lttoolbox/tmx_preprocessor.cc

   1 /*
   2  * Copyright (C) 2005 Universitat d'Alacant / Universidad de Alicante
   3  *
   4  * This program is free software; you can redistribute it and/or
   5  * modify it under the terms of the GNU General Public License as
   6  * published by the Free Software Foundation; either version 2 of the
   7  * License, or (at your option) any later version.
   8  *
   9  * This program is distributed in the hope that it will be useful, but
  10  * WITHOUT ANY WARRANTY; without even the implied warranty of
  11  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  12  * General Public License for more details.
  13  *
  14  * You should have received a copy of the GNU General Public License
  15  * along with this program; if not, write to the Free Software
  16  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
  17  * 02111-1307, USA.
  18  */
  19 #include <lttoolbox/tmx_preprocessor.h>
  20 #include <lttoolbox/lt_locale.h>
  21 #include <lttoolbox/xml_parse_util.h>
  22
  23 #include <cstdlib>
  24 #include <iostream>
  25 #include <libxml/encoding.h>
  26
  27 using namespace std;
  28
  29 wstring const TMXPreprocessor::TMX_BODY_ELEM    = L"body";
  30 wstring const TMXPreprocessor::TMX_HEADER_ELEM  = L"header";
  31 wstring const TMXPreprocessor::TMX_MAP_ELEM     = L"map";
  32 wstring const TMXPreprocessor::TMX_NOTE_ELEM    = L"note";
  33 wstring const TMXPreprocessor::TMX_PROP_ELEM    = L"prop";
  34 wstring const TMXPreprocessor::TMX_SEG_ELEM     = L"seg";
  35 wstring const TMXPreprocessor::TMX_TMX_ELEM     = L"tmx";
  36 wstring const TMXPreprocessor::TMX_TU_ELEM      = L"tu";
  37 wstring const TMXPreprocessor::TMX_TUV_ELEM     = L"tuv";
  38 wstring const TMXPreprocessor::TMX_UDE_ELEM     = L"ude";
  39 wstring const TMXPreprocessor::TMX_BPT_ELEM     = L"bpt";
  40 wstring const TMXPreprocessor::TMX_EPT_ELEM     = L"ept";
  41 wstring const TMXPreprocessor::TMX_HI_ELEM      = L"hi";
  42 wstring const TMXPreprocessor::TMX_IT_ELEM      = L"it";
  43 wstring const TMXPreprocessor::TMX_PH_ELEM      = L"ph";
  44 wstring const TMXPreprocessor::TMX_SUB_ELEM     = L"sub";
  45 wstring const TMXPreprocessor::TMX_UT_ELEM      = L"ut";
  46 wstring const TMXPreprocessor::TMX_ADMINLANG_ATTR       = L"adminlang";
  47 wstring const TMXPreprocessor::TMX_ASSOC_ATTR           = L"assoc";
  48 wstring const TMXPreprocessor::TMX_BASE_ATTR            = L"base";
  49 wstring const TMXPreprocessor::TMX_CHANGEDATE_ATTR      = L"changedate";
  50 wstring const TMXPreprocessor::TMX_CHANGEID_ATTR        = L"changeid";
  51 wstring const TMXPreprocessor::TMX_CODE_ATTR            = L"code";
  52 wstring const TMXPreprocessor::TMX_CREATIONDATE_ATTR    = L"creationdate";
  53 wstring const TMXPreprocessor::TMX_CREATIONID_ATTR      = L"creationid";
  54 wstring const TMXPreprocessor::TMX_CREATIONTOOL_ATTR    = L"creationtool";
  55 wstring const TMXPreprocessor::TMX_CREATIONTOOLVERSION_ATTR = L"creationtoolversion";
  56 wstring const TMXPreprocessor::TMX_DATATYPE_ATTR        = L"datatype";
  57 wstring const TMXPreprocessor::TMX_ENT_ATTR             = L"ent";
  58 wstring const TMXPreprocessor::TMX_I_ATTR               = L"i";
  59 wstring const TMXPreprocessor::TMX_LASTUSAGEDATE_ATTR   = L"lastusagedate";
  60 wstring const TMXPreprocessor::TMX_NAME_ATTR            = L"name";
  61 wstring const TMXPreprocessor::TMX_O_ENCODING_ATTR      = L"o-encoding";
  62 wstring const TMXPreprocessor::TMX_O_TMF_ATTR           = L"o-tmf";
  63 wstring const TMXPreprocessor::TMX_POS_ATTR             = L"pos";
  64 wstring const TMXPreprocessor::TMX_SEGTYPE_ATTR         = L"segtype";
  65 wstring const TMXPreprocessor::TMX_SRCLANG_ATTR         = L"srclang";
  66 wstring const TMXPreprocessor::TMX_SUBST_ATTR           = L"subst";
