|
|
| version 1.85, 2010/08/27 02:53:24 | version 1.100, 2015/10/09 11:52:22 |
|---|---|
| Line 1 | Line 1 |
| /** @file | /** @file |
| Parser: Charset connection implementation. | Parser: Charset connection implementation. |
| Copyright(c) 2001-2009 ArtLebedev Group (http://www.artlebedev.com) | Copyright (c) 2001-2012 Art. Lebedev Studio (http://www.artlebedev.com) |
| Author: Alexander Petrosyan<paf@design.ru>(http://paf.design.ru) | Author: Alexander Petrosyan<paf@design.ru>(http://paf.design.ru) |
| */ | */ |
| static const char * const IDENT_CHARSET_C="$Date$"; | |
| #include "pa_charset.h" | #include "pa_charset.h" |
| #include "pa_charsets.h" | #include "pa_charsets.h" |
| // we are using some pcre_internal.h stuff as well | |
| #include "../lib/pcre/pa_pcre_internal.h" | |
| volatile const char * IDENT_PA_CHARSET_C="$Id$" IDENT_PA_CHARSET_H; | |
| #ifdef XML | #ifdef XML |
| #include "libxml/encoding.h" | #include "libxml/encoding.h" |
| #endif | #endif |
| Line 43 inline void prepare_case_tables(unsigned | Line 46 inline void prepare_case_tables(unsigned |
| for(int i=0; i<0x100; i++) | for(int i=0; i<0x100; i++) |
| lcc_table[i]=fcc_table[i]=(unsigned char)i; | lcc_table[i]=fcc_table[i]=(unsigned char)i; |
| } | } |
| inline void cstr2ctypes(unsigned char *tables, const unsigned char *cstr, | inline void cstr2ctypes(unsigned char *tables, const unsigned char *cstr, unsigned char bit) { |
| unsigned char bit) { | |
| unsigned char *ctypes_table=tables+ctypes_offset; | unsigned char *ctypes_table=tables+ctypes_offset; |
| ctypes_table[0]=bit; | ctypes_table[0]=bit; |
| for(; *cstr; cstr++) { | for(; *cstr; cstr++) { |
| Line 58 inline unsigned int to_wchar_code(const | Line 60 inline unsigned int to_wchar_code(const |
| if(cstr[1]==0) | if(cstr[1]==0) |
| return(unsigned int)(unsigned char)cstr[0]; | return(unsigned int)(unsigned char)cstr[0]; |
| char *error_pos; | return pa_atoui(cstr,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, unsigned char *tables, unsigned char bit, int group_offset=-1) { |
| unsigned char *tables, unsigned char bit, int group_offset=-1) { | |
| if(!belongs) | if(!belongs) |
| return; | return; |
| Line 75 static void element2ctypes(unsigned char | Line 75 static void element2ctypes(unsigned char |
| if(group_offset>=0) | if(group_offset>=0) |
| tables[cbits_offset+group_offset+c/8] |= 1<<(c%8); | tables[cbits_offset+group_offset+c/8] |= 1<<(c%8); |
| } | } |
| static void element2case(unsigned char from, unsigned char to, | static void element2case(unsigned char from, unsigned char to, unsigned char *tables) { |
| unsigned char *tables) { | |
| if(!to) | if(!to) |
| return; | return; |
| Line 86 static void element2case(unsigned char f | Line 85 static void element2case(unsigned char f |
| fcc_table[from]=to; fcc_table[to]=from; | fcc_table[from]=to; fcc_table[to]=from; |
| } | } |
| inline XMLByte *append_hex_8(XMLByte *dest, unsigned char c, const char* prefix=0) { | |
| if(prefix) { | |
| strcpy((char *)dest, prefix); | |
| dest+=strlen(prefix); | |
| } | |
| *dest++=hex_digits[c >> 4]; | |
| *dest++=hex_digits[c & 0x0F]; | |
| return dest; | |
| } | |
| inline XMLByte *append_hex_16(XMLByte *dest, unsigned int c, const char* prefix=0) { | |
| if(prefix) { | |
| strcpy((char *)dest, prefix); | |
| dest+=strlen(prefix); | |
| } | |
| *dest++=hex_digits[(c >> 12) & 0x0F]; | |
| *dest++=hex_digits[(c >> 8) & 0x0F]; | |
| *dest++=hex_digits[(c >> 4) & 0x0F]; | |
| *dest++=hex_digits[(c) & 0x0F]; | |
| return dest; | |
| } | |
| // methods | // methods |
| Charset::Charset(Request_charsets* charsets, const String::Body ANAME, const String* afile_spec): | Charset::Charset(Request_charsets* charsets, const String::Body ANAME, const String* afile_spec): |
| Line 101 Charset::Charset(Request_charsets* chars | Line 122 Charset::Charset(Request_charsets* chars |
| } else { | } else { |
| fisUTF8=true; | fisUTF8=true; |
| // grab default onces [for UTF-8 so to be able to make a-z =>A-Z | // grab default onces [for UTF-8 so to be able to make a-z =>A-Z |
| memcpy(pcre_tables, _pcre_default_tables, sizeof(pcre_tables)); | memcpy(pcre_tables, pa_pcre_default_tables, sizeof(pcre_tables)); |
| } | } |
