|
|
| version 1.9, 2001/12/28 14:06:51 | version 1.10, 2001/12/28 18:12:30 |
|---|---|
| Line 71 static void element2case(unsigned char f | Line 71 static void element2case(unsigned char f |
| fcc_table[from]=to; fcc_table[to]=from; | fcc_table[from]=to; fcc_table[to]=from; |
| } | } |
| /// @test custom encodings | |
| #ifdef XML | |
| /* | |
| template <class TType> class ENameMapFor2 : public ENameMap { | |
| public: | |
| // ----------------------------------------------------------------------- | |
| // Constructors and Destructor | |
| // ----------------------------------------------------------------------- | |
| ENameMapFor2( | |
| const XMLCh* const encodingName | |
| , const XMLCh* const fromTable | |
| , const XMLTransService::TransRec* const toTable | |
| , const unsigned int toTableSize | |
| ): ENameMap(encodingName), | |
| ffromTable(fromTable), | |
| ftoTable(toTable), | |
| ftoTableSize(toTableSize) {} | |
| // ----------------------------------------------------------------------- | |
| // Implementation of virtual factory method | |
| // ----------------------------------------------------------------------- | |
| virtual XMLTranscoder* makeNew(const unsigned int blockSize) const { | |
| return new TType( | |
| getKey(), | |
| blockSize, | |
| ffromTable, | |
| ftoTable, ftoTableSize); | |
| } | |
| private: | |
| const XMLCh* const ffromTable; | |
| const XMLTransService::TransRec* const ftoTable; | |
| const unsigned int ftoTableSize; | |
| private: | |
| // ----------------------------------------------------------------------- | |
| // Unimplemented constructors and operators | |
| // ----------------------------------------------------------------------- | |
| ENameMapFor2(); | |
| ENameMapFor2(const ENameMapFor2<TType>&); | |
| void operator=(const ENameMapFor2<TType>&); | |
| }; | |
| class XML256TableTranscoder2 : public XML256TableTranscoder { | |
| public : | |
| XML256TableTranscoder2( | |
| const XMLCh* const encodingName | |
| , const unsigned int blockSize | |
| , const XMLCh* const fromTable | |
| , const XMLTransService::TransRec* const toTable | |
| , const unsigned int toTableSize | |
| ) : XML256TableTranscoder(encodingName, blockSize, fromTable, toTable, toTableSize) {} | |
| private: | |
| XML256TableTranscoder2(); | |
| XML256TableTranscoder2(const XML256TableTranscoder2&); | |
| void operator=(const XML256TableTranscoder2&); | |
| }; | |
| */ | |
| #endif | |
| // methods | // methods |
| extern "C" unsigned char pcre_default_tables[]; // pcre/chartables.c | extern "C" unsigned char pcre_default_tables[]; // pcre/chartables.c |
| Charset::Charset(Pool& apool, const String& aname, const String *request_file_spec): Pooled(apool), | Charset::Charset(Pool& apool, const String& aname, const String *request_file_spec): Pooled(apool), |
| fname(aname) { | fname(aname) { |
| const char *name_cstr=fname.cstr(); | char *name_cstr=fname.cstr(); |
| for(char *c=name_cstr; *c; c++) | |
| *c = toupper(*c); | |
| if(request_file_spec) { | if(request_file_spec) { |
| fisUTF8=false; | fisUTF8=false; |
| Line 173 void Charset::loadDefinition(const Strin | Line 115 void Charset::loadDefinition(const Strin |
| cstr2ctypes(pcre_tables,(const unsigned char *)"*+?{^.$|()[", ctype_meta); | cstr2ctypes(pcre_tables,(const unsigned char *)"*+?{^.$|()[", ctype_meta); |
| // charset | // charset |
| memset(fromTable, 0, sizeof(fromTable)); | memset(tables.fromTable, 0, sizeof(tables.fromTable)); |
