--- parser3/src/main/pa_charset.C 2002/10/15 15:12:57 1.31 +++ parser3/src/main/pa_charset.C 2003/01/31 12:34:35 1.33.2.6 @@ -1,13 +1,14 @@ /** @file Parser: Charset connection implementation. - Copyright(c) 2001, 2002 ArtLebedev Group (http://www.artlebedev.com) + Copyright(c) 2001-2003 ArtLebedev Group (http://www.artlebedev.com) Author: Alexander Petrosyan(http://paf.design.ru) */ -static const char* IDENT_CHARSET_C="$Date: 2002/10/15 15:12:57 $"; +static const char* IDENT_CHARSET_C="$Date: 2003/01/31 12:34:35 $"; #include "pa_charset.h" +#include "pa_value_includes.h" #ifdef XML #include "libxml/encoding.h" @@ -15,6 +16,10 @@ static const char* IDENT_CHARSET_C="$Dat // globals +#define CHARSET_UTF8_NAME "UTF-8" + +CharsetPtr UTF8_charset(new Charset(StringPtr(new String(CHARSET_UTF8_NAME)), + StringPtr(0)/*no file=system*/)); // consts @@ -37,7 +42,7 @@ inline void cstr2ctypes(unsigned char *t 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) return 0; if(cstr[1]==0) @@ -46,7 +51,7 @@ inline unsigned int to_wchar_code(const char *error_pos; 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; } static void element2ctypes(unsigned char c, bool belongs, @@ -74,18 +79,18 @@ static void element2case(unsigned char f // methods extern "C" unsigned char pcre_default_tables[]; // pcre/chartables.c -Charset::Charset(Pool& apool, const String& aname, const String *request_file_spec): Pooled(apool), - fname(aname) { +Charset::Charset(StringPtr aname, StringPtr afile_spec): + fname(aname), + fname_cstr(aname->cstr()) { - char *name_cstr=fname.cstr(); - for(char *c=name_cstr; *c; c++) + for(char *c=fname_cstr; *c; c++) *c = toupper(*c); - if(request_file_spec) { + if(afile_spec) { fisUTF8=false; - loadDefinition(*request_file_spec); + load_definition(fname); #ifdef XML - addEncoding(name_cstr); + addEncoding(fname_cstr); #endif } else { fisUTF8=true; @@ -94,17 +99,18 @@ Charset::Charset(Pool& apool, const Stri } #ifdef XML - initTranscoder(&aname, name_cstr); + initTranscoder(fname, fname_cstr); #endif } Charset::~Charset() { + // @todonow unregister encodings #ifdef XML // not deleting transcoder, that's not our business #endif } -void Charset::loadDefinition(const String& request_file_spec) { +void Charset::load_definition(StringPtr afile_spec) { // pcre_tables // lowcase, flipcase, bits digit+word+whitespace, masks @@ -116,7 +122,7 @@ void Charset::loadDefinition(const Strin // charset memset(tables.fromTable, 0, sizeof(tables.fromTable)); - tables.toTable=(Charset_TransRec *)calloc(sizeof(Charset_TransRec)*MAX_CHARSET_UNI_CODES); + tables.toTable=(Charset_TransRec *)pool_for_load.calloc(sizeof(Charset_TransRec)*MAX_CHARSET_UNI_CODES); tables.toTableSize=0; // strangly vital tables.toTable[tables.toTableSize].intCh=0; @@ -124,7 +130,7 @@ void Charset::loadDefinition(const Strin tables.toTableSize++; // loading text - char *data=file_read_text(pool(), request_file_spec); + char *data=file_read_text(pool_for_load, *UTF8_charset, afile_spec); // ignore header getrow(&data); @@ -154,7 +160,7 @@ void Charset::loadDefinition(const Strin // charset if(tables.toTableSize>MAX_CHARSET_UNI_CODES) throw