| #ifdef XML | #ifdef XML |
| Line 152 void Charset::load_definition(Request_ch | Line 173 void Charset::load_definition(Request_ch |
| case 8: | case 8: |
| // charset | // charset |
| if(tables.toTableSize>MAX_CHARSET_UNI_CODES) | if(tables.toTableSize>MAX_CHARSET_UNI_CODES) |
| throw Exception(PARSER_RUNTIME, | throw Exception(PARSER_RUNTIME, &afile_spec, "charset must contain not more then %d unicode values", MAX_CHARSET_UNI_CODES); |
| &afile_spec, | |
| "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*/) |
| Line 171 void Charset::load_definition(Request_ch | Line 190 void Charset::load_definition(Request_ch |
| } | } |
| }; | }; |
| // parser charset tables declare only white-space before 0x20, thus adding the missing chars | |
| for(uint i=0; i<0x20; i++) | |
| if(!tables.fromTable[i]){ | |
| tables.fromTable[i]=i; | |
| tables.toTable[tables.toTableSize].intCh=i; | |
| tables.toTable[tables.toTableSize].extCh=(XMLByte)i; | |
| tables.toTableSize++; | |
| } | |
| // sort by the Unicode code point | // sort by the Unicode code point |
| sort_ToTable(); | sort_ToTable(); |
| } | } |
| Line 182 static int sort_cmp_Trans_rec_intCh(cons | Line 210 static int sort_cmp_Trans_rec_intCh(cons |
| } | } |
| void Charset::sort_ToTable() { | void Charset::sort_ToTable() { |
| _qsort(tables.toTable, tables.toTableSize, sizeof(*tables.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(tables.toTable, tables.toTableSize, sizeof(*tables.toTable), f); | //fwrite(tables.toTable, tables.toTableSize, sizeof(*tables.toTable), f); |
| //fclose(f); | //fclose(f); |
| } | } |
| // @todo: precache for spedup searching | // @todo: precache for spedup searching |
| static XMLByte xlatOneTo(const XMLCh toXlat, | static XMLByte xlatOneTo(const XMLCh toXlat, const Charset::Tables& tables, XMLByte not_found) { |
| const Charset::Tables& tables, | |
| XMLByte not_found) { | |
| int lo = 0; | int lo = 0; |
| int hi = tables.toTableSize - 1; | int hi = tables.toTableSize - 1; |
| while(lo<=hi) { | while(lo<=hi) { |
| Line 211 static XMLByte xlatOneTo(const XMLCh toX | Line 236 static XMLByte xlatOneTo(const XMLCh toX |
| return not_found; | return not_found; |
| } | } |
| String::C Charset::transcode(const String::C src, | String::C Charset::transcode(const String::C src, const Charset& source_charset, const Charset& dest_charset) { |
| const Charset& source_charset, | |
| const Charset& dest_charset) { | |
| if(!src.length) | if(!src.length) |
| return String::C("", 0); | return String::C("", 0); |
| Line 270 static const XMLByte gFirstByteMark[7] = | Line 293 static const XMLByte gFirstByteMark[7] = |
| 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC | 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC |
| }; | }; |
| static int transcodeToUTF8(const XMLByte* srcData, int& srcLen, | static int transcodeToUTF8(const XMLByte* srcData, int& srcLen, XMLByte *toFill, int& toFillLen, const Charset::Tables& tables) { |
| XMLByte *toFill, int& toFillLen, | |
| const Charset::Tables& tables) { | |
| const XMLByte* srcPtr=srcData; | const XMLByte* srcPtr=srcData; |
| const XMLByte* srcEnd=srcData+srcLen; | const XMLByte* srcEnd=srcData+srcLen; |
| XMLByte* outPtr=toFill; | XMLByte* outPtr=toFill; |
| Line 353 of ocetes consumed. | Line 374 of ocetes consumed. |
| return 0; | return 0; |
| } | } |
| /// @todo digital entites only when xml/html output [at output in html/xml mode, in html part of a letter] | /// @todo digital entites only when xml/html output [at output in html/xml mode, in html part of a letter] |
| static int transcodeFromUTF8(const XMLByte* srcData, int& srcLen, | static int transcodeFromUTF8(const XMLByte* srcData, int& srcLen, XMLByte* toFill, int& toFillLen, const Charset::Tables& tables) { |
| XMLByte* toFill, int& toFillLen, | |
| const Charset::Tables& tables) { | |
| const XMLByte* srcPtr=srcData; | const XMLByte* srcPtr=srcData; |
| const XMLByte* srcEnd=srcData+srcLen; | const XMLByte* srcEnd=srcData+srcLen; |
| XMLByte* outPtr=toFill; | XMLByte* outPtr=toFill; |
| Line 394 static int transcodeFromUTF8(const XMLBy | Line 413 static int transcodeFromUTF8(const XMLBy |
| break; | break; |
| default: | default: |
