncodingState state; - Encoding *encodingPtr; + const Encoding *encodingPtr; int flags, dstLen, result, soFar, srcRead, dstWrote, dstChars; Tcl_DStringInit(dstPtr); @@ -1133,15 +1130,14 @@ Tcl_ExternalToUtfDString( if (src == NULL) { srcLen = 0; } else if (srcLen < 0) { - srcLen = (*encodingPtr->lengthProc)(src); + srcLen = encodingPtr->lengthProc(src); } flags = TCL_ENCODING_START | TCL_ENCODING_END; while (1) { - result = (*encodingPtr->toUtfProc)(encodingPtr->clientData, src, - srcLen, flags, &state, dst, dstLen, &srcRead, &dstWrote, - &dstChars); + result = encodingPtr->toUtfProc(encodingPtr->clientData, src, srcLen, + flags, &state, dst, dstLen, &srcRead, &dstWrote, &dstChars); soFar = dst + dstWrote - Tcl_DStringValue(dstPtr); if (result != TCL_CONVERT_NOSPACE) { @@ -1209,8 +1205,11 @@ Tcl_ExternalToUtf( * correspond to the bytes stored in the * output buffer. */ { - Encoding *encodingPtr; - int result, srcRead, dstWrote, dstChars; + const Encoding *encodingPtr; + int result, srcRead, dstWrote, dstChars = 0; + int noTerminate = flags & TCL_ENCODING_NO_TERMINATE; + int charLimited = (flags & TCL_ENCODING_CHAR_LIMIT) && dstCharsPtr; + int maxChars = INT_MAX; Tcl_EncodingState state; if (encoding == NULL) { @@ -1221,7 +1220,7 @@ Tcl_ExternalToUtf( if (src == NULL) { srcLen = 0; } else if (srcLen < 0) { - srcLen = (*encodingPtr->lengthProc)(src); + srcLen = encodingPtr->lengthProc(src); } if (statePtr == NULL) { flags |= TCL_ENCODING_START | TCL_ENCODING_END; @@ -1235,19 +1234,40 @@ Tcl_ExternalToUtf( } if (dstCharsPtr == NULL) { dstCharsPtr = &dstChars; + flags &= ~TCL_ENCODING_CHAR_LIMIT; + } else if (charLimited) { + maxChars = *dstCharsPtr; } - /* - * If there are any null characters in the middle of the buffer, they will - * converted to the UTF-8 null character (\xC080). To get the actual \0 at - * the end of the destination buffer, we need to append it manually. - */ + if (!noTerminate) { + /* + * If there are any null characters in the middle of the buffer, + * they will converted to the UTF-8 null character (\xC080). To get + * the actual \0 at the end of the destination buffer, we need to + * append it manually. First make room for it... + */ - dstLen--; - result = (*encodingPtr->toUtfProc)(encodingPtr->clientData, src, srcLen, - flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, - dstCharsPtr); - dst[*dstWrotePtr] = '\0'; + dstLen--; + } + do { + int savedFlags = flags; + Tcl_EncodingState savedState = *statePtr; + + result = encodingPtr->toUtfProc(encodingPtr->clientData, src, srcLen, + flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, + dstCharsPtr); + if (*dstCharsPtr <= maxChars) { + break; + } + dstLen = Tcl_UtfAtIndex(dst, maxChars) - 1 - dst + TCL_UTF_MAX; + flags = savedFlags; + *statePtr = savedState; + } while (1); + if (!noTerminate) { + /* ...and then append it */ + + dst[*dstWrotePtr] = '\0'; + } return result; } @@ -1285,7 +1305,7 @@ Tcl_UtfToExternalDString( { char *dst; Tcl_EncodingState state; - Encoding *encodingPtr; + const Encoding *encodingPtr; int flags, dstLen, result, soFar, srcRead, dstWrote, dstChars; Tcl_DStringInit(dstPtr); @@ -1304,7 +1324,7 @@ Tcl_UtfToExternalDString( } flags = TCL_ENCODING_START | TCL_ENCODING_END; while (1) { - result = (*encodingPtr->fromUtfProc)(encodingPtr->clientData, src, + result = encodingPtr->fromUtfProc(encodingPtr->clientData, src, srcLen, flags, &state, dst, dstLen, &srcRead, &dstWrote, &dstChars); soFar = dst + dstWrote - Tcl_DStringValue(dstPtr); @@ -1377,7 +1397,7 @@ Tcl_UtfToExternal( * correspond to the bytes stored in the * output buffer. */ { - Encoding *encodingPtr; + const Encoding *encodingPtr; int result, srcRead, dstWrote, dstChars; Tcl_EncodingState state; @@ -1406,7 +1426,7 @@ Tcl_UtfToExternal( } dstLen -= encodingPtr->nullSize; - result = (*encodingPtr->fromUtfProc)(encodingPtr->clientData, src, srcLen, + result = encodingPtr->fromUtfProc(encodingPtr->clientData, src, srcLen, flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr); if (encodingPtr->nullSize == 2) { @@ -1434,7 +1454,7 @@ Tcl_UtfToExternal( * *--------------------------------------------------------------------------- */ - +#undef Tcl_FindExecutable void Tcl_FindExecutable( const char *argv0) /* The value of the application's argv[0] @@ -1453,9 +1473,9 @@ Tcl_FindExecutable( * Open the file believed to hold data for the encoding, "name". * * Results: - * Returns the readable Tcl_Channel from opening the file, or NULL if the - * file could not be successfully opened. If NULL was returned, an error - * message is left in interp's result object, unless interp was NULL. + * Returns the readable Tcl_Channel from opening the file, or NULL if the + * file could not be successfully opened. If NULL was returned, an error + * message is left in interp's result object, unless interp was NULL. * * Side effects: * Channel may be opened. Information about the filesystem may be cached @@ -1498,6 +1518,7 @@ OpenEncodingFileChannel( } if (!verified) { const char *dirString = Tcl_GetString(directory); + for (i=0; i<numDirs && !verified; i++) { if (strcmp(dirString, Tcl_GetString(dir[i])) == 0) { verified = 1; @@ -1550,7 +1571,8 @@ OpenEncodingFileChannel( } if ((NULL == chan) && (interp != NULL)) { - Tcl_AppendResult(interp, "unknown encoding \"", name, "\"", NULL); + Tcl_SetObjResult(interp, Tcl_ObjPrintf( + "unknown encoding \"%s\"", name)); Tcl_SetErrorCode(interp, "TCL", "LOOKUP", "ENCODING", name, NULL); } Tcl_DecrRefCount(fileNameObj); @@ -1624,7 +1646,9 @@ LoadEncodingFile( break; } if ((encoding == NULL) && (interp != NULL)) { - Tcl_AppendResult(interp, "invalid encoding file \"", name, "\"", NULL); + Tcl_SetObjResult(interp, Tcl_ObjPrintf( + "invalid encoding file \"%s\"", name)); + Tcl_SetErrorCode(interp, "TCL", "LOOKUP", "ENCODING", name, NULL); } Tcl_Close(NULL, chan); @@ -1663,11 +1687,11 @@ LoadTableEncoding( Tcl_DString lineString; Tcl_Obj *objPtr; char *line; - int i, hi, lo, numPages, symbol, fallback; + int i, hi, lo, numPages, symbol, fallback, len; unsigned char used[256]; - unsigned int size; + unsigned size; TableEncodingData *dataPtr; - unsigned short *pageMemPtr; + unsigned short *pageMemPtr, *page; Tcl_EncodingType encType; /* @@ -1714,7 +1738,7 @@ LoadTableEncoding( #undef PAGESIZE #define PAGESIZE (256 * sizeof(unsigned short)) - dataPtr = (TableEncodingData *) ckalloc(sizeof(TableEncodingData)); + dataPtr = ckalloc(sizeof(TableEncodingData)); memset(dataPtr, 0, sizeof(TableEncodingData)); dataPtr->fallback = fallback; @@ -1726,7 +1750,7 @@ LoadTableEncoding( */ size = 256 * sizeof(unsigned short *) + numPages * PAGESIZE; - dataPtr->toUnicode = (unsigned short **) ckalloc(size); + dataPtr->toUnicode = ckalloc(size); memset(dataPtr->toUnicode, 0, size); pageMemPtr = (unsigned short *) (dataPtr->toUnicode + 256); @@ -1734,7 +1758,7 @@ LoadTableEncoding( Tcl_IncrRefCount(objPtr); for (i = 0; i < numPages; i++) { int ch; - char *p; + const char *p; Tcl_ReadChars(chan, objPtr, 3 + 16 * (16 * 4 + 1), 0); p = Tcl_GetString(objPtr); @@ -1784,29 +1808,26 @@ LoadTableEncoding( } } size = 256 * sizeof(unsigned short *) + numPages * PAGESIZE; - dataPtr->fromUnicode = (unsigned short **) ckalloc(size); + dataPtr->fromUnicode = ckalloc(size); memset(dataPtr->fromUnicode, 0, size); pageMemPtr = (unsigned short *) (dataPtr->fromUnicode + 256); for (hi = 0; hi < 256; hi++) { if (dataPtr->toUnicode[hi] == NULL) { dataPtr->toUnicode[hi] = emptyPage; - } else { - for (lo = 0; lo < 256; lo++) { - int ch; - - ch = dataPtr->toUnicode[hi][lo]; - if (ch != 0) { - unsigned short *page; - - page = dataPtr->fromUnicode[ch >> 8]; - if (page == NULL) { - page = pageMemPtr; - pageMemPtr += 256; - dataPtr->fromUnicode[ch >> 8] = page; - } - page[ch & 0xff] = (unsigned short) ((hi << 8) + lo); + continue; + } + for (lo = 0; lo < 256; lo++) { + int ch = dataPtr->toUnicode[hi][lo]; + + if (ch != 0) { + page = dataPtr->fromUnicode[ch >> 8]; + if (page == NULL) { + page = pageMemPtr; + pageMemPtr += 256; + dataPtr->fromUnicode[ch >> 8] = page; } + page[ch & 0xff] = (unsigned short) ((hi << 8) + lo); } } } @@ -1825,8 +1846,6 @@ LoadTableEncoding( } } if (symbol) { - unsigned short *page; - /* * Make a special symbol encoding that not only maps the symbol * characters from their Unicode code points down into page 0, but @@ -1834,7 +1853,7 @@ LoadTableEncoding( * is so that a symbol font can be used to display a simple string * like "abcd" and have alpha, beta, chi, delta show up, rather than * have "unknown" chars show up because strictly speaking the symbol - * font doesn't have glyphs for those low ascii chars. + * font doesn't have glyphs for those low ASCII chars. */ page = dataPtr->fromUnicode[0]; @@ -1859,57 +1878,77 @@ LoadTableEncoding( */ Tcl_DStringInit(&lineString); - do { - int len; + + /* + * Skip leading empty lines. + */ + + while ((len = Tcl_Gets(chan, &lineString)) == 0) { + /* empty body */ + } + if (len < 0) { + goto doneParse; + } + + /* + * Require that it starts with an 'R'. + */ + + line = Tcl_DStringValue(&lineString); + if (line[0] != 'R') { + goto doneParse; + } + + /* + * Read lines from the encoding until EOF. + */ + + for (TclDStringClear(&lineString); + (len = Tcl_Gets(chan, &lineString)) >= 0; + TclDStringClear(&lineString)) { + const unsigned char *p; + int to, from; /* - * Skip leading empty lines. + * Skip short lines. */ - while ((len = Tcl_Gets(chan, &lineString)) == 0) { - /* empty body */ + if (len < 5) { + continue; } - if (len < 0) { - break; - } - line = Tcl_DStringValue(&lineString); - if (line[0] != 'R') { - break; - } - for (Tcl_DStringSetLength(&lineString, 0); - (len = Tcl_Gets(chan, &lineString)) >= 0; - Tcl_DStringSetLength(&lineString, 0)) { - unsigned char* p; - int to, from; + /* + * Parse the line as a sequence of hex digits. + */ - if (len < 5) { - continue; - } - p = (unsigned char*) Tcl_DStringValue(&lineString); - to = (staticHex[p[0]] << 12) + (staticHex[p[1]] << 8) + p = (const unsigned char *) Tcl_DStringValue(&lineString); + to = (staticHex[p[0]] << 12) + (staticHex[p[1]] << 8) + + (staticHex[p[2]] << 4) + staticHex[p[3]]; + if (to == 0) { + continue; + } + for (p += 5, len -= 5; len >= 0 && *p; p += 5, len -= 5) { + from = (staticHex[p[0]] << 12) + (staticHex[p[1]] << 8) + (staticHex[p[2]] << 4) + staticHex[p[3]]; - if (to == 0) { - continue; - } - for (p += 5, len -= 5; len >= 0 && *p; p += 5, len -= 5) { - from = (staticHex[p[0]] << 12) + (staticHex[p[1]] << 8) - + (staticHex[p[2]] << 4) + staticHex[p[3]]; - if (from == 0) { - continue; - } - dataPtr->fromUnicode[from >> 8][from & 0xff] = to; + if (from == 0) { + continue; } + dataPtr->fromUnicode[from >> 8][from & 0xff] = to; } - } while (0); + } + doneParse: Tcl_DStringFree(&lineString); + /* + * Package everything into an encoding structure. + */ + encType.encodingName = name; encType.toUtfProc = TableToUtfProc; encType.fromUtfProc = TableFromUtfProc; encType.freeProc = TableFreeProc; encType.nullSize = (type == ENCODING_DOUBLEBYTE) ? 