|
|
| version 1.9, 2001/12/28 14:06:51 | version 1.33.2.16, 2003/03/07 11:02:11 |
|---|---|
| Line 1 | Line 1 |
| /** @file | /** @file |
| Parser: Charset connection implementation. | Parser: Charset connection implementation. |
| Copyright(c) 2001 ArtLebedev Group(http://www.artlebedev.com) | Copyright(c) 2001-2003 ArtLebedev Group (http://www.artlebedev.com) |
| Author: Alexander Petrosyan<paf@design.ru>(http://paf.design.ru) | Author: Alexander Petrosyan<paf@design.ru>(http://paf.design.ru) |
| $Id$ | |
| */ | */ |
| static const char* IDENT_CHARSET_C="$Date$"; | |
| #include "pa_charset.h" | #include "pa_charset.h" |
| #include "pa_charsets.h" | |
| #ifdef XML | #ifdef XML |
| #include "libxml/encoding.h" | #include "libxml/encoding.h" |
| #endif | #endif |
| // globals | |
| // consts | |
| #define MAX_CHARSET_UNI_CODES 500 | |
| // helpers | // helpers |
| inline void prepare_case_tables(unsigned char *tables) { | inline void prepare_case_tables(unsigned char *tables) { |
| Line 37 inline void cstr2ctypes(unsigned char *t | Line 31 inline void cstr2ctypes(unsigned char *t |
| ctypes_table[c]|=bit; | ctypes_table[c]|=bit; |
| } | } |
| } | } |
| inline unsigned int to_wchar_code(const char *cstr) { | inline unsigned int to_wchar_code(const char* cstr) { |
| if(!cstr || !*cstr) | if(!cstr || !*cstr) |
| return 0; | return 0; |
| if(cstr[1]==0) | if(cstr[1]==0) |
| Line 46 inline unsigned int to_wchar_code(const | Line 40 inline unsigned int to_wchar_code(const |
| char *error_pos; | char *error_pos; |
| return(unsigned int)strtol(cstr, &error_pos, 0); | return(unsigned int)strtol(cstr, &error_pos, 0); |
| } | } |
| inline bool to_bool(const char *cstr) { | inline bool to_bool(const char* cstr) { |
| return cstr && *cstr!=0; | return cstr && *cstr!=0; |
| } | } |
| static void element2ctypes(unsigned char c, bool belongs, | static void element2ctypes(unsigned char c, bool belongs, |
| Line 71 static void element2case(unsigned char f | Line 65 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(StringPtr aname, StringPtr afile_spec): |
| fname(aname) { | fname(aname), |
| fname_cstr(aname->cstr()) { | |
| const char *name_cstr=fname.cstr(); | for(char *c=fname_cstr; *c; c++) |
| *c = toupper(*c); | |
| if(request_file_spec) { | if(afile_spec) { |
| fisUTF8=false; | fisUTF8=false; |
| loadDefinition(*request_file_spec); | load_definition(afile_spec); |
| #ifdef XML | #ifdef XML |
| addEncoding(name_cstr); | addEncoding(fname_cstr); |
| #endif | #endif |
| } else { | } else { |
| fisUTF8=true; | fisUTF8=true; |
| Line 152 Charset::Charset(Pool& apool, const Stri | Line 88 Charset::Charset(Pool& apool, const Stri |
| } | } |
| #ifdef XML | #ifdef XML |
| initTranscoder(&aname, name_cstr); | initTranscoder(fname, fname_cstr); |
| #endif | #endif |
| } | } |
| Charset::~Charset() { | Charset::~Charset() { |
| // @todonow unregister encodings | |
| #ifdef XML | #ifdef XML |
| // not deleting transcoder, that's not our business | // not deleting transcoder, that's not our business |
| #endif | #endif |
| } | } |
| void Charset::loadDefinition(const String& request_file_spec) { | void Charset::load_definition(StringPtr afile_spec) { |
| // pcre_tables | // pcre_tables |
| // lowcase, flipcase, bits digit+word+whitespace, masks | // lowcase, flipcase, bits digit+word+whitespace, masks |
| Line 173 void Charset::loadDefinition(const Strin | Line 110 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, 0, sizeof(tables)); |
| toTable=(Charset_TransRec *)calloc(sizeof(Charset_TransRec)*MAX_CHARSET_UNI_CODES); | |