| throw Exception(0, | throw Exception(0, 0, "transcodeFromUTF8 error: wrong trailingBytes value(%d)", trailingBytes); // never |
| 0, | |
| "transcodeFromUTF8 error: wrong trailingBytes value(%d)", trailingBytes); // never | |
| } | } |
| tmpVal-=gUTFOffsets[trailingBytes]; | tmpVal-=gUTFOffsets[trailingBytes]; |
| Line 438 static bool need_escape(XMLByte c){ | Line 455 static bool need_escape(XMLByte c){ |
| !( | !( |
| (c<=127) | (c<=127) |
| && ( | && ( |
| ((c>='0') && (c<='9')) | pa_isalnum((unsigned char)c) |
| || ((c>='A') && (c<='Z')) | |
| || ((c>='a') && (c<='z')) | |
| || strchr("*@-_+./", c)!=0 | || strchr("*@-_+./", c)!=0 |
| ) | ) |
| ); | ); |
| Line 531 size_t Charset::calc_escaped_length_UTF8 | Line 546 size_t Charset::calc_escaped_length_UTF8 |
| return dest_length; | return dest_length; |
| } | } |
| size_t Charset::calc_escaped_length(XMLByte* src, size_t src_length, const Charset::Tables& tables){ | size_t Charset::calc_escaped_length(const XMLByte* src, size_t src_length, const Charset::Tables& tables){ |
| size_t dest_length=0; | const XMLByte* src_end=src+src_length; |
| const XMLByte* src_ptr=src; | |
| const XMLByte* src_end=src_ptr+src_length; | |
| XMLByte first_byte; | XMLByte first_byte; |
| XMLCh UTF8_char; | XMLCh UTF8_char; |
| while(uint char_size=readChar(src_ptr, src_end, first_byte, UTF8_char, tables)){ | size_t dest_length=0; |
| while(uint char_size=readChar(src, src_end, first_byte, UTF8_char, tables)){ | |
| if(char_size==1) | if(char_size==1) |
| dest_length+=(!first_byte/*replacement char '?'*/ || !need_escape(first_byte))?1:3/*'%XX'*/; | dest_length+=(!first_byte/*replacement char '?'*/ || !need_escape(first_byte))?1:3/*'%XX'*/; |
| else | else |
| Line 549 size_t Charset::calc_escaped_length(XMLB | Line 563 size_t Charset::calc_escaped_length(XMLB |
| } | } |
| size_t Charset::calc_escaped_length(const String::C src, const Charset& source_charset){ | size_t Charset::calc_escaped_length(const String::C src, const Charset& source_charset){ |
| size_t src_length=src.length; | if(!src.length) |
| if(!src_length) | |
| return 0; | return 0; |
| #ifdef PRECALCULATE_DEST_LENGTH | #ifdef PRECALCULATE_DEST_LENGTH |
| if(source_charset.isUTF8()) | if(source_charset.isUTF8()) |
| return calc_escaped_length_UTF8((XMLByte *)src.str, src_length); | return calc_escaped_length_UTF8((XMLByte *)src.str, src.length); |
| else | else |
| return calc_escaped_length((XMLByte *)src.str, src_length, source_charset.tables); | return calc_escaped_length((XMLByte *)src.str, src.length, source_charset.tables); |
| #else | #else |
| return src_length*6; // enough for %uXXXX but too memory-hungry | return src_length*6; // enough for %uXXXX but too memory-hungry |
| #endif | #endif |
| Line 567 size_t Charset::calc_escaped_length(cons | Line 580 size_t Charset::calc_escaped_length(cons |
| if(char_size==1) \ | if(char_size==1) \ |
| if(first_byte){ \ | if(first_byte){ \ |
| if(need_escape(first_byte)) \ | if(need_escape(first_byte)) \ |
| dest_ptr+=sprintf((char*)dest_ptr, "%%%02X", first_byte); /* %XX */ \ | dest_ptr=append_hex_8(dest_ptr, first_byte, "%"); /* %XX */ \ |
| else \ | else \ |
| *dest_ptr++=first_byte; /*as is*/ \ | *dest_ptr++=first_byte; /*as is*/ \ |
| } else \ | } else \ |
| *dest_ptr++='?'; /* replacement char '?' */ \ | *dest_ptr++='?'; /* replacement char '?' */ \ |
| else \ | else \ |
| dest_ptr+=sprintf((char*)dest_ptr, "%%u%04X", UTF8_char); // %uXXXX | dest_ptr=append_hex_16(dest_ptr, UTF8_char, "%u"); /* %uXXXX */ |
| size_t Charset::escape_UTF8(const XMLByte* src, size_t src_length, XMLByte* dest) { | size_t Charset::escape_UTF8(const XMLByte* src, size_t src_length, XMLByte* dest) { |
| Line 587 size_t Charset::escape_UTF8(const XMLByt | Line 600 size_t Charset::escape_UTF8(const XMLByt |
| } | } |
| size_t Charset::escape(const XMLByte* src, size_t src_length, XMLByte* dest, const Charset::Tables& tables) { | size_t Charset::escape(const XMLByte* src, size_t src_length, XMLByte* dest, const Charset::Tables& tables) { |
| const XMLByte* src_ptr=src; | |
| const XMLByte* src_end=src+src_length; | const XMLByte* src_end=src+src_length; |
| XMLByte* dest_ptr=dest; | XMLByte* dest_ptr=dest; |