| toTable=(Charset_TransRec *)calloc(sizeof(Charset_TransRec)*MAX_CHARSET_UNI_CODES); | tables.toTable=(Charset_TransRec *)calloc(sizeof(Charset_TransRec)*MAX_CHARSET_UNI_CODES); |
| toTableSize=0; | tables.toTableSize=0; |
| // strangly vital | // strangly vital |
| toTable[toTableSize].intCh=0; | tables.toTable[tables.toTableSize].intCh=0; |
| toTable[toTableSize].extCh=(XMLByte)0; | tables.toTable[tables.toTableSize].extCh=(XMLByte)0; |
| toTableSize++; | tables.toTableSize++; |
| // loading text | // loading text |
| char *data=file_read_text(pool(), request_file_spec); | char *data=file_read_text(pool(), request_file_spec); |
| Line 210 void Charset::loadDefinition(const Strin | Line 152 void Charset::loadDefinition(const Strin |
| case 7: | case 7: |
| case 8: | case 8: |
| // charset | // charset |
| if(toTableSize>MAX_CHARSET_UNI_CODES) | if(tables.toTableSize>MAX_CHARSET_UNI_CODES) |
| throw Exception(0, 0, | throw Exception(0, 0, |
| &request_file_spec, | &request_file_spec, |
| "charset must contain not more then %d unicode values", MAX_CHARSET_UNI_CODES); | "charset must contain not more then %d unicode values", MAX_CHARSET_UNI_CODES); |
| Line 219 void Charset::loadDefinition(const Strin | Line 161 void Charset::loadDefinition(const Strin |
| if(!unicode && column==7/*unicode1 column*/) | if(!unicode && column==7/*unicode1 column*/) |
| unicode=(XMLCh)c; | unicode=(XMLCh)c; |
| if(unicode) { | if(unicode) { |
| if(!fromTable[c]) | if(!tables.fromTable[c]) |
| fromTable[c]=unicode; | tables.fromTable[c]=unicode; |
| toTable[toTableSize].intCh=unicode; | tables.toTable[tables.toTableSize].intCh=unicode; |
| toTable[toTableSize].extCh=(XMLByte)c; | tables.toTable[tables.toTableSize].extCh=(XMLByte)c; |
| toTableSize++; | tables.toTableSize++; |
| } | } |
| break; | break; |
| } | } |
| Line 234 void Charset::loadDefinition(const Strin | Line 176 void Charset::loadDefinition(const Strin |
| sort_ToTable(); | sort_ToTable(); |
| } | } |
| #ifdef XML | |
| void Charset::addEncoding(const char *name_cstr) { | |
| /* | |
| // addEncoding | |
| XalanDOMString sencoding(name_cstr); | |
| const XMLCh* const auto_encoding_cstr=sencoding.c_str(); | |
| int size=sizeof(XMLCh)*(sencoding.size()+1); | |
| XMLCh* pool_encoding_cstr=(XMLCh*)malloc(size); | |
| memcpy(pool_encoding_cstr, auto_encoding_cstr, size); | |
| XMLString::upperCase(pool_encoding_cstr); | |
| XMLPlatformUtils::fgTransService->addEncoding( | |
| pool_encoding_cstr, | |
| new ENameMapFor2<XML256TableTranscoder2>( | |
| pool_encoding_cstr | |
| , fromTable | |
| , toTable | |
| , toTableSize | |
| )); | |
| */ | |
| } | |
| void Charset::initTranscoder(const String *source, const char *name_cstr) { | |
| transcoder=xmlFindCharEncodingHandler(name_cstr); | |
| if(!transcoder) | |
| throw Exception(0, 0, | |
| source, | |
| "unsupported encoding"); | |
| } | |
| #endif | |
| static int sort_cmp_Trans_rec_intCh(const void *a, const void *b) { | static int sort_cmp_Trans_rec_intCh(const void *a, const void *b) { |
| return | return |
| static_cast<const Charset_TransRec *>(a)->intCh- | static_cast<const Charset_TransRec *>(a)->intCh- |
| Line 272 static int sort_cmp_Trans_rec_intCh(cons | Line 183 static int sort_cmp_Trans_rec_intCh(cons |
| } | } |
| void Charset::sort_ToTable() { | void Charset::sort_ToTable() { |
| _qsort(toTable, toTableSize, sizeof(*toTable), | _qsort(tables.toTable, tables.toTableSize, sizeof(*tables.toTable), |
| sort_cmp_Trans_rec_intCh); | sort_cmp_Trans_rec_intCh); |