| 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); | Pool pool_for_load_only; |
| char *data=file_read_text(pool_for_load_only, *UTF8_charset, afile_spec); | |
| // ignore header | // ignore header |
| getrow(&data); | getrow(&data); |
| Line 210 void Charset::loadDefinition(const Strin | Line 146 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("parser.runtime", |
| &request_file_spec, | afile_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); |
| XMLCh unicode=(XMLCh)to_wchar_code(cell); | XMLCh unicode=(XMLCh)to_wchar_code(cell); |
| 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 170 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 177 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, | |
| XMLByte not_found) { | |
| 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 197 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 not_found; |
| } | } |
| void Charset::transcode(Pool& pool, | void Charset::transcode(Pool& pool, |
| const Charset& source_charset, const void *source_body, size_t source_content_length, | const Charset& source_charset, const void* source_body, size_t source_content_length, |
| const Charset& dest_charset, const void *& dest_body, size_t& dest_content_length | const Charset& dest_charset, const void *& dest_body, size_t& dest_content_length |
| ) { | ) { |
| if(!source_content_length) { | if(!source_content_length) { |
| Line 375 static const XMLByte gFirstByteMark[7] = | Line 282 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 int transcodeToUTF8( |
| void Charset::transcodeToUTF8(Pool& pool, | const XMLByte* srcData, size_t& srcLen, |
| const void *source_body, size_t source_content_length, | XMLByte *toFill, size_t& toFillLen, |
| const void *& adest_body, size_t& adest_content_length) const { | const Charset::Tables& tables) { |
| const XMLByte* srcPtr=srcData; | |
| size_t dest_content_length=0; | const XMLByte* srcEnd=srcData+srcLen; |
| XMLByte *dest_body=(XMLByte*)pool.malloc(source_content_length*6/*so that surly enough*/); | XMLByte* outPtr=toFill; |
| XMLByte* outEnd=toFill+toFillLen; | |
| const XMLByte* srcPtr=(const XMLByte*)source_body; | |
| const XMLByte* srcEnd=(const XMLByte*)source_body+source_content_length; | |
| XMLByte* outPtr=dest_body; | |
| 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 321 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 349 void Charset::transcodeToUTF8(Pool& pool |
| outPtr+= encodedBytes; | outPtr+= encodedBytes; |
| } | } |
| // return | // Update the bytes eaten |
| adest_body=dest_body; | srcLen = srcPtr - srcData; |
| adest_content_length=outPtr-dest_body; | |
| // Return the characters read | |
| toFillLen = outPtr - toFill; | |
| //return srcPtr==srcEnd?(int)toFillLen:-1; | |
| /* | |
| xmlCharEncodingInputFunc | |
| Returns : | |
| the number of byte written, or -1 by lack of space, or -2 if the transcoding failed. The value of inlen after return is the | |
| number of octets consumed as the return value is positive, else unpredictiable. The value of outlen after return is the number | |
| of ocetes consumed. | |
| */ | |
| return 0; | |
| } | } |
| void Charset::transcodeFromUTF8(Pool& pool, | /// @todo digital entites only when xml/html output [at output in html/xml mode, in html part of a letter] |
| const void *source_body, size_t source_content_length, | static int transcodeFromUTF8( |
| const void *& adest_body, size_t& adest_content_length) const { | const XMLByte* srcData, size_t& srcLen, |
| size_t dest_content_length=0; | XMLByte* toFill, size_t& toFillLen, |
