diff options
Diffstat (limited to 'generic/tclEncoding.c')
| -rw-r--r-- | generic/tclEncoding.c | 967 |
1 files changed, 520 insertions, 447 deletions
diff --git a/generic/tclEncoding.c b/generic/tclEncoding.c index 646713d..8108771 100644 --- a/generic/tclEncoding.c +++ b/generic/tclEncoding.c @@ -8,7 +8,7 @@ * See the file "license.terms" for information on usage and redistribution * of this file, and for a DISCLAIMER OF ALL WARRANTIES. * - * RCS: @(#) $Id: tclEncoding.c,v 1.35 2005/05/10 18:34:34 kennykb Exp $ + * RCS: @(#) $Id: tclEncoding.c,v 1.36 2005/07/17 21:17:40 dkf Exp $ */ #include "tclInt.h" @@ -16,8 +16,8 @@ typedef size_t (LengthProc)_ANSI_ARGS_((CONST char *src)); /* - * The following data structure represents an encoding, which describes how - * to convert between various character sets and UTF-8. + * The following data structure represents an encoding, which describes how to + * convert between various character sets and UTF-8. */ typedef struct Encoding { @@ -28,8 +28,8 @@ typedef struct Encoding { * Tcl_EncodingType structure may not be * persistent. */ Tcl_EncodingConvertProc *toUtfProc; - /* Procedure to convert from external - * encoding into UTF-8. */ + /* Procedure to convert from external encoding + * into UTF-8. */ Tcl_EncodingConvertProc *fromUtfProc; /* Procedure to convert from UTF-8 into * external encoding. */ @@ -61,9 +61,9 @@ typedef struct Encoding { */ typedef struct TableEncodingData { - int fallback; /* Character (in this encoding) to - * substitute when this encoding cannot - * represent a UTF-8 character. */ + int fallback; /* Character (in this encoding) to substitute + * when this encoding cannot represent a UTF-8 + * character. */ char prefixBytes[256]; /* If a byte in the input stream is a lead * byte for a 2-byte sequence, the * corresponding entry in this array is 1, @@ -73,7 +73,8 @@ typedef struct TableEncodingData { * Each element of the toUnicode array points * to an array of 256 shorts. If there is no * corresponding character in Unicode, the - * value in the matrix is 0x0000. malloc'd. */ + * value in the matrix is 0x0000. + * malloc'd. */ unsigned short **fromUnicode; /* Two dimensional sparse matrix to map * characters from Unicode to the encoding. @@ -86,11 +87,11 @@ typedef struct TableEncodingData { /* * The following structures is the clientData for a dynamically-loaded, - * escape-driven encoding that is itself comprised of other simpler - * encodings. An example is "iso-2022-jp", which uses escape sequences to - * switch between ascii, jis0208, jis0212, gb2312, and ksc5601. Note that - * "escape-driven" does not necessarily mean that the ESCAPE character is - * the character used for switching character sets. + * escape-driven encoding that is itself comprised of other simpler encodings. + * An example is "iso-2022-jp", which uses escape sequences to switch between + * ascii, jis0208, jis0212, gb2312, and ksc5601. Note that "escape-driven" + * does not necessarily mean that the ESCAPE character is the character used + * for switching character sets. */ typedef struct EscapeSubTable { @@ -103,25 +104,25 @@ typedef struct EscapeSubTable { } EscapeSubTable; typedef struct EscapeEncodingData { - int fallback; /* Character (in this encoding) to - * substitute when this encoding cannot - * represent a UTF-8 character. */ + int fallback; /* Character (in this encoding) to substitute + * when this encoding cannot represent a UTF-8 + * character. */ unsigned int 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. */ - 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 character of one of the escape - * sequences in the following array, the - * corresponding entry in this array is 1, - * otherwise it is 0. */ + 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 + * character of one of the escape sequences in + * the following array, the corresponding + * entry in this array is 1, otherwise it is + * 0. */ int numSubTables; /* Length of following array. */ - EscapeSubTable subTables[1];/* Information about each EscapeSubTable - * used by this encoding type. The actual - * size will be as large as necessary to - * hold all EscapeSubTables. */ + EscapeSubTable subTables[1];/* Information about each EscapeSubTable used + * by this encoding type. The actual size + * will be as large as necessary to hold all + * EscapeSubTables. */ } EscapeEncodingData; /* @@ -135,49 +136,51 @@ typedef struct EscapeEncodingData { #define ENCODING_ESCAPE 3 /* - * A list of directories in which Tcl should look for *.enc files. - * This list is shared by all threads. Access is governed by a - * mutex lock. + * A list of directories in which Tcl should look for *.enc files. This list + * is shared by all threads. Access is governed by a mutex lock. */ -static TclInitProcessGlobalValueProc InitializeEncodingSearchPath; -static ProcessGlobalValue encodingSearchPath = - {0, 0, NULL, NULL, InitializeEncodingSearchPath, NULL, NULL}; +static TclInitProcessGlobalValueProc InitializeEncodingSearchPath; +static ProcessGlobalValue encodingSearchPath = { + 0, 0, NULL, NULL, InitializeEncodingSearchPath, NULL, NULL +}; /* - * A map from encoding names to the directories in which their data - * files have been seen. The string value of the map is shared by all - * threads. Access to the shared string is governed by a mutex lock. + * A map from encoding names to the directories in which their data files have + * been seen. The string value of the map is shared by all threads. Access + * to the shared string is governed by a mutex lock. */ -static ProcessGlobalValue encodingFileMap = - {0, 0, NULL, NULL, NULL, NULL, NULL}; +static ProcessGlobalValue encodingFileMap = { + 0, 0, NULL, NULL, NULL, NULL, NULL +}; /* - * A list of directories making up the "library path". Historically - * this search path has served many uses, but the only one remaining - * is a base for the encodingSearchPath above. If the application - * does not explicitly set the encodingSearchPath, then it will be - * initialized by appending /encoding to each directory in this - * "libraryPath". + * A list of directories making up the "library path". Historically this + * search path has served many uses, but the only one remaining is a base for + * the encodingSearchPath above. If the application does not explicitly set + * the encodingSearchPath, then it will be initialized by appending /encoding + * to each directory in this "libraryPath". */ -static ProcessGlobalValue libraryPath = - {0, 0, NULL, NULL, TclpInitLibraryPath, NULL, NULL}; + +static ProcessGlobalValue libraryPath = { + 0, 0, NULL, NULL, TclpInitLibraryPath, NULL, NULL +}; static int encodingsInitialized = 0; /* - * Hash table that keeps track of all loaded Encodings. Keys are - * the string names that represent the encoding, values are (Encoding *). + * Hash table that keeps track of all loaded Encodings. Keys are the string + * names that represent the encoding, values are (Encoding *). */ - + static Tcl_HashTable encodingTable; TCL_DECLARE_MUTEX(encodingMutex) /* - * The following are used to hold the default and current system encodings. - * If NULL is passed to one of the conversion routines, the current setting - * of the system encoding will be used to perform the conversion. + * The following are used to hold the default and current system encodings. + * If NULL is passed to one of the conversion routines, the current setting of + * the system encoding will be used to perform the conversion. */ static Tcl_Encoding defaultEncoding; @@ -221,10 +224,10 @@ static Tcl_Encoding LoadEncodingFile _ANSI_ARGS_((Tcl_Interp *interp, CONST char *name)); static Tcl_Encoding LoadTableEncoding _ANSI_ARGS_((CONST char *name, int type, Tcl_Channel chan)); -static Tcl_Encoding LoadEscapeEncoding _ANSI_ARGS_((CONST char *name, +static Tcl_Encoding LoadEscapeEncoding _ANSI_ARGS_((CONST char *name, Tcl_Channel chan)); -static Tcl_Channel OpenEncodingFileChannel _ANSI_ARGS_((Tcl_Interp *interp, - CONST char *name)); +static Tcl_Channel OpenEncodingFileChannel _ANSI_ARGS_(( + Tcl_Interp *interp, CONST char *name)); static void TableFreeProc _ANSI_ARGS_((ClientData clientData)); static int TableFromUtfProc _ANSI_ARGS_((ClientData clientData, CONST char *src, int srcLen, int flags, @@ -264,35 +267,35 @@ static int UtfExtToUtfIntProc _ANSI_ARGS_((ClientData clientData, int *dstCharsPtr)); /* - * A Tcl_ObjType for holding a cached Tcl_Encoding as the intrep. - * This should help the lifetime of encodings be more useful. - * See concerns raised in [Bug 1077262]. + * A Tcl_ObjType for holding a cached Tcl_Encoding as the intrep. This should + * help the lifetime of encodings be more useful. See concerns raised in [Bug + * 1077262]. */ static Tcl_ObjType EncodingType = { "encoding", FreeEncodingIntRep, DupEncodingIntRep, NULL, NULL }; - /* *---------------------------------------------------------------------- * * TclGetEncodingFromObj -- * - * Writes to (*encodingPtr) the Tcl_Encoding value of (*objPtr), - * if possible, and returns TCL_OK. If no such encoding exists, - * TCL_ERROR is returned, and if interp is non-NULL, an error message - * is written there. + * Writes to (*encodingPtr) the Tcl_Encoding value of (*objPtr), if + * possible, and returns TCL_OK. If no such encoding exists, TCL_ERROR + * is returned, and if interp is non-NULL, an error message is written + * there. * * Results: - * Standard Tcl return code. + * Standard Tcl return code. * * Side effects: * Caches the Tcl_Encoding value as the internal rep of (*objPtr). * *---------------------------------------------------------------------- */ -int + +int TclGetEncodingFromObj(interp, objPtr, encodingPtr) Tcl_Interp *interp; Tcl_Obj *objPtr; @@ -318,10 +321,11 @@ TclGetEncodingFromObj(interp, objPtr, encodingPtr) * * FreeEncodingIntRep -- * - * The Tcl_FreeInternalRepProc for the "encoding" Tcl_ObjType. + * The Tcl_FreeInternalRepProc for the "encoding" Tcl_ObjType. * *---------------------------------------------------------------------- */ + static void FreeEncodingIntRep(objPtr) Tcl_Obj *objPtr; @@ -334,10 +338,11 @@ FreeEncodingIntRep(objPtr) * * DupEncodingIntRep -- * - * The Tcl_DupInternalRepProc for the "encoding" Tcl_ObjType. + * The Tcl_DupInternalRepProc for the "encoding" Tcl_ObjType. * *---------------------------------------------------------------------- */ + static void DupEncodingIntRep(srcPtr, dupPtr) Tcl_Obj *srcPtr; @@ -352,12 +357,11 @@ DupEncodingIntRep(srcPtr, dupPtr) * * TclGetEncodingSearchPath -- * - * Keeps the per-thread copy of the encoding search path current - * with changes to the global copy. + * Keeps the per-thread copy of the encoding search path current with + * changes to the global copy. * * Results: - * Returns a "list" (Tcl_Obj *) that contains the encoding - * search path. + * Returns a "list" (Tcl_Obj *) that contains the encoding search path. * *---------------------------------------------------------------------- */ @@ -372,15 +376,15 @@ TclGetEncodingSearchPath() { * * TclSetEncodingSearchPath -- * - * Keeps the per-thread copy of the encoding search path current - * with changes to the global copy. + * Keeps the per-thread copy of the encoding search path current with + * changes to the global copy. * *---------------------------------------------------------------------- */ -int +int TclSetEncodingSearchPath(searchPath) - Tcl_Obj *searchPath; + Tcl_Obj *searchPath; { int dummy; @@ -396,11 +400,11 @@ TclSetEncodingSearchPath(searchPath) * * TclGetLibraryPath -- * - * Keeps the per-thread copy of the library path current - * with changes to the global copy. + * Keeps the per-thread copy of the library path current with changes to + * the global copy. * * Results: - * Returns a "list" (Tcl_Obj *) that contains the library path. + * Returns a "list" (Tcl_Obj *) that contains the library path. * *---------------------------------------------------------------------- */ @@ -415,19 +419,19 @@ TclGetLibraryPath() { * * TclSetLibraryPath -- * - * Keeps the per-thread copy of the library path current - * with changes to the global copy. + * Keeps the per-thread copy of the library path current with changes to + * the global copy. * - * NOTE: this routine returns void, so there's no way to - * report the error that searchPath is not a valid list. - * In that case, this routine will silently do nothing. + * NOTE: this routine returns void, so there's no way to report the error + * that searchPath is not a valid list. In that case, this routine will + * silently do nothing. * *---------------------------------------------------------------------- */ void TclSetLibraryPath(path) - Tcl_Obj *path; + Tcl_Obj *path; { int dummy; @@ -442,18 +446,17 @@ TclSetLibraryPath(path) * * 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 encoding name. + * Scan the directories on the encoding search path, find the *.enc + * files, and store the found pathnames in a map associated with the + * encoding name. * - * In particular, if $dir is on the encoding search path, and the - * file $dir/foo.enc is found, then store a "foo" -> $dir entry - * in the map. Later, any need for the "foo" encoding will quickly - * be able to construct the $dir/foo.enc pathname for reading the - * encoding data. + * In particular, if $dir is on the encoding search path, and the file + * $dir/foo.enc is found, then store a "foo" -> $dir entry in the map. + * Later, any need for the "foo" encoding will quickly * be able to + * construct the $dir/foo.enc pathname for reading the encoding data. * * Results: * None. @@ -475,23 +478,25 @@ FillEncodingFileMap() Tcl_ListObjLength(NULL, searchPath, &numDirs); map = Tcl_NewDictObj(); Tcl_IncrRefCount(map); + for (i = numDirs-1; i >= 0; i--) { - /* - * Iterate backwards through the search path so as we - * overwrite entries found, we favor files earlier on - * the search path. + /* + * Iterate backwards through the search path so as we overwrite + * entries found, we favor files earlier on the search path. */ + int j, numFiles; Tcl_Obj *directory, *matchFileList = Tcl_NewObj(); Tcl_Obj **filev; - Tcl_GlobTypeData readableFiles = - {TCL_GLOB_TYPE_FILE, TCL_GLOB_PERM_R, NULL, NULL}; + Tcl_GlobTypeData readableFiles = { + TCL_GLOB_TYPE_FILE, TCL_GLOB_PERM_R, NULL, NULL + }; Tcl_ListObjIndex(NULL, searchPath, i, &directory); Tcl_IncrRefCount(directory); Tcl_IncrRefCount(matchFileList); - Tcl_FSMatchInDirectory(NULL, matchFileList, - directory, "*.enc", &readableFiles); + Tcl_FSMatchInDirectory(NULL, matchFileList, directory, "*.enc", + &readableFiles); Tcl_ListObjGetElements(NULL, matchFileList, &numFiles, &filev); for (j=0; j<numFiles; j++) { @@ -517,7 +522,7 @@ FillEncodingFileMap() * TclInitEncodingSubsystem -- * * Initialize all resources used by this subsystem on a per-process - * basis. + * basis. * * Results: * None. @@ -540,11 +545,11 @@ TclInitEncodingSubsystem() Tcl_MutexLock(&encodingMutex); Tcl_InitHashTable(&encodingTable, TCL_STRING_KEYS); Tcl_MutexUnlock(&encodingMutex); - + /* - * Create a few initial encodings. Note that the UTF-8 to UTF-8 - * translation is not a no-op, because it will turn a stream of - * improperly formed UTF-8 into a properly formed stream. + * Create a few initial encodings. Note that the UTF-8 to UTF-8 + * translation is not a no-op, because it will turn a stream of improperly + * formed UTF-8 into a properly formed stream. */ type.encodingName = "identity"; @@ -599,19 +604,22 @@ TclFinalizeEncodingSubsystem() Tcl_HashEntry *hPtr; Tcl_MutexLock(&encodingMutex); - encodingsInitialized = 0; + encodingsInitialized = 0; FreeEncoding(systemEncoding); + hPtr = Tcl_FirstHashEntry(&encodingTable, &search); while (hPtr != NULL) { /* * Call FreeEncoding instead of doing it directly to handle refcounts - * like escape encodings use. [Bug #524674] - * Make sure to call Tcl_FirstHashEntry repeatedly so that all - * encodings are eventually cleaned up. + * like escape encodings use. [Bug #524674] Make sure to call + * Tcl_FirstHashEntry repeatedly so that all encodings are eventually + * cleaned up. */ + FreeEncoding((Tcl_Encoding) Tcl_GetHashValue(hPtr)); hPtr = Tcl_FirstHashEntry(&encodingTable, &search); } + Tcl_DeleteHashTable(&encodingTable); Tcl_MutexUnlock(&encodingMutex); } @@ -621,12 +629,12 @@ TclFinalizeEncodingSubsystem() * * 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. + * The directory pathname, as a string, or NULL for an empty encoding + * search path. * * Side effects: * None. @@ -654,8 +662,8 @@ Tcl_GetDefaultEncodingDir() * * 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. @@ -689,16 +697,16 @@ Tcl_SetDefaultEncodingDir(path) * * Results: * Returns a token that represents the encoding. If the name didn't - * refer to any known or loadable encoding, NULL is returned. If - * NULL was returned, an error message is left in interp's result - * object, unless interp was NULL. + * refer to any known or loadable encoding, NULL is returned. If NULL + * was returned, an error message is left in interp's result object, + * unless interp was NULL. * * Side effects: * The new encoding type is entered into a table visible to all - * interpreters, keyed off the encoding's name. For each call to - * this procedure, there should eventually be a call to - * Tcl_FreeEncoding, so that the database can be cleaned up when - * encodings aren't needed anymore. + * interpreters, keyed off the encoding's name. For each call to this + * procedure, there should eventually be a call to Tcl_FreeEncoding, so + * that the database can be cleaned up when encodings aren't needed + * anymore. * *------------------------------------------------------------------------- */ @@ -727,6 +735,7 @@ Tcl_GetEncoding(interp, name) return (Tcl_Encoding) encodingPtr; } Tcl_MutexUnlock(&encodingMutex); + return LoadEncodingFile(interp, name); } @@ -742,8 +751,8 @@ Tcl_GetEncoding(interp, name) * None. * * Side effects: - * The reference count associated with the encoding is decremented - * and the encoding may be deleted if nothing is using it anymore. + * The reference count associated with the encoding is decremented and + * the encoding may be deleted if nothing is using it anymore. * *--------------------------------------------------------------------------- */ @@ -762,15 +771,15 @@ Tcl_FreeEncoding(encoding) * * FreeEncoding -- * - * This procedure is called to release an encoding by procedures - * that already have the encodingMutex. + * This procedure is called to release an encoding by procedures that + * already have the encodingMutex. * * Results: * None. * * Side effects: - * The reference count associated with the encoding is decremented - * and the encoding may be deleted if nothing is using it anymore. + * The reference count associated with the encoding is decremented and + * the encoding may be deleted if nothing is using it anymore. * *---------------------------------------------------------------------- */ @@ -780,7 +789,7 @@ FreeEncoding(encoding) Tcl_Encoding encoding; { Encoding *encodingPtr; - + encodingPtr = (Encoding *) encoding; if (encodingPtr == NULL) { return; @@ -803,8 +812,8 @@ FreeEncoding(encoding) * * Tcl_GetEncodingName -- * - * Given an encoding, return the name that was used to constuct - * the encoding. + * Given an encoding, return the name that was used to constuct the + * encoding. * * Results: * The name of the encoding. @@ -819,13 +828,11 @@ CONST char * Tcl_GetEncodingName(encoding) Tcl_Encoding encoding; /* The encoding whose name to fetch. */ { - Encoding *encodingPtr; - if (encoding == NULL) { encoding = systemEncoding; } - encodingPtr = (Encoding *) encoding; - return encodingPtr->name; + + return ((Encoding *) encoding)->name; } /* @@ -833,8 +840,8 @@ Tcl_GetEncodingName(encoding) * * Tcl_GetEncodingNames -- * - * Get the list of all known encodings, including the ones stored - * as files on disk in the encoding path. + * Get the list of all known encodings, including the ones stored as + * files on disk in the encoding path. * * Results: * Modifies interp's result object to hold a list of all the available @@ -859,7 +866,10 @@ Tcl_GetEncodingNames(interp) Tcl_InitObjHashTable(&table); - /* Copy encoding names from loaded encoding table to table */ + /* + * Copy encoding names from loaded encoding table to table. + */ + Tcl_MutexLock(&encodingMutex); for (hPtr = Tcl_FirstHashEntry(&encodingTable, &search); hPtr != NULL; hPtr = Tcl_NextHashEntry(&search)) { @@ -872,16 +882,22 @@ Tcl_GetEncodingNames(interp) FillEncodingFileMap(); map = TclGetProcessGlobalValue(&encodingFileMap); - /* Copy encoding names from encoding file map to table */ + /* + * Copy encoding names from encoding file map to table. + */ + Tcl_DictObjFirst(NULL, map, &mapSearch, &name, NULL, &done); for (; !done; Tcl_DictObjNext(&mapSearch, &name, NULL, &done)) { Tcl_CreateHashEntry(&table, (char *) name, &dummy); } - /* Pull all encoding names from table into the result list */ + /* + * Pull all encoding names from table into the result list. + */ + for (hPtr = Tcl_FirstHashEntry(&table, &search); hPtr != NULL; hPtr = Tcl_NextHashEntry(&search)) { - Tcl_ListObjAppendElement(NULL, result, + Tcl_ListObjAppendElement(NULL, result, (Tcl_Obj *) Tcl_GetHashKey(&table, hPtr)); } Tcl_SetObjResult(interp, result); @@ -893,21 +909,21 @@ Tcl_GetEncodingNames(interp) * * Tcl_SetSystemEncoding -- * - * Sets the default encoding that should be used whenever the user - * passes a NULL value in to one of the conversion routines. - * If the supplied name is NULL, the system encoding is reset to the - * default system encoding. + * Sets the default encoding that should be used whenever the user passes + * a NULL value in to one of the conversion routines. If the supplied + * name is NULL, the system encoding is reset to the default system + * encoding. * * Results: - * The return value is TCL_OK if the system encoding was successfully - * set to the encoding specified by name, TCL_ERROR otherwise. If - * TCL_ERROR is returned, an error message is left in interp's result - * object, unless interp was NULL. + * The return value is TCL_OK if the system encoding was successfully set + * to the encoding specified by name, TCL_ERROR otherwise. If TCL_ERROR + * is returned, an error message is left in interp's result object, + * unless interp was NULL. * * Side effects: - * The reference count of the new system encoding is incremented. - * The reference count of the old system encoding is decremented and - * it may be freed. + * The reference count of the new system encoding is incremented. The + * reference count of the old system encoding is decremented and it may + * be freed. * *------------------------------------------------------------------------ */ @@ -948,25 +964,25 @@ Tcl_SetSystemEncoding(interp, name) * Tcl_CreateEncoding -- * * This procedure is called to define a new encoding and the procedures - * that are used to convert between the specified encoding and Unicode. + * that are used to convert between the specified encoding and Unicode. * * Results: - * Returns a token that represents the encoding. If an encoding with - * the same name already existed, the old encoding token remains - * valid and continues to behave as it used to, and will eventually - * be garbage collected when the last reference to it goes away. Any - * subsequent calls to Tcl_GetEncoding with the specified name will - * retrieve the most recent encoding token. + * Returns a token that represents the encoding. If an encoding with the + * same name already existed, the old encoding token remains valid and + * continues to behave as it used to, and will eventually be garbage + * collected when the last reference to it goes away. Any subsequent + * calls to Tcl_GetEncoding with the specified name will retrieve the + * most recent encoding token. * * Side effects: * The new encoding type is entered into a table visible to all - * interpreters, keyed off the encoding's name. For each call to - * this procedure, there should eventually be a call to - * Tcl_FreeEncoding, so that the database can be cleaned up when - * encodings aren't needed anymore. + * interpreters, keyed off the encoding's name. For each call to this + * procedure, there should eventually be a call to Tcl_FreeEncoding, so + * that the database can be cleaned up when encodings aren't needed + * anymore. * *--------------------------------------------------------------------------- - */ + */ Tcl_Encoding Tcl_CreateEncoding(typePtr) @@ -981,16 +997,16 @@ Tcl_CreateEncoding(typePtr) hPtr = Tcl_CreateHashEntry(&encodingTable, typePtr->encodingName, &new); if (new == 0) { /* - * Remove old encoding from hash table, but don't delete it until - * last reference goes away. + * Remove old encoding from hash table, but don't delete it until last + * reference goes away. */ - + encodingPtr = (Encoding *) Tcl_GetHashValue(hPtr); encodingPtr->hPtr = NULL; } name = ckalloc((unsigned) strlen(typePtr->encodingName) + 1); - + encodingPtr = (Encoding *) ckalloc(sizeof(Encoding)); encodingPtr->name = strcpy(name, typePtr->encodingName); encodingPtr->toUtfProc = typePtr->toUtfProc; @@ -1017,15 +1033,15 @@ Tcl_CreateEncoding(typePtr) * * Tcl_ExternalToUtfDString -- * - * Convert a source buffer from the specified encoding into UTF-8. - * If any of the bytes in the source buffer are invalid or cannot - * be represented in the target encoding, a default fallback - * character will be substituted. + * Convert a source buffer from the specified encoding into UTF-8. If any + * of the bytes in the source buffer are invalid or cannot be represented + * in the target encoding, a default fallback character will be + * substituted. * * Results: * The converted bytes are stored in the DString, which is then NULL - * terminated. The return value is a pointer to the value stored - * in the DString. + * terminated. The return value is a pointer to the value stored in the + * DString. * * Side effects: * None. @@ -1033,15 +1049,15 @@ Tcl_CreateEncoding(typePtr) *------------------------------------------------------------------------- */ -char * +char * Tcl_ExternalToUtfDString(encoding, src, srcLen, dstPtr) - Tcl_Encoding encoding; /* The encoding for the source string, or - * NULL for the default system encoding. */ + Tcl_Encoding encoding; /* The encoding for the source string, or NULL + * for the default system encoding. */ CONST char *src; /* Source string in specified encoding. */ int srcLen; /* Source string length in bytes, or < 0 for * encoding-specific string length. */ - Tcl_DString *dstPtr; /* Uninitialized or free DString in which - * the converted string is stored. */ + Tcl_DString *dstPtr; /* Uninitialized or free DString in which the + * converted string is stored. */ { char *dst; Tcl_EncodingState state; @@ -1051,7 +1067,7 @@ Tcl_ExternalToUtfDString(encoding, src, srcLen, dstPtr) Tcl_DStringInit(dstPtr); dst = Tcl_DStringValue(dstPtr); dstLen = dstPtr->spaceAvl - 1; - + if (encoding == NULL) { encoding = systemEncoding; } @@ -1062,16 +1078,20 @@ Tcl_ExternalToUtfDString(encoding, src, srcLen, dstPtr) } else if (srcLen < 0) { 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); soFar = dst + dstWrote - Tcl_DStringValue(dstPtr); + if (result != TCL_CONVERT_NOSPACE) { Tcl_DStringSetLength(dstPtr, soFar); return Tcl_DStringValue(dstPtr); } + flags &= ~TCL_ENCODING_START; src += srcRead; srcLen -= srcRead; @@ -1093,11 +1113,11 @@ Tcl_ExternalToUtfDString(encoding, src, srcLen, dstPtr) * * Results: * The return value is one of TCL_OK, TCL_CONVERT_MULTIBYTE, - * TCL_CONVERT_SYNTAX, TCL_CONVERT_UNKNOWN, or TCL_CONVERT_NOSPACE, - * as documented in tcl.h. + * TCL_CONVERT_SYNTAX, TCL_CONVERT_UNKNOWN, or TCL_CONVERT_NOSPACE, as + * documented in tcl.h. * * Side effects: - * The converted bytes are stored in the output buffer. + * The converted bytes are stored in the output buffer. * *------------------------------------------------------------------------- */ @@ -1106,19 +1126,19 @@ int Tcl_ExternalToUtf(interp, encoding, src, srcLen, flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr) Tcl_Interp *interp; /* Interp for error return, if not NULL. */ - Tcl_Encoding encoding; /* The encoding for the source string, or - * NULL for the default system encoding. */ + Tcl_Encoding encoding; /* The encoding for the source string, or NULL + * for the default system encoding. */ CONST char *src; /* Source string in specified encoding. */ int srcLen; /* Source string length in bytes, or < 0 for * encoding-specific string length. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ - char *dst; /* Output buffer in which converted string - * is stored. */ + char *dst; /* Output buffer in which converted string is + * stored. */ int dstLen; /* The maximum length of output buffer in * bytes. */ int *srcReadPtr; /* Filled with the number of bytes from the @@ -1136,7 +1156,7 @@ Tcl_ExternalToUtf(interp, encoding, src, srcLen, flags, statePtr, dst, Encoding *encodingPtr; int result, srcRead, dstWrote, dstChars; Tcl_EncodingState state; - + if (encoding == NULL) { encoding = systemEncoding; } @@ -1163,8 +1183,8 @@ Tcl_ExternalToUtf(interp, encoding, src, srcLen, flags, statePtr, dst, /* * 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. + * 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. */ dstLen--; @@ -1172,6 +1192,7 @@ Tcl_ExternalToUtf(interp, encoding, src, srcLen, flags, statePtr, dst, flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr); dst[*dstWrotePtr] = '\0'; + return result; } @@ -1180,15 +1201,15 @@ Tcl_ExternalToUtf(interp, encoding, src, srcLen, flags, statePtr, dst, * * Tcl_UtfToExternalDString -- * - * Convert a source buffer from UTF-8 into the specified encoding. - * If any of the bytes in the source buffer are invalid or cannot - * be represented in the target encoding, a default fallback - * character will be substituted. + * Convert a source buffer from UTF-8 into the specified encoding. If + * any of the bytes in the source buffer are invalid or cannot be + * represented in the target encoding, a default fallback character will + * be substituted. * * Results: - * The converted bytes are stored in the DString, which is then - * NULL terminated in an encoding-specific manner. The return value - * is a pointer to the value stored in the DString. + * The converted bytes are stored in the DString, which is then NULL + * terminated in an encoding-specific manner. The return value is a + * pointer to the value stored in the DString. * * Side effects: * None. @@ -1198,19 +1219,19 @@ Tcl_ExternalToUtf(interp, encoding, src, srcLen, flags, statePtr, dst, char * Tcl_UtfToExternalDString(encoding, src, srcLen, dstPtr) - Tcl_Encoding encoding; /* The encoding for the converted string, - * or NULL for the default system encoding. */ + Tcl_Encoding encoding; /* The encoding for the converted string, or + * NULL for the default system encoding. */ CONST char *src; /* Source string in UTF-8. */ int srcLen; /* Source string length in bytes, or < 0 for * strlen(). */ - Tcl_DString *dstPtr; /* Uninitialized or free DString in which - * the converted string is stored. */ + Tcl_DString *dstPtr; /* Uninitialized or free DString in which the + * converted string is stored. */ { char *dst; Tcl_EncodingState state; Encoding *encodingPtr; int flags, dstLen, result, soFar, srcRead, dstWrote, dstChars; - + Tcl_DStringInit(dstPtr); dst = Tcl_DStringValue(dstPtr); dstLen = dstPtr->spaceAvl - 1; @@ -1231,13 +1252,15 @@ Tcl_UtfToExternalDString(encoding, src, srcLen, dstPtr) srcLen, flags, &state, dst, dstLen, &srcRead, &dstWrote, &dstChars); soFar = dst + dstWrote - Tcl_DStringValue(dstPtr); + if (result != TCL_CONVERT_NOSPACE) { if (encodingPtr->nullSize == 2) { - Tcl_DStringSetLength(dstPtr, soFar + 1); + Tcl_DStringSetLength(dstPtr, soFar + 1); } Tcl_DStringSetLength(dstPtr, soFar); return Tcl_DStringValue(dstPtr); } + flags &= ~TCL_ENCODING_START; src += srcRead; srcLen -= srcRead; @@ -1259,11 +1282,11 @@ Tcl_UtfToExternalDString(encoding, src, srcLen, dstPtr) * * Results: * The return value is one of TCL_OK, TCL_CONVERT_MULTIBYTE, - * TCL_CONVERT_SYNTAX, TCL_CONVERT_UNKNOWN, or TCL_CONVERT_NOSPACE, - * as documented in tcl.h. + * TCL_CONVERT_SYNTAX, TCL_CONVERT_UNKNOWN, or TCL_CONVERT_NOSPACE, as + * documented in tcl.h. * * Side effects: - * The converted bytes are stored in the output buffer. + * The converted bytes are stored in the output buffer. * *------------------------------------------------------------------------- */ @@ -1272,14 +1295,14 @@ int Tcl_UtfToExternal(interp, encoding, src, srcLen, flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr) Tcl_Interp *interp; /* Interp for error return, if not NULL. */ - Tcl_Encoding encoding; /* The encoding for the converted string, - * or NULL for the default system encoding. */ + Tcl_Encoding encoding; /* The encoding for the converted string, or + * NULL for the default system encoding. */ CONST char *src; /* Source string in UTF-8. */ int srcLen; /* Source string length in bytes, or < 0 for * strlen(). */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ @@ -1302,7 +1325,7 @@ Tcl_UtfToExternal(interp, encoding, src, srcLen, flags, statePtr, dst, Encoding *encodingPtr; int result, srcRead, dstWrote, dstChars; Tcl_EncodingState state; - + if (encoding == NULL) { encoding = systemEncoding; } @@ -1335,7 +1358,7 @@ Tcl_UtfToExternal(interp, encoding, src, srcLen, flags, statePtr, dst, dst[*dstWrotePtr + 1] = '\0'; } dst[*dstWrotePtr] = '\0'; - + return result; } @@ -1351,8 +1374,8 @@ Tcl_UtfToExternal(interp, encoding, src, srcLen, flags, statePtr, dst, * None. * * Side effects: - * The absolute pathname for the application is computed and stored - * to be returned later be [info nameofexecutable]. + * The absolute pathname for the application is computed and stored to be + * returned later be [info nameofexecutable]. * *--------------------------------------------------------------------------- */ @@ -1375,14 +1398,14 @@ Tcl_FindExecutable(argv0) * 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 to speed later calls. + * Channel may be opened. Information about the filesystem may be cached + * to speed later calls. * *--------------------------------------------------------------------------- */ @@ -1390,8 +1413,8 @@ Tcl_FindExecutable(argv0) static Tcl_Channel OpenEncodingFileChannel(interp, name) Tcl_Interp *interp; /* Interp for error reporting, if not NULL. */ - CONST char *name; /* The name of the encoding file on disk - * and also the name for new encoding. */ + CONST char *name; /* The name of the encoding file on disk and + * also the name for new encoding. */ { Tcl_Obj *nameObj = Tcl_NewStringObj(name, -1); Tcl_Obj *fileNameObj = Tcl_DuplicateObj(nameObj); @@ -1407,7 +1430,10 @@ OpenEncodingFileChannel(interp, name) Tcl_IncrRefCount(fileNameObj); Tcl_DictObjGet(NULL, map, nameObj, &directory); - /* Check that any cached directory is still on the encoding search path */ + /* + * Check that any cached directory is still on the encoding search path. + */ + if (NULL != directory) { int verified = 0; @@ -1425,7 +1451,10 @@ OpenEncodingFileChannel(interp, name) } } if (!verified) { - /* Directory no longer on the search path. Remove from cache */ + /* + * Directory no longer on the search path. Remove from cache. + */ + map = Tcl_DuplicateObj(map); Tcl_DictObjRemove(NULL, map, nameObj); TclSetProcessGlobalValue(&encodingFileMap, map, NULL); @@ -1434,7 +1463,10 @@ OpenEncodingFileChannel(interp, name) } if (NULL != directory) { - /* Got a directory from the cache. Try to use it first */ + /* + * Got a directory from the cache. Try to use it first. + */ + Tcl_IncrRefCount(directory); path = Tcl_FSJoinToPath(directory, 1, &fileNameObj); Tcl_IncrRefCount(path); @@ -1443,25 +1475,33 @@ OpenEncodingFileChannel(interp, name) Tcl_DecrRefCount(path); } - /* Scan the search path until we find it. */ + /* + * Scan the search path until we find it. + */ + for (i=0; i<numDirs && (chan == NULL); i++) { path = Tcl_FSJoinToPath(dir[i], 1, &fileNameObj); Tcl_IncrRefCount(path); chan = Tcl_FSOpenFileChannel(NULL, path, "r", 0); Tcl_DecrRefCount(path); if (chan != NULL) { - /* Save directory in the cache */ + /* + * Save directory in the cache. + */ + map = Tcl_DuplicateObj(TclGetProcessGlobalValue(&encodingFileMap)); Tcl_DictObjPut(NULL, map, nameObj, dir[i]); TclSetProcessGlobalValue(&encodingFileMap, map, NULL); } } + if ((NULL == chan) && (interp != NULL)) { Tcl_AppendResult(interp, "unknown encoding \"", name, "\"", NULL); } Tcl_DecrRefCount(fileNameObj); Tcl_DecrRefCount(nameObj); Tcl_DecrRefCount(searchPath); + return chan; } @@ -1470,17 +1510,17 @@ OpenEncodingFileChannel(interp, name) * * LoadEncodingFile -- * - * Read a file that describes an encoding and create a new Encoding - * from the data. + * Read a file that describes an encoding and create a new Encoding from + * the data. * * Results: - * The return value is the newly loaded Encoding, or NULL if - * the file didn't exist of was in the incorrect format. If NULL was - * returned, an error message is left in interp's result object, - * unless interp was NULL. + * The return value is the newly loaded Encoding, or NULL if the file + * didn't exist of was in the incorrect format. If NULL was returned, an + * error message is left in interp's result object, unless interp was + * NULL. * * Side effects: - * File read from disk. + * File read from disk. * *--------------------------------------------------------------------------- */ @@ -1488,8 +1528,8 @@ OpenEncodingFileChannel(interp, name) static Tcl_Encoding LoadEncodingFile(interp, name) Tcl_Interp *interp; /* Interp for error reporting, if not NULL. */ - CONST char *name; /* The name of the encoding file on disk - * and also the name for new encoding. */ + CONST char *name; /* The name of the encoding file on disk and + * also the name for new encoding. */ { Tcl_Channel chan = NULL; Tcl_Encoding encoding = NULL; @@ -1515,27 +1555,24 @@ LoadEncodingFile(interp, name) } switch (ch) { - case 'S': { - encoding = LoadTableEncoding(name, ENCODING_SINGLEBYTE, chan); - break; - } - case 'D': { - encoding = LoadTableEncoding(name, ENCODING_DOUBLEBYTE, chan); - break; - } - case 'M': { - encoding = LoadTableEncoding(name, ENCODING_MULTIBYTE, chan); - break; - } - case 'E': { - encoding = LoadEscapeEncoding(name, chan); - break; - } + case 'S': + encoding = LoadTableEncoding(name, ENCODING_SINGLEBYTE, chan); + break; + case 'D': + encoding = LoadTableEncoding(name, ENCODING_DOUBLEBYTE, chan); + break; + case 'M': + encoding = LoadTableEncoding(name, ENCODING_MULTIBYTE, chan); + break; + case 'E': + encoding = LoadEscapeEncoding(name, chan); + break; } if ((encoding == NULL) && (interp != NULL)) { Tcl_AppendResult(interp, "invalid encoding file \"", name, "\"", NULL); } Tcl_Close(NULL, chan); + return encoding; } @@ -1544,17 +1581,17 @@ LoadEncodingFile(interp, name) * * LoadTableEncoding -- * - * Helper function for LoadEncodingTable(). Loads a table to that - * converts between Unicode and some other encoding and creates an + * Helper function for LoadEncodingTable(). Loads a table to that + * converts between Unicode and some other encoding and creates an * encoding (using a TableEncoding structure) from that information. * - * File contains binary data, but begins with a marker to indicate - * byte-ordering, so that same binary file can be read on either - * endian platforms. + * File contains binary data, but begins with a marker to indicate + * byte-ordering, so that same binary file can be read on either endian + * platforms. * * Results: - * The return value is the new encoding, or NULL if the encoding - * could not be created (because the file contained invalid data). + * The return value is the new encoding, or NULL if the encoding could + * not be created (because the file contained invalid data). * * Side effects: * None. @@ -1629,8 +1666,8 @@ LoadTableEncoding(name, type, chan) /* * Read the table that maps characters to Unicode. Performs a single - * malloc to get the memory for the array and all the pages needed by - * the array. + * malloc to get the memory for the array and all the pages needed by the + * array. */ size = 256 * sizeof(unsigned short *) + numPages * PAGESIZE; @@ -1646,15 +1683,15 @@ LoadTableEncoding(name, type, chan) Tcl_ReadChars(chan, objPtr, 3 + 16 * (16 * 4 + 1), 0); p = Tcl_GetString(objPtr); - hi = (staticHex[(unsigned int)p[0]] << 4) + staticHex[(unsigned int)p[1]]; + hi = (staticHex[UCHAR(p[0])] << 4) + staticHex[UCHAR(p[1])]; dataPtr->toUnicode[hi] = pageMemPtr; p += 2; for (lo = 0; lo < 256; lo++) { if ((lo & 0x0f) == 0) { p++; } - ch = (staticHex[(unsigned int)p[0]] << 12) + (staticHex[(unsigned int)p[1]] << 8) - + (staticHex[(unsigned int)p[2]] << 4) + staticHex[(unsigned int)p[3]]; + ch = (staticHex[UCHAR(p[0])] << 12) + (staticHex[UCHAR(p[1])] << 8) + + (staticHex[UCHAR(p[2])] << 4) + staticHex[UCHAR(p[3])]; if (ch != 0) { used[ch >> 8] = 1; } @@ -1664,7 +1701,7 @@ LoadTableEncoding(name, type, chan) } } TclDecrRefCount(objPtr); - + if (type == ENCODING_DOUBLEBYTE) { memset(dataPtr->prefixBytes, 1, sizeof(dataPtr->prefixBytes)); } else { @@ -1677,9 +1714,9 @@ LoadTableEncoding(name, type, chan) /* * Invert toUnicode array to produce the fromUnicode array. Performs a - * single malloc to get the memory for the array and all the pages - * needed by the array. While reading in the toUnicode array, we - * remembered what pages that would be needed for the fromUnicode array. + * single malloc to get the memory for the array and all the pages needed + * by the array. While reading in the toUnicode array, we remembered what + * pages that would be needed for the fromUnicode array. */ if (symbol) { @@ -1706,7 +1743,7 @@ LoadTableEncoding(name, type, chan) ch = dataPtr->toUnicode[hi][lo]; if (ch != 0) { unsigned short *page; - + page = dataPtr->fromUnicode[ch >> 8]; if (page == NULL) { page = pageMemPtr; @@ -1734,16 +1771,15 @@ LoadTableEncoding(name, type, chan) } 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 - * also ensure that the characters on page 0 map to themselves. - * This 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. + * also ensure that the characters on page 0 map to themselves. This + * 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. */ page = dataPtr->fromUnicode[0]; @@ -1762,15 +1798,22 @@ LoadTableEncoding(name, type, chan) dataPtr->fromUnicode[hi] = emptyPage; } } + /* * For trailing 'R'everse encoding, see [Patch #689341] */ + Tcl_DStringInit(&lineString); do { int len; - /* skip leading empty lines */ + + /* + * Skip leading empty lines. + */ + while ((len = Tcl_Gets(chan, &lineString)) == 0) ; + if (len < 0) { break; } @@ -1779,16 +1822,17 @@ LoadTableEncoding(name, type, chan) break; } for (Tcl_DStringSetLength(&lineString, 0); - (len = Tcl_Gets(chan, &lineString)) >= 0; - Tcl_DStringSetLength(&lineString, 0)) { + (len = Tcl_Gets(chan, &lineString)) >= 0; + Tcl_DStringSetLength(&lineString, 0)) { unsigned char* p; int to, from; + if (len < 5) { continue; } p = (unsigned char*) Tcl_DStringValue(&lineString); to = (staticHex[p[0]] << 12) + (staticHex[p[1]] << 8) - + (staticHex[p[2]] << 4) + staticHex[p[3]]; + + (staticHex[p[2]] << 4) + staticHex[p[3]]; if (to == 0) { continue; } @@ -1810,6 +1854,7 @@ LoadTableEncoding(name, type, chan) encType.freeProc = TableFreeProc; encType.nullSize = (type == ENCODING_DOUBLEBYTE) ? 