| Line 595 size_t Charset::escape(const XMLByte* sr | Line 607 size_t Charset::escape(const XMLByte* sr |
| XMLCh UTF8_char; | XMLCh UTF8_char; |
| uint char_size; | uint char_size; |
| while(char_size=readChar(src_ptr, src_end, first_byte, UTF8_char, tables)) | while(char_size=readChar(src, src_end, first_byte, UTF8_char, tables)) |
| escape_char(dest_ptr, char_size, first_byte, UTF8_char) | escape_char(dest_ptr, char_size, first_byte, UTF8_char) |
| return dest_ptr - dest; | return dest_ptr - dest; |
| } | } |
| String::C Charset::escape(const String::C src, const Charset& source_charset){ | String::C Charset::escape(const String::C src, const Charset& source_charset){ |
| size_t src_length=src.length; | if(!src.length) |
| if(!src_length) | |
| return String::C("", 0); | return String::C("", 0); |
| size_t dest_calculated_length=calc_escaped_length(src, source_charset); | size_t dest_calculated_length=calc_escaped_length(src, source_charset); |
| Line 611 String::C Charset::escape(const String:: | Line 622 String::C Charset::escape(const String:: |
| size_t dest_length; | size_t dest_length; |
| if(source_charset.isUTF8()) | if(source_charset.isUTF8()) |
| dest_length=escape_UTF8((XMLByte *)src.str, src_length, dest_body); | dest_length=escape_UTF8((XMLByte *)src.str, src.length, dest_body); |
| else | else |
| dest_length=escape((XMLByte *)src.str, src_length, dest_body, source_charset.tables); | dest_length=escape((XMLByte *)src.str, src.length, dest_body, source_charset.tables); |
| if(dest_length>dest_calculated_length) | if(dest_length>dest_calculated_length) |
| throw Exception(0, 0, "Charset::escape buffer overflow"); | throw Exception(0, 0, "Charset::escape buffer overflow"); |
| Line 623 String::C Charset::escape(const String:: | Line 634 String::C Charset::escape(const String:: |
| } | } |
| String::Body Charset::escape(const String::Body src, const Charset& source_charset) { | String::Body Charset::escape(const String::Body src, const Charset& source_charset) { |
| const char *src_ptr=src.cstr(); | String::C dest=Charset::escape(String::C(src.cstr(), src.length()), source_charset); |
| size_t src_size=src.length(); | |
| String::C dest=Charset::escape(String::C(src_ptr, src_size), source_charset); | |
| return String::Body(dest.length ? dest.str:0); | return String::Body(dest.length ? dest.str:0); |
| } | } |
| Line 646 size_t Charset::calc_JSON_escaped_length | Line 653 size_t Charset::calc_JSON_escaped_length |
| size_t dest_length=0; | size_t dest_length=0; |
| for(UTF8_string_iterator i(src, src_length); i.has_next(); ){ | for(UTF8_string_iterator i(src, src_length); i.has_next(); ){ |
| if(i.getCharSize()==1) | if(i.getCharSize()==1){ |
| dest_length+=need_json_escape(i.getFirstByte()) ? 2 : 1; | XMLByte first_byte=i.getFirstByte(); |
| else | dest_length+=need_json_escape(first_byte) ? 2 : (first_byte < 0x20 && first_byte /* 0 replacement char is '?' */) ? 6 : 1; |
| } else | |
| dest_length+=6; // \uXXXX | dest_length+=6; // \uXXXX |
| } | } |
| return dest_length; | return dest_length; |
| } | } |
| size_t Charset::calc_JSON_escaped_length(XMLByte* src, size_t src_length, const Charset::Tables& tables){ | size_t Charset::calc_JSON_escaped_length(const XMLByte* src, size_t src_length, const Charset::Tables& tables){ |
| const XMLByte* src_ptr=src; | const XMLByte* src_end=src+src_length; |
| const XMLByte* src_end=src_ptr+src_length; | |
| XMLByte first_byte; | XMLByte first_byte; |
| XMLCh UTF8_char; | XMLCh UTF8_char; |
| size_t dest_length=0; | size_t dest_length=0; |
| while(uint char_size=readChar(src_ptr, src_end, first_byte, UTF8_char, tables)){ | while(uint char_size=readChar(src, src_end, first_byte, UTF8_char, tables)){ |
| if(char_size==1) | if(char_size==1) |
| dest_length+=(!first_byte/*replacement char '?'*/ || !need_json_escape(first_byte))? 1 : 2; | dest_length+=need_json_escape(first_byte) ? 2 : (first_byte < 0x20 && first_byte /* 0 replacement char is '?' */) ? 6 : 1; |
| else | else |
| dest_length+=6; // \uXXXX | dest_length+=6; // \uXXXX |
| } | } |
| Line 673 size_t Charset::calc_JSON_escaped_length | Line 680 size_t Charset::calc_JSON_escaped_length |
| } | } |