| XMLByte *dest_body=(XMLByte*)pool.malloc(source_content_length/*surly enough*/); | const Charset::Tables& tables) { |
| const XMLByte* srcPtr=srcData; | |
| const XMLByte* srcPtr=(const XMLByte*)source_body; | const XMLByte* srcEnd=srcData+srcLen; |
| const XMLByte* srcEnd=(const XMLByte*)source_body+source_content_length; | XMLByte* outPtr=toFill; |
| XMLByte* outPtr=dest_body; | XMLByte* outEnd=toFill+toFillLen; |
| // 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; |
| // Special-case ASCII, which is a leading byte value of<= 127 | // Special-case ASCII, which is a leading byte value of<= 127 |
| if(firstByte<= 127) { | if(firstByte<= 127) { |
| Line 491 void Charset::transcodeFromUTF8(Pool& po | Line 408 void Charset::transcodeFromUTF8(Pool& po |
| break; | break; |
| default: | default: |
| throw Exception(0, 0, | throw Exception(0, |
| 0, | Exception::undefined_source, |
| "transcodeFromUTF8 error: wrong trailingBytes value(%d)", trailingBytes); | "transcodeFromUTF8 error: wrong trailingBytes value(%d)", trailingBytes); |
| } | } |
| tmpVal-=gUTFOffsets[trailingBytes]; | tmpVal-=gUTFOffsets[trailingBytes]; |
| Line 500 void Charset::transcodeFromUTF8(Pool& po | Line 417 void Charset::transcodeFromUTF8(Pool& po |
| // If it will fit into a single char, then put it in. Otherwise | // If it will fit into a single char, then put it in. Otherwise |
| // 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); | if(XMLByte xlat=xlatOneTo(tmpVal, tables, 0)) |
| else | *outPtr++=xlat; |
| throw Exception(0, 0, | else |
| 0, | outPtr+=sprintf((char *)outPtr, "&#%d;", tmpVal); // &#decimal; |
| } else | |
| throw Exception(0, | |
| Exception::undefined_source, | |
| "transcodeFromUTF8 error: too big tmpVal(0x%08X)", tmpVal); | "transcodeFromUTF8 error: too big tmpVal(0x%08X)", tmpVal); |
| } | } |
| // Update the bytes eaten | |
| srcLen = srcPtr - srcData; | |
| // Return the characters read | |
| toFillLen = outPtr - toFill; | |
| //return srcPtr==srcEnd?(int)toFillLen:-1; | |
| /* | |
| xmlCharEncodingOutputFunc | |
| Returns : | |
| the number of byte written, or -1 by lack of space, or -2 if the transcoding failed. The value of inlen after return is the | |
| number of octets consumed as the return value is positive, else unpredictiable. The value of outlen after return is the number | |
| of ocetes consumed. | |
| */ | |
| return 0; | |
| } | |
| /// @todo not so memory-hungry with prescan | |
| void Charset::transcodeToUTF8(Pool& pool, | |
| const void* source_body, size_t source_content_length, | |
| const void *& adest_body, size_t& dest_content_length) const { | |
| dest_content_length=source_content_length*6/*so that surly enough, max utf8 seq len=6*/; | |
| XMLByte *dest_body=new(pool) XMLByte[dest_content_length]; | |
| if(::transcodeToUTF8( | |
| (XMLByte *)source_body, source_content_length, | |
| dest_body, dest_content_length, | |
| tables)<0) | |
| throw(0, 0, | |
| 0, | |
| "Charset::transcodeToUTF8 buffer overflow"); | |
| // return | |
| adest_body=dest_body; | |
| } | |
| void Charset::transcodeFromUTF8(Pool& pool, | |
| const void* source_body, size_t source_content_length, | |
| const void *& adest_body, size_t& dest_content_length) const { | |
| dest_content_length=source_content_length*6/*so that surly enough, "ÿ" has max ratio */; | |
| XMLByte *dest_body=new(pool) XMLByte[dest_content_length]; | |
| if(::transcodeFromUTF8( | |
| (XMLByte *)source_body, source_content_length, | |
| dest_body, dest_content_length, | |
| tables)<0) | |
| throw(0, 0, | |
| 0, | |
| "Charset::transcodeToUTF8 buffer overflow"); | |
| // return | // return |