2 : 1; encType.clientData = (ClientData) dataPtr; + return Tcl_CreateEncoding(&encType); } @@ -1818,16 +1863,16 @@ LoadTableEncoding(name, type, chan) * * LoadEscapeEncoding -- * - * Helper function for LoadEncodingTable(). Loads a state machine - * that converts between Unicode and some other encoding. + * Helper function for LoadEncodingTable(). Loads a state machine that + * converts between Unicode and some other encoding. * - * File contains text data that describes the escape sequences that - * are used to choose an encoding and the associated names for the + * File contains text data that describes the escape sequences that are + * used to choose an encoding and the associated names for the * sub-encodings. * * Results: - * The return value is the new encoding, or NULL if the encoding - * could not be created (because the file contained invalid data). + * The return value is the new encoding, or NULL if the encoding could + * not be created (because the file contained invalid data). * * Side effects: * None. @@ -1856,13 +1901,13 @@ LoadEscapeEncoding(name, chan) CONST char **argv; char *line; Tcl_DString lineString; - + Tcl_DStringInit(&lineString); if (Tcl_Gets(chan, &lineString) < 0) { break; } line = Tcl_DStringValue(&lineString); - if (Tcl_SplitList(NULL, line, &argc, &argv) != TCL_OK) { + if (Tcl_SplitList(NULL, line, &argc, &argv) != TCL_OK) { continue; } if (argc >= 2) { @@ -1884,7 +1929,10 @@ LoadEscapeEncoding(name, chan) strncpy(est.name, argv[0], sizeof(est.name)); est.name[sizeof(est.name) - 1] = '\0'; - /* To avoid infinite recursion in [encoding system iso2022-*]*/ + /* + * To avoid infinite recursion in [encoding system iso2022-*] + */ + Tcl_GetEncoding(NULL, est.name); est.encodingPtr = NULL; @@ -1895,14 +1943,15 @@ LoadEscapeEncoding(name, chan) Tcl_DStringFree(&lineString); } - size = sizeof(EscapeEncodingData) - - sizeof(EscapeSubTable) + Tcl_DStringLength(&escapeData); + size = sizeof(EscapeEncodingData) - sizeof(EscapeSubTable) + + Tcl_DStringLength(&escapeData); dataPtr = (EscapeEncodingData *) ckalloc(size); dataPtr->initLen = strlen(init); strcpy(dataPtr->init, init); dataPtr->finalLen = strlen(final); strcpy(dataPtr->final, final); - dataPtr->numSubTables = Tcl_DStringLength(&escapeData) / sizeof(EscapeSubTable); + dataPtr->numSubTables = + Tcl_DStringLength(&escapeData) / sizeof(EscapeSubTable); memcpy((VOID *) dataPtr->subTables, (VOID *) Tcl_DStringValue(&escapeData), (size_t) Tcl_DStringLength(&escapeData)); Tcl_DStringFree(&escapeData); @@ -1933,9 +1982,9 @@ LoadEscapeEncoding(name, chan) * * BinaryProc -- * - * The default conversion when no other conversion is specified. - * No translation is done; source bytes are copied directly to - * destination bytes. + * The default conversion when no other conversion is specified. No + * translation is done; source bytes are copied directly to destination + * bytes. * * Results: * Returns TCL_OK if conversion was successful. @@ -1953,13 +2002,13 @@ BinaryProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, CONST char *src; /* Source string (unknown encoding). */ int srcLen; /* Source string length in bytes. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ - char *dst; /* Output buffer in which converted string - * is stored. */ + char *dst; /* Output buffer in which converted string is + * stored. */ int dstLen; /* The maximum length of output buffer in * bytes. */ int *srcReadPtr; /* Filled with the number of bytes from the @@ -1992,14 +2041,13 @@ BinaryProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, return result; } - /* *------------------------------------------------------------------------- * * UtfExtToUtfIntProc -- * - * Convert from UTF-8 to UTF-8. While converting null-bytes from - * the Tcl's internal representation (0xc0, 0x80) to the official + * Convert from UTF-8 to UTF-8. While converting null-bytes from the + * Tcl's internal representation (0xc0, 0x80) to the official * representation (0x00). See UtfToUtfProc for details. * * Results: @@ -2010,15 +2058,16 @@ BinaryProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, * *------------------------------------------------------------------------- */ -static int + +static int UtfIntToUtfExtProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, - srcReadPtr, dstWrotePtr, dstCharsPtr) + srcReadPtr, dstWrotePtr, dstCharsPtr) ClientData clientData; /* Not used. */ CONST char *src; /* Source string in UTF-8. */ int srcLen; /* Source string length in bytes. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ @@ -2039,7 +2088,7 @@ UtfIntToUtfExtProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, * output buffer. */ { return UtfToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, - srcReadPtr, dstWrotePtr, dstCharsPtr, 1); + srcReadPtr, dstWrotePtr, dstCharsPtr, 1); } /* @@ -2059,20 +2108,20 @@ UtfIntToUtfExtProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, * *------------------------------------------------------------------------- */ -static int +static int UtfExtToUtfIntProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, - srcReadPtr, dstWrotePtr, dstCharsPtr) + srcReadPtr, dstWrotePtr, dstCharsPtr) ClientData clientData; /* Not used. */ CONST char *src; /* Source string in UTF-8. */ int srcLen; /* Source string length in bytes. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ - char *dst; /* Output buffer in which converted string - * is stored. */ + char *dst; /* Output buffer in which converted string is + * stored. */ int dstLen; /* The maximum length of output buffer in * bytes. */ int *srcReadPtr; /* Filled with the number of bytes from the @@ -2088,7 +2137,7 @@ UtfExtToUtfIntProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, * output buffer. */ { return UtfToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, - srcReadPtr, dstWrotePtr, dstCharsPtr, 0); + srcReadPtr, dstWrotePtr, dstCharsPtr, 0); } /* @@ -2096,9 +2145,9 @@ UtfExtToUtfIntProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, * * UtfToUtfProc -- * - * Convert from UTF-8 to UTF-8. Note that the UTF-8 to UTF-8 - * translation is not a no-op, because it will turn a stream of - * improperly formed UTF-8 into a properly formed stream. + * Convert from UTF-8 to UTF-8. Note that the UTF-8 to UTF-8 translation + * is not a no-op, because it will turn a stream of improperly formed + * UTF-8 into a properly formed stream. * * Results: * Returns TCL_OK if conversion was successful. @@ -2109,37 +2158,36 @@ UtfExtToUtfIntProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, *------------------------------------------------------------------------- */ -static int +static int UtfToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, - srcReadPtr, dstWrotePtr, dstCharsPtr, pureNullMode) + srcReadPtr, dstWrotePtr, dstCharsPtr, pureNullMode) ClientData clientData; /* Not used. */ CONST char *src; /* Source string in UTF-8. */ int srcLen; /* Source string length in bytes. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ - char *dst; /* Output buffer in which converted string - * is stored. */ + char *dst; /* Output buffer in which converted string is + * stored. */ int dstLen; /* The maximum length of output buffer in * bytes. */ int *srcReadPtr; /* Filled with the number of bytes from the - * source string that were converted. This - * may be less than the original source length - * if there was a problem converting some - * source characters. */ + * source string that were converted. This may + * be less than the original source length if + * there was a problem converting some source + * characters. */ int *dstWrotePtr; /* Filled with the number of bytes that were * stored in the output buffer as a result of * the conversion. */ int *dstCharsPtr; /* Filled with the number of characters that * correspond to the bytes stored in the * output buffer. */ - int pureNullMode; /* Convert embedded nulls from - * internal representation to real - * null-bytes or vice versa */ - + int pureNullMode; /* Convert embedded nulls from internal + * representation to real null-bytes or vice + * versa. */ { CONST char *srcStart, *srcEnd, *srcClose; char *dstStart, *dstEnd; @@ -2147,7 +2195,7 @@ UtfToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, Tcl_UniChar ch; result = TCL_OK; - + srcStart = src; srcEnd = src + srcLen; srcClose = srcEnd; @@ -2172,19 +2220,19 @@ UtfToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, result = TCL_CONVERT_NOSPACE; break; } - if (UCHAR(*src) < 0x80 && - !(UCHAR(*src) == 0 && pureNullMode == 0)) { + if (UCHAR(*src) < 0x80 && !(UCHAR(*src) == 0 && pureNullMode == 0)) { /* - * Copy 7bit chatacters, but skip null-bytes when we are - * in input mode, so that they get converted to 0xc080. + * Copy 7bit chatacters, but skip null-bytes when we are in input + * mode, so that they get converted to 0xc080. */ + *dst++ = *src++; - } else if (pureNullMode == 1 && - UCHAR(*src) == 0xc0 && - UCHAR(*(src+1)) == 0x80) { - /* + } else if (pureNullMode == 1 && UCHAR(*src) == 0xc0 && + UCHAR(*(src+1)) == 0x80) { + /* * Convert 0xc080 to real nulls when we are in output mode. */ + *dst++ = 0; src += 2; } else { @@ -2215,20 +2263,20 @@ UtfToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, *------------------------------------------------------------------------- */ -static int +static int UnicodeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr) ClientData clientData; /* Not used. */ CONST char *src; /* Source string in Unicode. */ int srcLen; /* Source string length in bytes. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ - char *dst; /* Output buffer in which converted string - * is stored. */ + char *dst; /* Output buffer in which converted string is + * stored. */ int dstLen; /* The maximum length of output buffer in * bytes. */ int *srcReadPtr; /* Filled with the number of bytes from the @@ -2246,7 +2294,7 @@ UnicodeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, CONST Tcl_UniChar *wSrc, *wSrcStart, *wSrcEnd; char *dstEnd, *dstStart; int result, numChars; - + result = TCL_OK; if ((srcLen % sizeof(Tcl_UniChar)) != 0) { result = TCL_CONVERT_MULTIBYTE; @@ -2267,9 +2315,11 @@ UnicodeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, result = TCL_CONVERT_NOSPACE; break; } + /* * Special case for 1-byte utf chars for speed. */ + if (*wSrc && *wSrc < 0x80) { *dst++ = (char) *wSrc; } else { @@ -2300,20 +2350,21 @@ UnicodeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, *------------------------------------------------------------------------- */ -static int +static int UtfToUnicodeProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr) - ClientData clientData; /* TableEncodingData that specifies encoding. */ + ClientData clientData; /* TableEncodingData that specifies + * encoding. */ CONST char *src; /* Source string in UTF-8. */ int srcLen; /* Source string length in bytes. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ - char *dst; /* Output buffer in which converted string - * is stored. */ + char *dst; /* Output buffer in which converted string is + * stored. */ int dstLen; /* The maximum length of output buffer in * bytes. */ int *srcReadPtr; /* Filled with the number of bytes from the @@ -2331,7 +2382,7 @@ UtfToUnicodeProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, CONST char *srcStart, *srcEnd, *srcClose; Tcl_UniChar *wDst, *wDstStart, *wDstEnd; int result, numChars; - + srcStart = src; srcEnd = src + srcLen; srcClose = srcEnd; @@ -2357,10 +2408,11 @@ UtfToUnicodeProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, if (wDst > wDstEnd) { result = TCL_CONVERT_NOSPACE; break; - } + } src += TclUtfToUniChar(src, wDst); wDst++; } + *srcReadPtr = src - srcStart; *dstWrotePtr = (char *) wDst - (char *) wDstStart; *dstCharsPtr = numChars; @@ -2384,7 +2436,7 @@ UtfToUnicodeProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, *------------------------------------------------------------------------- */ -static int +static int TableToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr) ClientData clientData; /* TableEncodingData that specifies @@ -2392,13 +2444,13 @@ TableToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, CONST char *src; /* Source string in specified encoding. */ int srcLen; /* Source string length in bytes. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ - char *dst; /* Output buffer in which converted string - * is stored. */ + char *dst; /* Output buffer in which converted string is + * stored. */ int dstLen; /* The maximum length of output buffer in * bytes. */ int *srcReadPtr; /* Filled with the number of bytes from the @@ -2420,7 +2472,7 @@ TableToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, unsigned short **toUnicode; unsigned short *pageZero; TableEncodingData *dataPtr; - + srcStart = src; srcEnd = src + srcLen; @@ -2434,10 +2486,10 @@ TableToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, result = TCL_OK; for (numChars = 0; src < srcEnd; numChars++) { - if (dst > dstEnd) { - result = TCL_CONVERT_NOSPACE; - break; - } + if (dst > dstEnd) { + result = TCL_CONVERT_NOSPACE; + break; + } byte = *((unsigned char *) src); if (prefixBytes[byte]) { src++; @@ -2468,8 +2520,9 @@ TableToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, } else { dst += Tcl_UniCharToUtf(ch, dst); } - src++; + src++; } + *srcReadPtr = src - srcStart; *dstWrotePtr = dst - dstStart; *dstCharsPtr = numChars; @@ -2493,7 +2546,7 @@ TableToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, *------------------------------------------------------------------------- */ -static int +static int TableFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr) ClientData clientData; /* TableEncodingData that specifies @@ -2501,13 +2554,13 @@ TableFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, CONST char *src; /* Source string in UTF-8. */ int srcLen; /* Source string length in bytes. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ - char *dst; /* Output buffer in which converted string - * is stored. */ + char *dst; /* Output buffer in which converted string is + * stored. */ int dstLen; /* The maximum length of output buffer in * bytes. */ int *srcReadPtr; /* Filled with the number of bytes from the @@ -2528,13 +2581,13 @@ TableFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, int result, len, word, numChars; TableEncodingData *dataPtr; unsigned short **fromUnicode; - - result = TCL_OK; + + result = TCL_OK; dataPtr = (TableEncodingData *) clientData; prefixBytes = dataPtr->prefixBytes; fromUnicode = dataPtr->fromUnicode; - + srcStart = src; srcEnd = src + srcLen; srcClose = srcEnd; @@ -2559,9 +2612,10 @@ TableFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, #if TCL_UTF_MAX > 3 /* - * This prevents a crash condition. More evaluation is required - * for full support of int Tcl_UniChar. [Bug 1004065] + * This prevents a crash condition. More evaluation is required for + * full support of int Tcl_UniChar. [Bug 1004065] */ + if (ch & 0xffff0000) { word = 0; } else @@ -2573,7 +2627,7 @@ TableFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, result = TCL_CONVERT_UNKNOWN; break; } - word = dataPtr->fallback; + word = dataPtr->fallback; } if (prefixBytes[(word >> 8)] != 0) { if (dst + 1 > dstEnd) { @@ -2590,9 +2644,10 @@ TableFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, } dst[0] = (char) word; dst++; - } + } src += len; } + *srcReadPtr = src - srcStart; *dstWrotePtr = dst - dstStart; *dstCharsPtr = numChars; @@ -2604,8 +2659,8 @@ TableFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, * * TableFreeProc -- * - * This procedure is invoked when an encoding is deleted. It deletes - * the memory used by the TableEncodingData. + * This procedure is invoked when an encoding is deleted. It deletes the + * memory used by the TableEncodingData. * * Results: * None. @@ -2650,7 +2705,7 @@ TableFreeProc(clientData) *------------------------------------------------------------------------- */ -static int +static int EscapeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr) ClientData clientData; /* EscapeEncodingData that specifies @@ -2658,13 +2713,13 @@ EscapeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, CONST char *src; /* Source string in specified encoding. */ int srcLen; /* Source string length in bytes. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ - char *dst; /* Output buffer in which converted string - * is stored. */ + char *dst; /* Output buffer in which converted string is + * stored. */ int dstLen; /* The maximum length of output buffer in * bytes. */ int *srcReadPtr; /* Filled with the number of bytes from the @@ -2710,54 +2765,56 @@ EscapeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, for (numChars = 0; src < srcEnd; ) { int byte, hi, lo, ch; - if (dst > dstEnd) { - result = TCL_CONVERT_NOSPACE; - break; - } + if (dst > dstEnd) { + result = TCL_CONVERT_NOSPACE; + break; + } byte = *((unsigned char *) src); if (prefixBytes[byte]) { unsigned int left, len, longest; int checked, i; EscapeSubTable *subTablePtr; - + /* - * Saw the beginning of an escape sequence. + * Saw the beginning of an escape sequence. */ - + left = srcEnd - src; len = dataPtr->initLen; longest = len; checked = 0; + if (len <= left) { checked++; - if ((len > 0) && - (memcmp(src, dataPtr->init, len) == 0)) { + if ((len > 0) && (memcmp(src, dataPtr->init, len) == 0)) { /* * If we see initialization string, skip it, even if we're - * not at the beginning of the buffer. + * not at the beginning of the buffer. */ - + src += len; continue; } } + len = dataPtr->finalLen; if (len > longest) { longest = len; } + if (len <= left) { checked++; - if ((len > 0) && - (memcmp(src, dataPtr->final, len) == 0)) { + if ((len > 0) && (memcmp(src, dataPtr->final, len) == 0)) { /* * If we see finalization string, skip it, even if we're - * not at the end of the buffer. + * not at the end of the buffer. */ - + src += len; continue; } } + subTablePtr = dataPtr->subTables; for (i = 0; i < dataPtr->numSubTables; i++) { len = subTablePtr->sequenceLen; @@ -2766,7 +2823,7 @@ EscapeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, } if (len <= left) { checked++; - if ((len > 0) && + if ((len > 0) && (memcmp(src, subTablePtr->sequence, len) == 0)) { state = i; encodingPtr = NULL; @@ -2777,6 +2834,7 @@ EscapeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, } subTablePtr++; } + if (subTablePtr == NULL) { /* * A match was found, the escape sequence was consumed, and @@ -2788,8 +2846,8 @@ EscapeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, /* * We have a split-up or unrecognized escape sequence. If we - * checked all the sequences, then it's a syntax error, - * otherwise we need more bytes to determine a match. + * checked all the sequences, then it's a syntax error, otherwise + * we need more bytes to determine a match. */ if ((checked == dataPtr->numSubTables + 2) @@ -2817,6 +2875,7 @@ EscapeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, tablePrefixBytes = tableDataPtr->prefixBytes; tableToUnicode = tableDataPtr->toUnicode; } + if (tablePrefixBytes[byte]) { src++; if (src >= srcEnd) { @@ -2830,6 +2889,7 @@ EscapeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, hi = 0; lo = byte; } + ch = tableToUnicode[hi][lo]; dst += Tcl_UniCharToUtf(ch, dst); src++; @@ -2860,7 +2920,7 @@ EscapeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, *------------------------------------------------------------------------- */ -static int +static int EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr) ClientData clientData; /* EscapeEncodingData that specifies @@ -2868,20 +2928,20 @@ EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, CONST char *src; /* Source string in UTF-8. */ int srcLen; /* Source string length in bytes. */ int flags; /* Conversion control flags. */ - Tcl_EncodingState *statePtr;/* Place for conversion routine to store - * state information used during a piecewise + Tcl_EncodingState *statePtr;/* Place for conversion routine to store state + * information used during a piecewise * conversion. Contents of statePtr are * initialized and/or reset by conversion * routine under control of flags argument. */ - char *dst; /* Output buffer in which converted string - * is stored. */ + char *dst; /* Output buffer in which converted string is + * stored. */ int dstLen; /* The maximum length of output buffer in * bytes. */ int *srcReadPtr; /* Filled with the number of bytes from the - * source string that were converted. This - * may be less than the original source length - * if there was a problem converting some - * source characters. */ + * source string that were converted. This may + * be less than the original source length if + * there was a problem converting some source + * characters. */ int *dstWrotePtr; /* Filled with the number of bytes that were * stored in the output buffer as a result of * the conversion. */ @@ -2897,8 +2957,8 @@ EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, TableEncodingData *tableDataPtr; char *tablePrefixBytes; unsigned short **tableFromUnicode; - - result = TCL_OK; + + result = TCL_OK; dataPtr = (EscapeEncodingData *) clientData; @@ -2924,11 +2984,10 @@ EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, *dstWrotePtr = 0; return TCL_CONVERT_NOSPACE; } - memcpy((VOID *) dst, (VOID *) dataPtr->init, - (size_t) dataPtr->initLen); + memcpy((VOID *)dst, (VOID *)dataPtr->init, (size_t)dataPtr->initLen); dst += dataPtr->initLen; } else { - state = (int) *statePtr; + state = (int) *statePtr; } encodingPtr = GetTableEncoding(dataPtr, state); @@ -2940,7 +2999,7 @@ EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, unsigned int len; int word; Tcl_UniChar ch; - + if ((src > srcClose) && (!Tcl_UtfCharComplete(src, srcEnd - src))) { /* * If there is more string to follow, this will ensure that the @@ -2956,7 +3015,7 @@ EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, if ((word == 0) && (ch != 0)) { int oldState; EscapeSubTable *subTablePtr; - + oldState = state; for (state = 0; state < dataPtr->numSubTables; state++) { encodingPtr = GetTableEncoding(dataPtr, state); @@ -2976,16 +3035,17 @@ EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, encodingPtr = GetTableEncoding(dataPtr, state); tableDataPtr = (TableEncodingData *) encodingPtr->clientData; word = tableDataPtr->fallback; - } - + } + tablePrefixBytes = tableDataPtr->prefixBytes; tableFromUnicode = tableDataPtr->fromUnicode; /* * The state variable has the value of oldState when word is 0. - * In this case, the escape sequense should not be copied to dst + * In this case, the escape sequense should not be copied to dst * because the current character set is not changed. */ + if (state != oldState) { subTablePtr = &dataPtr->subTables[state]; if ((dst + subTablePtr->sequenceLen) > dstEnd) { @@ -2995,6 +3055,7 @@ EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, * variable because this escape sequence must be written * in the next conversion. */ + state = oldState; result = TCL_CONVERT_NOSPACE; break; @@ -3020,7 +3081,7 @@ EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, } dst[0] = (char) word; dst++; - } + } src += len; } @@ -3052,7 +3113,7 @@ EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen, * * EscapeFreeProc -- * - * This procedure is invoked when an EscapeEncodingData encoding is + * This procedure is invoked when an EscapeEncodingData encoding is * deleted. It deletes the memory used by the encoding. * * Results: @@ -3097,9 +3158,9 @@ EscapeFreeProc(clientData) * The return value is the encoding. * * Side effects: - * If the encoding that represents the specified state has not - * already been used by this EscapeEncoding, it will be loaded - * and cached in the dataPtr. + * If the encoding that represents the specified state has not already + * been used by this EscapeEncoding, it will be loaded and cached in the + * dataPtr. * *--------------------------------------------------------------------------- */ @@ -3111,17 +3172,19 @@ GetTableEncoding(dataPtr, state) { EscapeSubTable *subTablePtr; Encoding *encodingPtr; - + subTablePtr = &dataPtr->subTables[state]; encodingPtr = subTablePtr->encodingPtr; + if (encodingPtr == NULL) { encodingPtr = (Encoding *) Tcl_GetEncoding(NULL, subTablePtr->name); - if ((encodingPtr == NULL) + if ((encodingPtr == NULL) || (encodingPtr->toUtfProc != TableToUtfProc)) { Tcl_Panic("EscapeToUtfProc: invalid sub table"); } subTablePtr->encodingPtr = encodingPtr; } + return encodingPtr; } @@ -3130,9 +3193,9 @@ GetTableEncoding(dataPtr, state) * * unilen -- * - * A helper function for the Tcl_ExternalToUtf functions. This - * function is similar to strlen for double-byte characters: it - * returns the number of bytes in a 0x0000 terminated string. + * A helper function for the Tcl_ExternalToUtf functions. This function + * is similar to strlen for double-byte characters: it returns the number + * of bytes in a 0x0000 terminated string. * * Results: * As above. @@ -3161,28 +3224,27 @@ unilen(src) * * InitializeEncodingSearchPath -- * - * This is the fallback routine that sets the default value - * of the encoding search path if the application has not set - * one via a call to TclSetEncodingSearchPath() by the first - * time the search path is needed to load encoding data. + * This is the fallback routine that sets the default value of the + * encoding search path if the application has not set one via a call to + * TclSetEncodingSearchPath() by the first time the search path is needed + * to load encoding data. * - * The default encoding search path is produced by taking each - * directory in the library path, appending a subdirectory - * named "encoding", and if the resulting directory exists, - * adding it to the encoding search path. + * The default encoding search path is produced by taking each directory + * in the library path, appending a subdirectory named "encoding", and if + * the resulting directory exists, adding it to the encoding search path. * * Results: * None. * * Side effects: - * Sets the encoding search path to an initial value. + * Sets the encoding search path to an initial value. * *------------------------------------------------------------------------- */ void InitializeEncodingSearchPath(valuePtr, lengthPtr, encodingPtr) - char **valuePtr; + char **valuePtr; int *lengthPtr; Tcl_Encoding *encodingPtr; { @@ -3196,18 +3258,20 @@ InitializeEncodingSearchPath(valuePtr, lengthPtr, encodingPtr) libPath = TclGetLibraryPath(); Tcl_IncrRefCount(libPath); Tcl_ListObjLength(NULL, libPath, &numDirs); + for (i = 0; i < numDirs; i++) { Tcl_Obj *directory, *path; Tcl_StatBuf stat; Tcl_ListObjIndex(NULL, libPath, i, &directory); - path = Tcl_FSJoinToPath(directory, 1, &encodingObj); + 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_DecrRefCount(path); } + Tcl_DecrRefCount(libPath); Tcl_DecrRefCount(encodingObj); *encodingPtr = libraryPath.encoding; @@ -3215,8 +3279,17 @@ InitializeEncodingSearchPath(valuePtr, lengthPtr, encodingPtr) ((Encoding *)(*encodingPtr))->refCount++; } bytes = Tcl_GetStringFromObj(searchPath, &numBytes); + *lengthPtr = numBytes; *valuePtr = ckalloc((unsigned int) numBytes + 1); memcpy((VOID *) *valuePtr, (VOID *) bytes, (size_t) numBytes + 1); Tcl_DecrRefCount(searchPath); } + +/* + * Local Variables: + * mode: c + * c-basic-offset: 4 + * fill-column: 78 + * End: + */ |