| //FILE *f=fopen("c:\\temp\\a", "wb"); | //FILE *f=fopen("c:\\temp\\a", "wb"); |
| //fwrite(toTable, toTableSize, sizeof(*toTable), f); | //fwrite(tables.toTable, tables.toTableSize, sizeof(*tables.toTable), f); |
| //fclose(f); | //fclose(f); |
| } | } |
| XMLByte Charset::xlatOneTo(const XMLCh toXlat) const { | static XMLByte xlatOneTo(const XMLCh toXlat, |
| const Charset::Tables& tables) { | |
| unsigned int lowOfs = 0; | unsigned int lowOfs = 0; |
| unsigned int hiOfs = toTableSize - 1; | unsigned int hiOfs = tables.toTableSize - 1; |
| XMLByte curByte = 0; | XMLByte curByte = 0; |
| do { | do { |
| // Calc the mid point of the low and high offset. | // Calc the mid point of the low and high offset. |
| Line 290 XMLByte Charset::xlatOneTo(const XMLCh t | Line 202 XMLByte Charset::xlatOneTo(const XMLCh t |
| // If our test char is greater than the mid point char, then | // If our test char is greater than the mid point char, then |
| // we move up to the upper half. Else we move to the lower | // we move up to the upper half. Else we move to the lower |
| // half. If its equal, then its our guy. | // half. If its equal, then its our guy. |
| if(toXlat>toTable[midOfs].intCh) | if(toXlat>tables.toTable[midOfs].intCh) |
| lowOfs = midOfs; | lowOfs = midOfs; |
| else if(toXlat<toTable[midOfs].intCh) | else if(toXlat<tables.toTable[midOfs].intCh) |
| hiOfs = midOfs; | hiOfs = midOfs; |
| else | else |
| return toTable[midOfs].extCh; | return tables.toTable[midOfs].extCh; |
| } while(lowOfs+1<hiOfs); | } while(lowOfs+1<hiOfs); |
| return '?'; | return '?'; |
| Line 375 static const XMLByte gFirstByteMark[7] = | Line 287 static const XMLByte gFirstByteMark[7] = |
| 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC | 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC |
| }; | }; |
| /// @todo not so memory-hungry with prescan | static size_t transcodeToUTF8( |
| void Charset::transcodeToUTF8(Pool& pool, | const void *source_body, size_t source_content_length, |
| const void *source_body, size_t source_content_length, | XMLByte *dest_body, size_t dest_buf_length, |
| const void *& adest_body, size_t& adest_content_length) const { | const Charset::Tables& tables) { |
| size_t dest_content_length=0; | |
| XMLByte *dest_body=(XMLByte*)pool.malloc(source_content_length*6/*so that surly enough*/); | |
| const XMLByte* srcPtr=(const XMLByte*)source_body; | const XMLByte* srcPtr=(const XMLByte*)source_body; |
| const XMLByte* srcEnd=(const XMLByte*)source_body+source_content_length; | const XMLByte* srcEnd=(const XMLByte*)source_body+source_content_length; |
| XMLByte* outPtr=dest_body; | XMLByte* outPtr=dest_body; |
| XMLByte* outEnd=dest_body+dest_buf_length; | |
| while(srcPtr<srcEnd) { | while(srcPtr<srcEnd) { |
| uint curVal = fromTable[*srcPtr]; | uint curVal = tables.fromTable[*srcPtr]; |
| if(!curVal) { | if(!curVal) { |
| // use the replacement character | // use the replacement character |
| *outPtr++= '?'; | *outPtr++= '?'; |
| Line 417 void Charset::transcodeToUTF8(Pool& pool | Line 326 void Charset::transcodeToUTF8(Pool& pool |
| continue; | continue; |
| } | } |
| // If we cannot fully get this char into the output buffer, | // If we cannot fully get this char into the output buffer |
| // never | if (outPtr + encodedBytes > outEnd) |
| break; | |
| // We can do it, so update the source index | // We can do it, so update the source index |
| srcPtr++; | srcPtr++; |
| Line 444 void Charset::transcodeToUTF8(Pool& pool | Line 354 void Charset::transcodeToUTF8(Pool& pool |