| adest_body=dest_body; | adest_body=dest_body; |
| adest_content_length=outPtr-dest_body; | |
| } | } |
| /// transcode using both charsets | /// transcode using both charsets |
| void Charset::transcodeToCharset(Pool& pool, | void Charset::transcodeToCharset(Pool& pool, |
| const Charset& dest_charset, | const Charset& dest_charset, |
| 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 { | const void *& adest_body, size_t& adest_content_length) const { |
| if(&dest_charset==this) { | if(&dest_charset==this) { |
| adest_body=source_body; | adest_body=source_body; |
| adest_content_length=source_content_length; | adest_content_length=source_content_length; |
| } else { | } else { |
| size_t dest_content_length=source_content_length; | size_t dest_content_length=source_content_length; |
| unsigned char *dest_body=(unsigned char *)pool.malloc(dest_content_length); | unsigned char *dest_body=new(pool) unsigned char[dest_content_length]; |
| const XMLByte* srcPtr=(const XMLByte*)source_body; | const XMLByte* srcPtr=(XMLByte *)source_body; |
| const XMLByte* srcEnd=(const XMLByte*)source_body+source_content_length; | const XMLByte* srcEnd=(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 512 void Charset::transcodeToCharset(Pool& p |
| } | } |
| #ifdef XML | #ifdef XML |
| const char *Charset::transcode_cstr(GdomeDOMString *s) { | static int xml256CharEncodingInputFunc ( |
| if(!transcoder) | unsigned char *out, |
| throw Exception(0, 0, | int *outlen, |
| 0, | const unsigned char *in, |
| "transcode_cstr no transcoder"); | int *inlen, |
| void *info) { | |
| return transcodeToUTF8( | |
| in, *(size_t*)inlen, | |
| out, *(size_t*)outlen, | |
| *(const Charset::Tables *)info); | |
| } | |
| static int xml256CharEncodingOutputFunc ( | |
| unsigned char *out, | |
| int *outlen, | |
| const unsigned char *in, | |
| int *inlen, | |
| void *info) { | |
| return transcodeFromUTF8( | |
| in, *(size_t*)inlen, | |
| out, *(size_t*)outlen, | |
| *(const Charset::Tables *)info); | |
| } | |
| void Charset::addEncoding(char *name_cstr) { | |
| xmlCharEncodingHandler* handler=new xmlCharEncodingHandler; | |
| fcreated_handler=xmlCharEncodingHandlerPtr(fcreated_handler); | |
| handler->name=name_cstr; | |
| handler->input=xml256CharEncodingInputFunc; handler->inputInfo=&tables; | |
| handler->output=xml256CharEncodingOutputFunc; handler->outputInfo=&tables; | |
| xmlRegisterCharEncodingHandler(handler); | |
| } | |
| void Charset::initTranscoder(StringPtr source, const char* name_cstr) { | |
| ftranscoder=xmlFindCharEncodingHandler(name_cstr); | |
| transcoder(source); // check right way | |
| } | |
| int inlen=gdome_str_length(s); | xmlCharEncodingHandler& Charset::transcoder(StringPtr source) { |
| if(!ftranscoder) | |
| throw Exception("parser.runtime", | |
| source, | |
| "unsupported encoding"); | |
| return *ftranscoder; | |
| } | |
| CharPtr Charset::transcode_cstr(xmlChar* s) { | |
| if(!s) | |
| return ""; | |
| int inlen=strlen((const char* )s); | |
| int outlen=inlen+1; // max | int outlen=inlen+1; // max |
| char *out=(char *)malloc(outlen*sizeof(char)); | char *out=new char[outlen]; |
| int size=transcoder->output( | int error; |
| (unsigned char*)out, &outlen, | if(xmlCharEncodingOutputFunc output=transcoder(Exception::undefined_source).output) { |
| (const unsigned char*)s->str, &inlen); | error=output( |
| if(size<0) | (unsigned char*)out, &outlen, |
| throw Exception(0, 0, | (const unsigned char*)s, &inlen, |
| 0, | transcoder(Exception::undefined_source).outputInfo); |
| "transcode_cstr failed (%d)", size); | } else { |