| size_t Charset::calc_JSON_escaped_length(const String::C src, const Charset& source_charset){ | size_t Charset::calc_JSON_escaped_length(const String::C src, const Charset& source_charset){ |
| size_t src_length=src.length; | if(!src.length) |
| if(!src_length) | |
| return 0; | return 0; |
| #ifdef PRECALCULATE_DEST_LENGTH | #ifdef PRECALCULATE_DEST_LENGTH |
| if(source_charset.isUTF8()) | if(source_charset.isUTF8()) |
| return calc_JSON_escaped_length_UTF8((XMLByte *)src.str, src_length); | return calc_JSON_escaped_length_UTF8((XMLByte *)src.str, src.length); |
| else | else |
| return calc_JSON_escaped_length((XMLByte *)src.str, src_length, source_charset.tables); | return calc_JSON_escaped_length((XMLByte *)src.str, src.length, source_charset.tables); |
| #else | #else |
| return src_length*6; // enough for \uXXXX but too memory-hungry | return src_length*6; // enough for \uXXXX but too memory-hungry |
| #endif | #endif |
| Line 699 size_t Charset::calc_JSON_escaped_length | Line 705 size_t Charset::calc_JSON_escaped_length |
| case '\b': *dest_ptr++='\\'; *dest_ptr++='b'; break; \ | case '\b': *dest_ptr++='\\'; *dest_ptr++='b'; break; \ |
| case '\f': *dest_ptr++='\\'; *dest_ptr++='f'; break; \ | case '\f': *dest_ptr++='\\'; *dest_ptr++='f'; break; \ |
| case 0 : *dest_ptr++='?'; break; /*replacement char*/ \ | case 0 : *dest_ptr++='?'; break; /*replacement char*/ \ |
| default : *dest_ptr++=first_byte; \ | default : if(first_byte < 0x20) dest_ptr=append_hex_16(dest_ptr, UTF8_char, "\\u"); \ |
| else *dest_ptr++=first_byte; \ | |
| } \ | } \ |
| else \ | else \ |
| dest_ptr+=sprintf((char*)dest_ptr, "\\u%04X", UTF8_char); // \uXXXX | dest_ptr=append_hex_16(dest_ptr, UTF8_char, "\\u"); // \uXXXX |
| size_t Charset::escape_JSON_UTF8(const XMLByte* src, size_t src_length, XMLByte* dest) { | size_t Charset::escape_JSON_UTF8(const XMLByte* src, size_t src_length, XMLByte* dest) { |
| Line 716 size_t Charset::escape_JSON_UTF8(const X | Line 723 size_t Charset::escape_JSON_UTF8(const X |
| } | } |
| size_t Charset::escape_JSON(const XMLByte* src, size_t src_length, XMLByte* dest, const Charset::Tables& tables) { | size_t Charset::escape_JSON(const XMLByte* src, size_t src_length, XMLByte* dest, const Charset::Tables& tables) { |
| const XMLByte* src_ptr=src; | |
| const XMLByte* src_end=src+src_length; | const XMLByte* src_end=src+src_length; |
| XMLByte* dest_ptr=dest; | XMLByte* dest_ptr=dest; |
| Line 724 size_t Charset::escape_JSON(const XMLByt | Line 730 size_t Charset::escape_JSON(const XMLByt |
| XMLCh UTF8_char; | XMLCh UTF8_char; |
| uint char_size; | uint char_size; |
| while(char_size=readChar(src_ptr, src_end, first_byte, UTF8_char, tables)) | while(char_size=readChar(src, src_end, first_byte, UTF8_char, tables)) |
| escape_char_JSON(dest_ptr, char_size, first_byte, UTF8_char) | escape_char_JSON(dest_ptr, char_size, first_byte, UTF8_char) |
| return dest_ptr - dest; | return dest_ptr - dest; |
| } | } |
| String::C Charset::escape_JSON(const String::C src, const Charset& source_charset){ | String::C Charset::escape_JSON(const String::C src, const Charset& source_charset){ |
| size_t src_length=src.length; | if(!src.length) |
| if(!src_length) | |
| return String::C("", 0); | return String::C("", 0); |
| size_t dest_calculated_length=calc_JSON_escaped_length(src, source_charset); | size_t dest_calculated_length=calc_JSON_escaped_length(src, source_charset); |
| XMLByte *dest_body=new(PointerFreeGC) XMLByte[dest_calculated_length+1/*terminator*/]; | XMLByte *dest_body=new(PointerFreeGC) XMLByte[dest_calculated_length+1/*terminator*/]; |
| size_t dest_length; | size_t dest_length; |
| if(source_charset.isUTF8()) | if(source_charset.isUTF8()) |
| dest_length=escape_JSON_UTF8((XMLByte *)src.str, src_length, dest_body); | dest_length=escape_JSON_UTF8((XMLByte *)src.str, src.length, dest_body); |
| else | else |
| dest_length=escape_JSON((XMLByte *)src.str, src_length, dest_body, source_charset.tables); | dest_length=escape_JSON((XMLByte *)src.str, src.length, dest_body, source_charset.tables); |
| if(dest_length>dest_calculated_length) | if(dest_length>dest_calculated_length) |
| throw Exception(0, 0, "Charset::escape_JSON buffer overflow"); | throw Exception(0, 0, "Charset::escape_JSON buffer overflow"); |