| outPtr+= encodedBytes; | outPtr+= encodedBytes; |
| } | } |
| // return | // eaten |
| adest_body=dest_body; | return outPtr-dest_body; |
| adest_content_length=outPtr-dest_body; | |
| } | } |
| void Charset::transcodeFromUTF8(Pool& pool, | static size_t transcodeFromUTF8( |
| const void *source_body, size_t source_content_length, | const void *source_body, size_t source_content_length, |
| const void *& adest_body, size_t& adest_content_length) const { | XMLByte* dest_body, size_t dest_buf_length, |
| size_t dest_content_length=0; | const Charset::Tables& tables) { |
| XMLByte *dest_body=(XMLByte*)pool.malloc(source_content_length/*surly enough*/); | |
| const XMLByte* srcPtr=(const XMLByte*)source_body; | const XMLByte* srcPtr=(const XMLByte*)source_body; |
| const XMLByte* srcEnd=(const XMLByte*)source_body+source_content_length; | const XMLByte* srcEnd=(const XMLByte*)source_body+source_content_length; |
| XMLByte* outPtr=dest_body; | XMLByte* outPtr=dest_body; |
| XMLByte* outEnd=dest_body+dest_buf_length; | |
| // We now loop until we either run out of input data | // We now loop until we either run out of input data, or room to store |
| while(srcPtr<srcEnd) { | while ((srcPtr < srcEnd) && (outPtr < outEnd)) { |
| // Get the next leading byte out | // Get the next leading byte out |
| const XMLByte firstByte = *srcPtr; | const XMLByte firstByte = *srcPtr; |
| Line 501 void Charset::transcodeFromUTF8(Pool& po | Line 409 void Charset::transcodeFromUTF8(Pool& po |
| // fail [*encode it as a surrogate pair. If its not valid, use the | // fail [*encode it as a surrogate pair. If its not valid, use the |
| // replacement char.*] | // replacement char.*] |
| if(!(tmpVal & 0xFFFF0000)) | if(!(tmpVal & 0xFFFF0000)) |
| *outPtr++= xlatOneTo(tmpVal); | *outPtr++= xlatOneTo(tmpVal, tables); |
| else | else |
| throw Exception(0, 0, | throw Exception(0, 0, |
| 0, | 0, |
| "transcodeFromUTF8 error: too big tmpVal(0x%08X)", tmpVal); | "transcodeFromUTF8 error: too big tmpVal(0x%08X)", tmpVal); |
| } | } |
| // eaten | |
| return outPtr-dest_body; | |
| } | |
| /// @todo not so memory-hungry with prescan | |
| /// @test buf overflow | |
| void Charset::transcodeToUTF8(Pool& pool, | |
| const void *source_body, size_t source_content_length, | |
| const void *& adest_body, size_t& adest_content_length) const { | |
| size_t dest_buf_length=source_content_length*6/*so that surly enough*/; | |
| XMLByte *dest_body=(XMLByte*)pool.malloc(dest_buf_length); | |
| size_t dest_content_length=::transcodeToUTF8( | |
| source_body, source_content_length, | |
| dest_body, dest_buf_length, | |
| tables); | |
| /*if(dest_content_length<dest_buf_length) | |
| throw(0, 0, | |
| 0, | |
| "Charset::transcodeToUTF8 error");*/ | |
| // return | |
| adest_body=dest_body; | |
| adest_content_length=dest_content_length; | |
| } | |
| /// @test buf overflow | |
| void Charset::transcodeFromUTF8(Pool& pool, | |
| const void *source_body, size_t source_content_length, | |
| const void *& adest_body, size_t& adest_content_length) const { | |
| size_t dest_buf_length=source_content_length/*surly enough*/; | |
| XMLByte *dest_body=(XMLByte*)pool.malloc(dest_buf_length); | |
| size_t dest_content_length=::transcodeFromUTF8( | |
| source_body, source_content_length, | |
| dest_body, dest_buf_length, | |
| tables); | |
| /* if(dest_content_length!=dest_buf_length) | |
| throw(0, 0, | |
| 0, | |
| "Charset::transcodeToUTF8 error");*/ | |
| // return | // return |