| memcpy(out, s, outlen=inlen); | |
| error=0; | |
| } | |
| if(error<0) | |
| throw Exception(Exception::undefined_type, | |
| Exception::undefined_source, | |
| "transcode_cstr failed (%d)", error); | |
| out[size]=0; | out[outlen/*surely would be less then on input*/]=0; |
| return out; | return out; |
| } | } |
| String& Charset::transcode(GdomeDOMString *s) { | StringPtr Charset::transcode(xmlChar* s |
| return *NEW String(pool(), transcode_cstr(s)); | #ifndef NO_STRING_ORIGIN |
| , StringPtr origin | |
| #endif | |
| ) { | |
| StringPtr result(new String()); | |
| result->APPEND_CLEAN(transcode_cstr(s), 0/*auto-size*/, origin->origin().file, origin->origin().line); | |
| return result; | |
| } | |
| const char* Charset::transcode_cstr(GdomeDOMString* s) { | |
| return s?transcode_cstr(BAD_CAST s->str):""; | |
| } | |
| StringPtr Charset::transcode(GdomeDOMString* s | |
| #ifndef NO_STRING_ORIGIN | |
| , StringPtr origin | |
| #endif | |
| ) { | |
| StringPtr result(new String()); | |
| result->APPEND_CLEAN(transcode_cstr(s), 0/*auto-size*/, origin->origin().file, origin->origin().line); | |
| return result; | |
| } | } |
| /// @test less memory using -maybe- xmlParserInputBufferCreateMem | /// @test less memory using -maybe- xmlParserInputBufferCreateMem |
| GdomeDOMString *Charset::transcode_buf(const char *buf, size_t buf_size) { | void* Charset::transcode_buf2mchar(transcode_buf_malloc_func malloc_func, |
| if(!transcoder) | const char* buf, size_t buf_size) { |
| throw Exception(0, 0, | unsigned char* out; |
| 0, | int outlen; |
| "transcode_buf no transcoder"); | int error; |
| if(xmlCharEncodingInputFunc input=transcoder(Exception::undefined_source).input) { | |
| outlen=buf_size*6/*max*/; | |
| out=(unsigned char*)malloc_func(outlen+1); | |
| error=input( | |
| out, &outlen, | |
| (const unsigned char *)buf, (int *)&buf_size, | |
| transcoder(Exception::undefined_source).inputInfo); | |
| } else { | |
| outlen=buf_size; | |
| out=(unsigned char*)malloc_func(outlen+1); | |
| memcpy(out, buf, outlen); | |
| error=0; | |
| } | |
| if(error<0) | |
| throw Exception(Exception::undefined_type, | |
| Exception::undefined_source, | |
| "transcode_buf failed (%d)", error); | |
| int outlen=buf_size*6/*max*/+1; | out[outlen/*surely would be less then on input*/]=0; |
| unsigned char *out=(unsigned char*)malloc(outlen*sizeof(unsigned char)); | return out; |
| int size=transcoder->input( | } |
| out, &outlen, | |
| (const unsigned char *)buf, (int *)&buf_size); | |
| if(size<0) | |
| throw Exception(0, 0, | |
| 0, | |
| "transcode_buf failed (%d)", size); | |
| out[size]=0; | xmlChar* Charset::transcode_buf2xchar(const char* buf, size_t buf_size) { |
| return gdome_str_mkref_own((gchar*)out); | return static_cast<xmlChar*>(transcode_buf2mchar(xmlMalloc, buf, buf_size)); |
| } | |
| static void* g_malloc_wrapper(size_t size) { | |
| return g_malloc(size); | |
| } | |
| gchar* Charset::transcode_buf2gchar(const char* buf, size_t buf_size) { | |
| return static_cast<gchar*>(transcode_buf2mchar(g_malloc_wrapper, buf, buf_size)); | |
| } | |
| GdomeDOMString_auto_ptr Charset::transcode_buf2dom(const char* buf, size_t buf_size) { | |
| return GdomeDOMString_auto_ptr(transcode_buf2gchar(buf, buf_size)); | |
| } | } |
| GdomeDOMString *Charset::transcode(const String& s) { | GdomeDOMString_auto_ptr Charset::transcode(StringPtr s) { |
| const char *cstr=s.cstr(String::UL_UNSPECIFIED); | CharPtr cstr=s->cstr(String::UL_UNSPECIFIED); |
| return transcode_buf(cstr, strlen(cstr)); | return transcode_buf2dom(cstr, strlen(cstr)); |
| } | } |
| #endif | #endif |