2 : 1; - encType.clientData = (ClientData) dataPtr; + encType.clientData = dataPtr; return Tcl_CreateEncoding(&encType); } @@ -1942,7 +1981,7 @@ LoadEscapeEncoding( Tcl_Channel chan) /* File containing new encoding. */ { int i; - unsigned int size; + unsigned size; Tcl_DString escapeData; char init[16], final[16]; EscapeEncodingData *dataPtr; @@ -1964,6 +2003,7 @@ LoadEscapeEncoding( } line = Tcl_DStringValue(&lineString); if (Tcl_SplitList(NULL, line, &argc, &argv) != TCL_OK) { + Tcl_DStringFree(&lineString); continue; } if (argc >= 2) { @@ -1991,8 +2031,8 @@ LoadEscapeEncoding( */ e = (Encoding *) Tcl_GetEncoding(NULL, est.name); - if (e && e->toUtfProc != TableToUtfProc && - e->toUtfProc != Iso88591ToUtfProc) { + if ((e != NULL) && (e->toUtfProc != TableToUtfProc) + && (e->toUtfProc != Iso88591ToUtfProc)) { Tcl_FreeEncoding((Tcl_Encoding) e); e = NULL; } @@ -2000,17 +2040,17 @@ LoadEscapeEncoding( Tcl_DStringAppend(&escapeData, (char *) &est, sizeof(est)); } } - ckfree((char *) argv); + ckfree(argv); Tcl_DStringFree(&lineString); } size = sizeof(EscapeEncodingData) - sizeof(EscapeSubTable) + Tcl_DStringLength(&escapeData); - dataPtr = (EscapeEncodingData *) ckalloc(size); + dataPtr = ckalloc(size); dataPtr->initLen = strlen(init); - strcpy(dataPtr->init, init); + memcpy(dataPtr->init, init, (unsigned) dataPtr->initLen + 1); dataPtr->finalLen = strlen(final); - strcpy(dataPtr->final, final); + memcpy(dataPtr->final, final, (unsigned) dataPtr->finalLen + 1); dataPtr->numSubTables = Tcl_DStringLength(&escapeData) / sizeof(EscapeSubTable); memcpy(dataPtr->subTables, Tcl_DStringValue(&escapeData), @@ -2028,12 +2068,16 @@ LoadEscapeEncoding( dataPtr->prefixBytes[UCHAR(dataPtr->final[0])] = 1; } + /* + * Package everything into an encoding structure. + */ + type.encodingName = name; type.toUtfProc = EscapeToUtfProc; type.fromUtfProc = EscapeFromUtfProc; type.freeProc = EscapeFreeProc; type.nullSize = 1; - type.clientData = (ClientData) dataPtr; + type.clientData = dataPtr; return Tcl_CreateEncoding(&type); } @@ -2087,6 +2131,9 @@ BinaryProc( if (dstLen < 0) { dstLen = 0; } + if ((flags & TCL_ENCODING_CHAR_LIMIT) && srcLen > *dstCharsPtr) { + srcLen = *dstCharsPtr; + } if (srcLen > dstLen) { srcLen = dstLen; result = TCL_CONVERT_NOSPACE; @@ -2165,6 +2212,7 @@ UtfIntToUtfExtProc( * *------------------------------------------------------------------------- */ + static int UtfExtToUtfIntProc( ClientData clientData, /* Not used. */ @@ -2245,8 +2293,8 @@ UtfToUtfProc( * versa. */ { const char *srcStart, *srcEnd, *srcClose; - char *dstStart, *dstEnd; - int result, numChars; + const char *dstStart, *dstEnd; + int result, numChars, charLimit = INT_MAX; Tcl_UniChar ch; result = TCL_OK; @@ -2257,11 +2305,14 @@ UtfToUtfProc( if ((flags & TCL_ENCODING_END) == 0) { srcClose -= TCL_UTF_MAX; } + if (flags & TCL_ENCODING_CHAR_LIMIT) { + charLimit = *dstCharsPtr; + } dstStart = dst; dstEnd = dst + dstLen - TCL_UTF_MAX; - for (numChars = 0; src < srcEnd; numChars++) { + for (numChars = 0; src < srcEnd && numChars <= charLimit; numChars++) { if ((src > srcClose) && (!Tcl_UtfCharComplete(src, srcEnd - src))) { /* * If there is more string to follow, this will ensure that the @@ -2356,10 +2407,13 @@ UnicodeToUtfProc( * output buffer. */ { const char *srcStart, *srcEnd; - char *dstEnd, *dstStart; - int result, numChars; + const char *dstEnd, *dstStart; + int result, numChars, charLimit = INT_MAX; Tcl_UniChar ch; + if (flags & TCL_ENCODING_CHAR_LIMIT) { + charLimit = *dstCharsPtr; + } result = TCL_OK; if ((srcLen % sizeof(Tcl_UniChar)) != 0) { result = TCL_CONVERT_MULTIBYTE; @@ -2373,15 +2427,17 @@ UnicodeToUtfProc( dstStart = dst; dstEnd = dst + dstLen - TCL_UTF_MAX; - for (numChars = 0; src < srcEnd; numChars++) { + for (numChars = 0; src < srcEnd && numChars <= charLimit; numChars++) { if (dst > dstEnd) { result = TCL_CONVERT_NOSPACE; break; } + /* - * Special case for 1-byte utf chars for speed. Make sure we - * work with Tcl_UniChar-size data. + * Special case for 1-byte utf chars for speed. Make sure we work with + * Tcl_UniChar-size data. */ + ch = *(Tcl_UniChar *)src; if (ch && ch < 0x80) { *dst++ = (ch & 0xFF); @@ -2471,11 +2527,13 @@ UtfToUnicodeProc( break; } src += TclUtfToUniChar(src, &ch); + /* - * Need to handle this in a way that won't cause misalignment - * by casting dst to a Tcl_UniChar. [Bug 1122671] + * Need to handle this in a way that won't cause misalignment by + * casting dst to a Tcl_UniChar. [Bug 1122671] * XXX: This hard-codes the assumed size of Tcl_UniChar as 2. */ + #ifdef WORDS_BIGENDIAN *dst++ = (ch >> 8); *dst++ = (ch & 0xFF); @@ -2536,26 +2594,28 @@ TableToUtfProc( * output buffer. */ { const char *srcStart, *srcEnd; - char *dstEnd, *dstStart, *prefixBytes; - int result, byte, numChars; + const char *dstEnd, *dstStart, *prefixBytes; + int result, byte, numChars, charLimit = INT_MAX; Tcl_UniChar ch; - unsigned short **toUnicode; - unsigned short *pageZero; - TableEncodingData *dataPtr; + const unsigned short *const *toUnicode; + const unsigned short *pageZero; + TableEncodingData *dataPtr = clientData; + if (flags & TCL_ENCODING_CHAR_LIMIT) { + charLimit = *dstCharsPtr; + } srcStart = src; srcEnd = src + srcLen; dstStart = dst; dstEnd = dst + dstLen - TCL_UTF_MAX; - dataPtr = (TableEncodingData *) clientData; - toUnicode = dataPtr->toUnicode; + toUnicode = (const unsigned short *const *) dataPtr->toUnicode; prefixBytes = dataPtr->prefixBytes; pageZero = toUnicode[0]; result = TCL_OK; - for (numChars = 0; src < srcEnd; numChars++) { + for (numChars = 0; src < srcEnd && numChars <= charLimit; numChars++) { if (dst > dstEnd) { result = TCL_CONVERT_NOSPACE; break; @@ -2582,9 +2642,11 @@ TableToUtfProc( } ch = (Tcl_UniChar) byte; } + /* * Special case for 1-byte utf chars for speed. */ + if (ch && ch < 0x80) { *dst++ = (char) ch; } else { @@ -2645,17 +2707,16 @@ TableFromUtfProc( * output buffer. */ { const char *srcStart, *srcEnd, *srcClose; - char *dstStart, *dstEnd, *prefixBytes; + const char *dstStart, *dstEnd, *prefixBytes; Tcl_UniChar ch; int result, len, word, numChars; - TableEncodingData *dataPtr; - unsigned short **fromUnicode; + TableEncodingData *dataPtr = clientData; + const unsigned short *const *fromUnicode; result = TCL_OK; - dataPtr = (TableEncodingData *) clientData; prefixBytes = dataPtr->prefixBytes; - fromUnicode = dataPtr->fromUnicode; + fromUnicode = (const unsigned short *const *) dataPtr->fromUnicode; srcStart = src; srcEnd = src + srcLen; @@ -2767,9 +2828,12 @@ Iso88591ToUtfProc( * output buffer. */ { const char *srcStart, *srcEnd; - char *dstEnd, *dstStart; - int result, numChars; + const char *dstEnd, *dstStart; + int result, numChars, charLimit = INT_MAX; + if (flags & TCL_ENCODING_CHAR_LIMIT) { + charLimit = *dstCharsPtr; + } srcStart = src; srcEnd = src + srcLen; @@ -2777,7 +2841,7 @@ Iso88591ToUtfProc( dstEnd = dst + dstLen - TCL_UTF_MAX; result = TCL_OK; - for (numChars = 0; src < srcEnd; numChars++) { + for (numChars = 0; src < srcEnd && numChars <= charLimit; numChars++) { Tcl_UniChar ch; if (dst > dstEnd) { @@ -2785,9 +2849,11 @@ Iso88591ToUtfProc( break; } ch = (Tcl_UniChar) *((unsigned char *) src); + /* * Special case for 1-byte utf chars for speed. */ + if (ch && ch < 0x80) { *dst++ = (char) ch; } else { @@ -2846,7 +2912,7 @@ Iso88591FromUtfProc( * output buffer. */ { const char *srcStart, *srcEnd, *srcClose; - char *dstStart, *dstEnd; + const char *dstStart, *dstEnd; int result, numChars; result = TCL_OK; @@ -2929,16 +2995,17 @@ TableFreeProc( ClientData clientData) /* TableEncodingData that specifies * encoding. */ { - TableEncodingData *dataPtr; + TableEncodingData *dataPtr = clientData; /* * Make sure we aren't freeing twice on shutdown. [Bug 219314] */ - dataPtr = (TableEncodingData *) clientData; - ckfree((char *) dataPtr->toUnicode); - ckfree((char *) dataPtr->fromUnicode); - ckfree((char *) dataPtr); + ckfree(dataPtr->toUnicode); + dataPtr->toUnicode = NULL; + ckfree(dataPtr->fromUnicode); + dataPtr->fromUnicode = NULL; + ckfree(dataPtr); } /* @@ -2986,20 +3053,19 @@ EscapeToUtfProc( * correspond to the bytes stored in the * output buffer. */ { - EscapeEncodingData *dataPtr; - char *prefixBytes, *tablePrefixBytes; - unsigned short **tableToUnicode; - Encoding *encodingPtr; - int state, result, numChars; - const char *srcStart, *srcEnd; - char *dstStart, *dstEnd; + EscapeEncodingData *dataPtr = clientData; + const char *prefixBytes, *tablePrefixBytes, *srcStart, *srcEnd; + const unsigned short *const *tableToUnicode; + const Encoding *encodingPtr; + int state, result, numChars, charLimit = INT_MAX; + const char *dstStart, *dstEnd; + if (flags & TCL_ENCODING_CHAR_LIMIT) { + charLimit = *dstCharsPtr; + } result = TCL_OK; - tablePrefixBytes = NULL; /* lint. */ tableToUnicode = NULL; /* lint. */ - - dataPtr = (EscapeEncodingData *) clientData; prefixBytes = dataPtr->prefixBytes; encodingPtr = NULL; @@ -3014,7 +3080,7 @@ EscapeToUtfProc( state = 0; } - for (numChars = 0; src < srcEnd; ) { + for (numChars = 0; src < srcEnd && numChars <= charLimit; ) { int byte, hi, lo, ch; if (dst > dstEnd) { @@ -3023,9 +3089,9 @@ EscapeToUtfProc( } byte = *((unsigned char *) src); if (prefixBytes[byte]) { - unsigned int left, len, longest; + unsigned left, len, longest; int checked, i; - EscapeSubTable *subTablePtr; + const EscapeSubTable *subTablePtr; /* * Saw the beginning of an escape sequence. @@ -3123,9 +3189,10 @@ EscapeToUtfProc( TableEncodingData *tableDataPtr; encodingPtr = GetTableEncoding(dataPtr, state); - tableDataPtr = (TableEncodingData *) encodingPtr->clientData; + tableDataPtr = encodingPtr->clientData; tablePrefixBytes = tableDataPtr->prefixBytes; - tableToUnicode = tableDataPtr->toUnicode; + tableToUnicode = (const unsigned short *const*) + tableDataPtr->toUnicode; } if (tablePrefixBytes[byte]) { @@ -3200,19 +3267,17 @@ EscapeFromUtfProc( * correspond to the bytes stored in the * output buffer. */ { - EscapeEncodingData *dataPtr; - Encoding *encodingPtr; + EscapeEncodingData *dataPtr = clientData; + const Encoding *encodingPtr; const char *srcStart, *srcEnd, *srcClose; - char *dstStart, *dstEnd; + const char *dstStart, *dstEnd; int state, result, numChars; - TableEncodingData *tableDataPtr; - char *tablePrefixBytes; - unsigned short **tableFromUnicode; + const TableEncodingData *tableDataPtr; + const char *tablePrefixBytes; + const unsigned short *const *tableFromUnicode; result = TCL_OK; - dataPtr = (EscapeEncodingData *) clientData; - srcStart = src; srcEnd = src + srcLen; srcClose = srcEnd; @@ -3224,7 +3289,7 @@ EscapeFromUtfProc( dstEnd = dst + dstLen - 1; /* - * RFC1468 states that the text starts in ASCII, and switches to Japanese + * RFC 1468 states that the text starts