Exception("parser.runtime", - &request_file_spec, + afile_spec, "charset must contain not more then %d unicode values", MAX_CHARSET_UNI_CODES); XMLCh unicode=(XMLCh)to_wchar_code(cell); @@ -415,7 +421,7 @@ static int transcodeFromUTF8( default: throw Exception(0, - 0, + Exception::undefined_source, "transcodeFromUTF8 error: wrong trailingBytes value(%d)", trailingBytes); } tmpVal-=gUTFOffsets[trailingBytes]; @@ -430,7 +436,7 @@ static int transcodeFromUTF8( outPtr+=sprintf((char *)outPtr, "&#%d;", tmpVal); // &#decimal; } else throw Exception(0, - 0, + Exception::undefined_source, "transcodeFromUTF8 error: too big tmpVal(0x%08X)", tmpVal); } @@ -553,7 +559,7 @@ void Charset::addEncoding(char *name_cst xmlRegisterCharEncodingHandler(handler); } -void Charset::initTranscoder(const String *source, const char *name_cstr) { +void Charset::initTranscoder(const String *source, const char* name_cstr) { ftranscoder=xmlFindCharEncodingHandler(name_cstr); transcoder(source); // check right way } @@ -566,11 +572,11 @@ xmlCharEncodingHandler *Charset::transco return ftranscoder; } -const char *Charset::transcode_cstr(xmlChar *s) { +const char* Charset::transcode_cstr(xmlChar *s) { if(!s) return ""; - int inlen=strlen((const char *)s); + int inlen=strlen((const char* )s); int outlen=inlen+1; // max char *out=(char *)malloc(outlen*sizeof(char)); @@ -601,7 +607,7 @@ String& Charset::transcode(xmlChar *s result.APPEND_CLEAN(transcode_cstr(s), 0/*auto-size*/, origin->origin().file, origin->origin().line); return result; } -const char *Charset::transcode_cstr(GdomeDOMString *s) { +const char* Charset::transcode_cstr(GdomeDOMString *s) { return s?transcode_cstr(BAD_CAST s->str):""; } String& Charset::transcode(GdomeDOMString *s @@ -615,20 +621,20 @@ String& Charset::transcode(GdomeDOMStrin } /// @test less memory using -maybe- xmlParserInputBufferCreateMem -xmlChar *Charset::transcode_buf2xchar(const char *buf, size_t buf_size) { +xmlChar *Charset::transcode_buf2xchar(const char* buf, size_t buf_size) { unsigned char *out; int outlen; int error; if(xmlCharEncodingInputFunc input=transcoder(0)->input) { - outlen=buf_size*6/*max*/+1; - out=(unsigned char*)malloc(outlen*sizeof(unsigned char)); + outlen=buf_size*6/*max*/; + out=(unsigned char*)malloc((outlen+1)*sizeof(unsigned char)); error=input( out, &outlen, (const unsigned char *)buf, (int *)&buf_size, transcoder(0)->inputInfo); } else { outlen=buf_size; - out=(unsigned char*)malloc(outlen*sizeof(unsigned char)); + out=(unsigned char*)malloc((outlen+1)*sizeof(unsigned char)); memcpy(out, buf, outlen); error=0; } @@ -641,11 +647,11 @@ xmlChar *Charset::transcode_buf2xchar(co out[outlen/*surely would be less then on input*/]=0; return (xmlChar *)out; } -GdomeDOMString_auto_ptr Charset::transcode_buf2dom(const char *buf, size_t buf_size) { +GdomeDOMString_auto_ptr Charset::transcode_buf2dom(const char* buf, size_t buf_size) { return GdomeDOMString_auto_ptr((gchar*)transcode_buf2xchar(buf, buf_size)); } GdomeDOMString_auto_ptr Charset::transcode(const String& s) { - const char *cstr=s.cstr(String::UL_UNSPECIFIED); + const char* cstr=s.cstr(String::UL_UNSPECIFIED); return transcode_buf2dom(cstr, strlen(cstr)); }