| Line 753 String::C Charset::escape_JSON(const Str | Line 757 String::C Charset::escape_JSON(const Str |
| } | } |
| String::Body Charset::escape_JSON(const String::Body src, const Charset& source_charset) { | String::Body Charset::escape_JSON(const String::Body src, const Charset& source_charset) { |
| const char *src_ptr=src.cstr(); | String::C dest=Charset::escape_JSON(String::C(src.cstr(), src.length()), source_charset); |
| size_t src_size=src.length(); | |
| String::C dest=Charset::escape_JSON(String::C(src_ptr, src_size), source_charset); | |
| return String::Body(dest.length ? dest.str:0); | return String::Body(dest.length ? dest.str:0); |
| } | } |
| Line 792 const String::C Charset::transcodeToUTF8 | Line 792 const String::C Charset::transcodeToUTF8 |
| (XMLByte *)src.str, src_length, | (XMLByte *)src.str, src_length, |
| dest_body, dest_length, | dest_body, dest_length, |
| tables)<0) | tables)<0) |
| throw Exception(0, | throw Exception(0, 0, "Charset::transcodeToUTF8 buffer overflow"); |
| 0, | |
| "Charset::transcodeToUTF8 buffer overflow"); | |
| assert(dest_length<=saved_dest_length); | assert(dest_length<=saved_dest_length); |
| dest_body[dest_length]=0; // terminator | dest_body[dest_length]=0; // terminator |
| Line 869 static void store_UTF8(XMLCh src, XMLByt | Line 867 static void store_UTF8(XMLCh src, XMLByt |
| outPtr+= encodedBytes; | outPtr+= encodedBytes; |
| } | } |
| static void change_case_UTF8(XMLCh src, XMLByte*& outPtr, | static void change_case_UTF8(XMLCh src, XMLByte*& outPtr, const Charset::UTF8CaseTable& table) { |
| const Charset::UTF8CaseTable& table) { | |
| store_UTF8(change_case_UTF8(src, table), outPtr); | store_UTF8(change_case_UTF8(src, table), outPtr); |
| }; | } |
| void change_case_UTF8(const XMLByte* srcData, size_t srcLen, | |
| XMLByte* toFill, size_t toFillLen, | void change_case_UTF8(const XMLByte* srcData, size_t srcLen, XMLByte* toFill, size_t toFillLen, const Charset::UTF8CaseTable& table) { |
| const Charset::UTF8CaseTable& table) { | |
| const XMLByte* srcPtr=srcData; | const XMLByte* srcPtr=srcData; |
| const XMLByte* srcEnd=srcData+srcLen; | const XMLByte* srcEnd=srcData+srcLen; |
| XMLByte* outPtr=toFill; | XMLByte* outPtr=toFill; |
| Line 907 void change_case_UTF8(const XMLByte* src | Line 903 void change_case_UTF8(const XMLByte* src |
| break; | break; |
| default: | default: |
| throw Exception(0, | throw Exception(0, 0, "change_case_UTF8 error: wrong trailingBytes value(%d)", trailingBytes); |
| 0, | |
| "change_case_UTF8 error: wrong trailingBytes value(%d)", trailingBytes); | |
| } | } |
| tmpVal-=gUTFOffsets[trailingBytes]; | tmpVal-=gUTFOffsets[trailingBytes]; |
| Line 919 void change_case_UTF8(const XMLByte* src | Line 913 void change_case_UTF8(const XMLByte* src |
| if(!(tmpVal & 0xFFFF0000)) | if(!(tmpVal & 0xFFFF0000)) |
| change_case_UTF8(tmpVal, outPtr, table); | change_case_UTF8(tmpVal, outPtr, table); |
| else | else |
| throw Exception(0, | throw Exception(0, 0, "change_case_UTF8 error: too big tmpVal(0x%08X)", tmpVal); |
| 0, | |
| "change_case_UTF8 error: too big tmpVal(0x%08X)", tmpVal); | |
| } | } |
| if(srcPtr!=outPtr) | if(srcPtr!=outPtr) |
| throw Exception(0, | throw Exception(0, 0, "change_case_UTF8 error: end pointers do not match"); |
| 0, | |
| "change_case_UTF8 error: end pointers do not match"); | |
| } | } |
| static size_t getDecNumLength(XMLCh UTF8Char){ | static size_t getDecNumLength(XMLCh UTF8Char){ |
| Line 946 const String::C Charset::transcodeFromUT | Line 936 const String::C Charset::transcodeFromUT |
| #ifdef PRECALCULATE_DEST_LENGTH | #ifdef PRECALCULATE_DEST_LENGTH |
| int dest_length=0; | int dest_length=0; |
| for(UTF8_string_iterator i((XMLByte *)src.str, src_length); i.has_next(); ){ | for(UTF8_string_iterator i((XMLByte *)src.str, src_length); i.has_next(); ){ |
| if(i.getCharSize()==1) | dest_length += ( i.next() & 0xFFFF0000 ) |
| dest_length++; | ? 3*i.getCharSize() // %XX for each byte |
| else | : ( xlatOneTo(i.next(), tables, 0) != 0 ) |
| dest_length+=(i.next() & 0xFFFF0000) | ? 1 // can convert it to a single char |