| adest_body=dest_body; | adest_body=dest_body; |
| adest_content_length=outPtr-dest_body; | adest_content_length=dest_content_length; |
| } | } |
| /// transcode using both charsets | /// transcode using both charsets |
| Line 529 void Charset::transcodeToCharset(Pool& p | Line 479 void Charset::transcodeToCharset(Pool& p |
| const XMLByte* srcEnd=(const XMLByte*)source_body+source_content_length; | const XMLByte* srcEnd=(const XMLByte*)source_body+source_content_length; |
| for(XMLByte* outPtr=dest_body; srcPtr<srcEnd; srcPtr++) { | for(XMLByte* outPtr=dest_body; srcPtr<srcEnd; srcPtr++) { |
| XMLCh curVal = fromTable[*srcPtr]; | XMLCh curVal = tables.fromTable[*srcPtr]; |
| if(curVal) | if(curVal) |
| *outPtr++=dest_charset.xlatOneTo(curVal); | *outPtr++=xlatOneTo(curVal, dest_charset.tables); |
| else { | else { |
| // use the replacement character | // use the replacement character |
| *outPtr++= '?'; | *outPtr++= '?'; |
| Line 544 void Charset::transcodeToCharset(Pool& p | Line 494 void Charset::transcodeToCharset(Pool& p |
| } | } |
| #ifdef XML | #ifdef XML |
| static int xml256CharEncodingInputFunc ( | |
| unsigned char *out, | |
| int *outlen, | |
| const unsigned char *in, | |
| int *inlen, | |
| void *info) { | |
| return transcodeToUTF8( | |
| in, *inlen, | |
| out, *(unsigned int*)outlen, | |
| *(const Charset::Tables *)info); | |
| } | |
| static int xml256CharEncodingOutputFunc ( | |
| unsigned char *out, | |
| int *outlen, | |
| const unsigned char *in, | |
| int *inlen, | |
| void *info) { | |
| return transcodeFromUTF8( | |
| in, *inlen, | |
| out, *outlen, | |
| *(const Charset::Tables *)info); | |
| } | |
| void Charset::addEncoding(char *name_cstr) { | |
| xmlCharEncodingHandler *handler= | |
| (xmlCharEncodingHandler *)malloc(sizeof(xmlCharEncodingHandler)); | |
| handler->name=name_cstr; | |
| handler->input=xml256CharEncodingInputFunc; handler->inputInfo=&tables; | |
| handler->output=xml256CharEncodingOutputFunc; handler->outputInfo=&tables; | |
| xmlRegisterCharEncodingHandler(handler); | |
| } | |
| void Charset::initTranscoder(const String *source, const char *name_cstr) { | |
| transcoder=xmlFindCharEncodingHandler(name_cstr); | |
| if(!transcoder) | |
| throw Exception(0, 0, | |
| source, | |
| "unsupported encoding"); | |
| } | |
| const char *Charset::transcode_cstr(GdomeDOMString *s) { | const char *Charset::transcode_cstr(GdomeDOMString *s) { |
| if(!transcoder) | if(!transcoder) |
| throw Exception(0, 0, | throw Exception(0, 0, |
| Line 556 const char *Charset::transcode_cstr(Gdom | Line 549 const char *Charset::transcode_cstr(Gdom |
| int size=transcoder->output( | int size=transcoder->output( |
| (unsigned char*)out, &outlen, | (unsigned char*)out, &outlen, |
| (const unsigned char*)s->str, &inlen); | (const unsigned char*)s->str, &inlen, |
| transcoder->outputInfo); | |
| if(size<0) | if(size<0) |
| throw Exception(0, 0, | throw Exception(0, 0, |
| 0, | 0, |
| Line 580 GdomeDOMString *Charset::transcode_buf(c | Line 574 GdomeDOMString *Charset::transcode_buf(c |
| unsigned char *out=(unsigned char*)malloc(outlen*sizeof(unsigned char)); | unsigned char *out=(unsigned char*)malloc(outlen*sizeof(unsigned char)); |
| int size=transcoder->input( | int size=transcoder->input( |
| out, &outlen, | out, &outlen, |
| (const unsigned char *)buf, (int *)&buf_size); | (const unsigned char *)buf, (int *)&buf_size, |
| transcoder->inputInfo); | |
| if(size<0) | if(size<0) |
| throw Exception(0, 0, | throw Exception(0, 0, |
| 0, | 0, |