in ASCII, and switches to Japanese * characters, and that the text must end in ASCII. [Patch 474358] */ @@ -3242,12 +3307,13 @@ EscapeFromUtfProc( } encodingPtr = GetTableEncoding(dataPtr, state); - tableDataPtr = (TableEncodingData *) encodingPtr->clientData; + tableDataPtr = encodingPtr->clientData; tablePrefixBytes = tableDataPtr->prefixBytes; - tableFromUnicode = tableDataPtr->fromUnicode; + tableFromUnicode = (const unsigned short *const *) + tableDataPtr->fromUnicode; for (numChars = 0; src < srcEnd; numChars++) { - unsigned int len; + unsigned len; int word; Tcl_UniChar ch; @@ -3265,13 +3331,13 @@ EscapeFromUtfProc( if ((word == 0) && (ch != 0)) { int oldState; - EscapeSubTable *subTablePtr; + const EscapeSubTable *subTablePtr; oldState = state; for (state = 0; state < dataPtr->numSubTables; state++) { encodingPtr = GetTableEncoding(dataPtr, state); - tableDataPtr = (TableEncodingData *) encodingPtr->clientData; - word = tableDataPtr->fromUnicode[(ch >> 8)][ch & 0xff]; + tableDataPtr = encodingPtr->clientData; + word = tableDataPtr->fromUnicode[(ch >> 8)][ch & 0xff]; if (word != 0) { break; } @@ -3284,12 +3350,13 @@ EscapeFromUtfProc( break; } encodingPtr = GetTableEncoding(dataPtr, state); - tableDataPtr = (TableEncodingData *) encodingPtr->clientData; + tableDataPtr = encodingPtr->clientData; word = tableDataPtr->fallback; } - tablePrefixBytes = tableDataPtr->prefixBytes; - tableFromUnicode = tableDataPtr->fromUnicode; + tablePrefixBytes = (const char *) tableDataPtr->prefixBytes; + tableFromUnicode = (const unsigned short *const *) + tableDataPtr->fromUnicode; /* * The state variable has the value of oldState when word is 0. @@ -3337,22 +3404,22 @@ EscapeFromUtfProc( } if ((result == TCL_OK) && (flags & TCL_ENCODING_END)) { - unsigned int len = dataPtr->subTables[0].sequenceLen; + unsigned len = dataPtr->subTables[0].sequenceLen; + /* - * Certain encodings like iso2022-jp need to write - * an escape sequence after all characters have - * been converted. This logic checks that enough - * room is available in the buffer for the escape bytes. - * The TCL_ENCODING_END flag is cleared after a final - * escape sequence has been added to the buffer so - * that another call to this method does not attempt - * to append escape bytes a second time. + * Certain encodings like iso2022-jp need to write an escape sequence + * after all characters have been converted. This logic checks that + * enough room is available in the buffer for the escape bytes. The + * TCL_ENCODING_END flag is cleared after a final escape sequence has + * been added to the buffer so that another call to this method does + * not attempt to append escape bytes a second time. */ + if ((dst + dataPtr->finalLen + (state?len:0)) > dstEnd) { result = TCL_CONVERT_NOSPACE; } else { if (state) { - memcpy(dst, dataPtr->subTables[0].sequence, (size_t) len); + memcpy(dst, dataPtr->subTables[0].sequence, len); dst += len; } memcpy(dst, dataPtr->final, (size_t) dataPtr->finalLen); @@ -3390,33 +3457,34 @@ EscapeFreeProc( ClientData clientData) /* EscapeEncodingData that specifies * encoding. */ { - EscapeEncodingData *dataPtr; + EscapeEncodingData *dataPtr = clientData; EscapeSubTable *subTablePtr; int i; - dataPtr = (EscapeEncodingData *) clientData; if (dataPtr == NULL) { return; } + /* - * The subTables should be freed recursively in normal operation but not - * during TclFinalizeEncodingSubsystem because they are also present as a - * weak reference in the toplevel encodingTable (ie they don't have a +1 - * refcount for this), and unpredictable nuking order could remove them - * from under