| ?i.getCharSize()*3 // '%XX' for each byte | : 3+getDecNumLength( i.next() ); // print char as &#XX;, &#XXX;, &#XXXX; or &#XXXXX; |
| :(xlatOneTo(i.next(), tables, 0)!=0) | |
| ?1 // can convert it to single char | |
| :getDecNumLength(i.next())+3; // &#XX; - &#XXXXX; | |
| } | } |
| #else | #else |
| // so that surly enough, "&#XXX;" has max ratio (huh? 8 bytes needed for '&#XXXXX;') | // so that surly enough, "&#XXX;" has max ratio (huh? 8 bytes needed for '&#XXXXX;') |
| Line 969 const String::C Charset::transcodeFromUT | Line 956 const String::C Charset::transcodeFromUT |
| (XMLByte *)src.str, src_length, | (XMLByte *)src.str, src_length, |
| dest_body, dest_length, | dest_body, dest_length, |
| tables)<0) | tables)<0) |
| throw Exception(0, | throw Exception(0, 0, "Charset::transcodeFromUTF8 buffer overflow"); |
| 0, | |
| "Charset::transcodeFromUTF8 buffer overflow"); | |
| assert(dest_length<=saved_dest_length); | assert(dest_length<=saved_dest_length); |
| dest_body[dest_length]=0; // terminator | dest_body[dest_length]=0; // terminator |
| Line 979 const String::C Charset::transcodeFromUT | Line 964 const String::C Charset::transcodeFromUT |
| } | } |
| /// transcode using both charsets | /// transcode using both charsets |
| const String::C Charset::transcodeToCharset(const String::C src, | const String::C Charset::transcodeToCharset(const String::C src, const Charset& dest_charset) const { |
| const Charset& dest_charset) const { | |
| if(&dest_charset==this) | if(&dest_charset==this) |
| return src; | return src; |
| else { | else { |
| Line 1079 static size_t handlers_count=0; | Line 1063 static size_t handlers_count=0; |
| void Charset::addEncoding(char *name_cstr) { | void Charset::addEncoding(char *name_cstr) { |
| if(handlers_count==MAX_CHARSETS) | if(handlers_count==MAX_CHARSETS) |
| throw Exception(0, | throw Exception(0, 0, "already allocated %d handlers, no space for new encoding '%s'", MAX_CHARSETS, name_cstr); |
| 0, | |
| "already allocated %d handlers, no space for new encoding '%s'", | |
| MAX_CHARSETS, name_cstr); | |
| xmlCharEncodingHandler* handler=new(UseGC) xmlCharEncodingHandler; | xmlCharEncodingHandler* handler=new xmlCharEncodingHandler; |
| { | { |
| handler->name=name_cstr; | handler->name=name_cstr; |
| handler->input=inputFuncs[handlers_count]; | handler->input=inputFuncs[handlers_count]; |
| Line 1104 void Charset::initTranscoder(const Strin | Line 1085 void Charset::initTranscoder(const Strin |
| xmlCharEncodingHandler& Charset::transcoder(const String::Body NAME) { | xmlCharEncodingHandler& Charset::transcoder(const String::Body NAME) { |
| if(!ftranscoder) | if(!ftranscoder) |
| throw Exception(PARSER_RUNTIME, | throw Exception(PARSER_RUNTIME, new String(NAME, String::L_TAINTED), "unsupported encoding"); |
| new String(NAME, String::L_TAINTED), | |
| "unsupported encoding"); | |
| return *ftranscoder; | return *ftranscoder; |
| } | } |
| Line 1135 String::C Charset::transcode_cstr(const | Line 1114 String::C Charset::transcode_cstr(const |
| error=0; | error=0; |
| } | } |
| if(error<0) | if(error<0) |
| throw Exception(0, | throw Exception(0, 0, "transcode_cstr failed (%d)", error); |
| 0, | |
| "transcode_cstr failed (%d)", error); | |
| assert(outlen<=saved_outlen); out[outlen]=0; | assert(outlen<=saved_outlen); out[outlen]=0; |
| return String::C(out, outlen); | return String::C(out, outlen); |
| Line 1179 xmlChar* Charset::transcode_buf2xchar(co | Line 1156 xmlChar* Charset::transcode_buf2xchar(co |
| } | } |
| if(error<0) | if(error<0) |
| throw Exception(0, | throw Exception(0, 0, "transcode_buf failed (%d)", error); |
| 0, | |
| "transcode_buf failed (%d)", error); | |
| assert(outlen<=saved_outlen); out[outlen]=0; | assert(outlen<=saved_outlen); out[outlen]=0; |
| return out; | return out; |
| Line 1197 xmlChar* Charset::transcode(const String | Line 1172 xmlChar* Charset::transcode(const String |
| } | } |
| #endif | #endif |
| String::Body Charset::transcode(const String::Body src, | String::Body Charset::transcode(const String::Body src, const Charset& source_transcoder, const Charset& dest_transcoder) { |
| const Charset& source_transcoder, | |
| const Charset& dest_transcoder) { | |
| const char *src_ptr=src.cstr(); | const char *src_ptr=src.cstr(); |
| size_t src_size=src.length(); | size_t src_size=src.length(); |
| Line 1209 String::Body Charset::transcode(const St | Line 1182 String::Body Charset::transcode(const St |
| return String::Body(dest.length ? dest.str:0); | return String::Body(dest.length ? dest.str:0); |
| } | } |
| String& Charset::transcode(const String& src, | String& Charset::transcode(const String& src, const Charset& source_transcoder, const Charset& dest_transcoder) { |
| const Charset& source_transcoder, | |
| const Charset& dest_transcoder) { | |
| if(src.is_empty()) | if(src.is_empty()) |
| return *new String(); | return *new String(); |
| return *new String(transcode((String::Body)src, source_transcoder, dest_transcoder), String::L_CLEAN); | return *new String(transcode((String::Body)src, source_transcoder, dest_transcoder), String::L_CLEAN); |
| } | } |
| void Charset::transcode(ArrayString& src, | void Charset::transcode(ArrayString& src, const Charset& source_transcoder, const Charset& dest_transcoder) { |
| const Charset& source_transcoder, | |
| const Charset& dest_transcoder) { | |
| for(size_t i=0; i<src.count(); i++) | for(size_t i=0; i<src.count(); i++) |
| src.put(i, &transcode(*src[i], source_transcoder, dest_transcoder)); | src.put(i, &transcode(*src[i], source_transcoder, dest_transcoder)); |
| } | } |
| Line 1231 struct Transcode_pair_info { | Line 1200 struct Transcode_pair_info { |
| const Charset* dest_transcoder; | const Charset* dest_transcoder; |
| }; | }; |
| #endif | #endif |
| static void transcode_pair(HashStringValue::key_type /*akey*/, | static void transcode_pair(HashStringValue::key_type /*akey*/, String::Body& avalue, Transcode_pair_info* info) { |
| String::Body& avalue, | avalue=Charset::transcode(avalue, *info->source_transcoder, *info->dest_transcoder); |
| Transcode_pair_info* info) { | |
| avalue=Charset::transcode(avalue, | |
| *info->source_transcoder, | |
| *info->dest_transcoder); | |
| } | } |
| void Charset::transcode(HashStringString& src, | void Charset::transcode(HashStringString& src, const Charset& source_transcoder, const Charset& dest_transcoder) { |
| const Charset& source_transcoder, | |
| const Charset& dest_transcoder) { | |
| Transcode_pair_info info={&source_transcoder, &dest_transcoder}; | Transcode_pair_info info={&source_transcoder, &dest_transcoder}; |
| src.for_each_ref<Transcode_pair_info*>(transcode_pair, &info); | src.for_each_ref<Transcode_pair_info*>(transcode_pair, &info); |
| } | } |
| Line 1264 size_t getUTF8CharPos(const XMLByte* src | Line 1227 size_t getUTF8CharPos(const XMLByte* src |
| } | } |
| // scan till end but position in bytes still too low | // scan till end but position in bytes still too low |
| throw Exception(0, | throw Exception(0, 0, "Error convertion byte pos to char pos"); |
| 0, | |
| "Error convertion byte pos to char pos"); | |
| } | } |
| size_t lengthUTF8(const XMLByte* srcBegin, const XMLByte* srcEnd){ | size_t lengthUTF8(const XMLByte* srcBegin, const XMLByte* srcEnd){ |
| Line 1281 unsigned int lengthUTF8Char(const XMLByt | Line 1242 unsigned int lengthUTF8Char(const XMLByt |
| return gUTFBytes[c]+1; | return gUTFBytes[c]+1; |
| } | } |
| const char *fixUTF8(const char *src){ | |
| if(src && *src){ | |
| size_t length=strlen(src); | |
| int error_offset; | |
| if(pa_pcre_valid_utf((unsigned char *)src, length, &error_offset)){ | |
| char *result=(char *)pa_malloc_atomic(length+1); | |
| char *dst=result; | |
| do { | |
| if(error_offset){ | |
| strncpy(dst, src, error_offset); | |
| dst+=error_offset; | |
| src+=error_offset; | |
| length-=error_offset; | |
| } | |
| *dst++='?'; | |
| src++; | |
| length--; | |
| } while (length && pa_pcre_valid_utf((unsigned char *)src, length, &error_offset)); | |
| if(length){ | |
| strcpy(dst, src); | |
| } else { | |
| *dst='\0'; | |
| } | |
| return result; | |
| } | |
| } | |
| return src; | |
| } | |
| bool UTF8_string_iterator::has_next(){ | bool UTF8_string_iterator::has_next(){ |
| fcharSize=readUTF8Char(fsrcPtr, fsrcEnd, ffirstByte, fUTF8Char); | fcharSize=readUTF8Char(fsrcPtr, fsrcEnd, ffirstByte, fUTF8Char); |
| return fcharSize!=0; | return fcharSize!=0; |