the following loop's feet [Bug 2891556]. - * - * The encodingsInitialized flag, being reset on entry to TFES, can serve - * as a "not in finalization" test. + * The subTables should be freed recursively in normal operation but not + * during TclFinalizeEncodingSubsystem because they are also present as a + * weak reference in the toplevel encodingTable (i.e., they don't have a + * +1 refcount for this), and unpredictable nuking order could remove them + * from under the following loop's feet. [Bug 2891556] + * + * The encodingsInitialized flag, being reset on entry to TFES, can serve + * as a "not in finalization" test. */ - if (encodingsInitialized) - { - subTablePtr = dataPtr->subTables; - for (i = 0; i < dataPtr->numSubTables; i++) { - FreeEncoding((Tcl_Encoding) subTablePtr->encodingPtr); - subTablePtr++; - } + + if (encodingsInitialized) { + subTablePtr = dataPtr->subTables; + for (i = 0; i < dataPtr->numSubTables; i++) { + FreeEncoding((Tcl_Encoding) subTablePtr->encodingPtr); + subTablePtr->encodingPtr = NULL; + subTablePtr++; } - ckfree((char *) dataPtr); + } + ckfree(dataPtr); } /* @@ -3444,11 +3512,8 @@ GetTableEncoding( EscapeEncodingData *dataPtr,/* Contains names of encodings. */ int state) /* Index in dataPtr of desired Encoding. */ { - EscapeSubTable *subTablePtr; - Encoding *encodingPtr; - - subTablePtr = &dataPtr->subTables[state]; - encodingPtr = subTablePtr->encodingPtr; + EscapeSubTable *subTablePtr = &dataPtr->subTables[state]; + Encoding *encodingPtr = subTablePtr->encodingPtr; if (encodingPtr == NULL) { encodingPtr = (Encoding *) Tcl_GetEncoding(NULL, subTablePtr->name); @@ -3523,43 +3588,43 @@ InitializeEncodingSearchPath( int *lengthPtr, Tcl_Encoding *encodingPtr) { - char *bytes; + const char *bytes; int i, numDirs, numBytes; - Tcl_Obj *libPath, *encodingObj, *searchPath; + Tcl_Obj *libPathObj, *encodingObj, *searchPathObj; TclNewLiteralStringObj(encodingObj, "encoding"); - TclNewObj(searchPath); + TclNewObj(searchPathObj); Tcl_IncrRefCount(encodingObj); - Tcl_IncrRefCount(searchPath); - libPath = TclGetLibraryPath(); - Tcl_IncrRefCount(libPath); - Tcl_ListObjLength(NULL, libPath, &numDirs); + Tcl_IncrRefCount(searchPathObj); + libPathObj = TclGetLibraryPath(); + Tcl_IncrRefCount(libPathObj); + Tcl_ListObjLength(NULL, libPathObj, &numDirs); for (i = 0; i < numDirs; i++) { - Tcl_Obj *directory, *path; + Tcl_Obj *directoryObj, *pathObj; Tcl_StatBuf stat; - Tcl_ListObjIndex(NULL, libPath, i, &directory); - path = Tcl_FSJoinToPath(directory, 1, &encodingObj); - Tcl_IncrRefCount(path); - if ((0 == Tcl_FSStat(path, &stat)) && S_ISDIR(stat.st_mode)) { - Tcl_ListObjAppendElement(NULL, searchPath, path); + Tcl_ListObjIndex(NULL, libPathObj, i, &directoryObj); + pathObj = Tcl_FSJoinToPath(directoryObj, 1, &encodingObj); + Tcl_IncrRefCount(pathObj); + if ((0 == Tcl_FSStat(pathObj, &stat)) && S_ISDIR(stat.st_mode)) { + Tcl_ListObjAppendElement(NULL, searchPathObj, pathObj); } - Tcl_DecrRefCount(path); + Tcl_DecrRefCount(pathObj); } - Tcl_DecrRefCount(libPath); + Tcl_DecrRefCount(libPathObj); Tcl_DecrRefCount(encodingObj); *encodingPtr = libraryPath.encoding; if (*encodingPtr) { ((Encoding *)(*encodingPtr))->refCount++; } - bytes = Tcl_GetStringFromObj(searchPath, &numBytes); + bytes = Tcl_GetStringFromObj(searchPathObj, &numBytes); *lengthPtr = numBytes; - *valuePtr = ckalloc((unsigned int) numBytes + 1); + *valuePtr = ckalloc(numBytes + 1); memcpy(*valuePtr, bytes, (size_t) numBytes + 1); - Tcl_DecrRefCount(searchPath); + Tcl_DecrRefCount(searchPathObj); } /* @@ -3569,4 +3634,3 @@ InitializeEncodingSearchPath( * fill-column: 78 * End: */ - |