  67 wstring const TMXPreprocessor::TMX_TUID_ATTR            = L"tuid";
  68 wstring const TMXPreprocessor::TMX_TYPE_ATTR            = L"type";
  69 wstring const TMXPreprocessor::TMX_UNICODE_ATTR         = L"unicode";
  70 wstring const TMXPreprocessor::TMX_USAGECOUNT_ATTR      = L"usagecount";
  71 wstring const TMXPreprocessor::TMX_VERSION_ATTR         = L"version";
  72 wstring const TMXPreprocessor::TMX_X_ATTR               = L"x";
  73 wstring const TMXPreprocessor::TMX_XML_LANG_ATTR        = L"xml:lang";
  74
  75 TMXPreprocessor::TMXPreprocessor()
  76 {
  77   LtLocale::tryToSetLocale();
  78 }
  79
  80 TMXPreprocessor::~TMXPreprocessor()
  81 {
  82 }
  83
  84 void
  85 TMXPreprocessor::parse(string const &filename, wstring const &dir)
  86 {
  87   reader = xmlReaderForFile(filename.c_str(), NULL, 0);
  88   if(reader == NULL)
  89   {
  90     cerr << "Error: Cannot open '";
  91     cerr << filename;
  92     cerr << "'." << endl;
  93     exit(EXIT_FAILURE);
  94   }
  95
  96   int ret = xmlTextReaderRead(reader);
  97   while(ret == 1)
  98   {
  99     procNode();
 100     ret = xmlTextReaderRead(reader);
 101   }
 102
 103   if(ret != 0)
 104   {
 105     wcerr << L"Error: Parse error at the end of input." << endl;
 106   }
 107
 108   xmlFreeTextReader(reader);
 109   xmlCleanupParser();
 110 }
 111
 112 void
 113 TMXPreprocessor::requireEmptyError(wstring const &name)
 114 {
 115   if(!xmlTextReaderIsEmptyElement(reader))
 116   {
 117     wcerr << L"Error (" << xmlTextReaderGetParserLineNumber(reader);
 118     wcerr << L"): Non-empty element '<" << name << L">' should be empty." << endl;
 119     exit(EXIT_FAILURE);
 120   }
 121 }
 122
 123 bool
 124 TMXPreprocessor::allBlanks()
 125 {
 126   bool flag = true;
 127   wstring text = XMLParseUtil::towstring(xmlTextReaderConstValue(reader));
 128
 129   for(unsigned int i = 0, limit = text.size(); i < limit; i++)
 130   {
 131     flag = flag && iswspace(text[i]);
 132   }
 133
 134   return flag;
 135 }
 136
 137 void
 138 TMXPreprocessor::skipBlanks(wstring &name)
 139 {
 140   while(name == L"#text" || name == L"#comment")
 141   {
 142     if(name != L"#comment")
 143     {
 144       if(!allBlanks())
 145       {
 146         wcerr << L"Error (" << xmlTextReaderGetParserLineNumber(reader);
 147         wcerr << L"): Invalid construction." << endl;
 148         exit(EXIT_FAILURE);
 149       }
 150     }
 151
 152     xmlTextReaderRead(reader);
 153     name = XMLParseUtil::towstring(xmlTextReaderConstName(reader));
 154   }
 155 }
 156
 157 void
 158 TMXPreprocessor::skip(wstring &name, wstring const &elem)
 159 {
 160   xmlTextReaderRead(reader);
 161   name = XMLParseUtil::towstring(xmlTextReaderConstName(reader));
 162
 163   while(name == L"#text" || name == L"#comment")
 164   {
 165     if(name != L"#comment")
 166     {
 167       if(!allBlanks())
 168       {
 169         wcerr << L"Error (" << xmlTextReaderGetParserLineNumber(reader);
 170         wcerr << L"): Invalid construction." << endl;
 171         exit(EXIT_FAILURE);
 172       }
 173     }
 174     xmlTextReaderRead(reader);
 175     name = XMLParseUtil::towstring(xmlTextReaderConstName(reader));
 176   }
 177
 178   if(name != elem)
 179   {
 180     wcerr << L"Error (" << xmlTextReaderGetParserLineNumber(reader);
 181     wcerr << L"): Expected '<" << elem << L">'." << endl;
 182     exit(EXIT_FAILURE);
 183   }
 184 }
 185
 186 wstring
 187 TMXPreprocessor::attrib(wstring const &name)
 188 {
 189   return XMLParseUtil::attrib(reader, name);
 190 }
 191
 192 void
 193 TMXPreprocessor::requireAttribute(wstring const &value, wstring const &attrname,
 194                            wstring const &elemname)
 195 {
 196   if(value == L"")
 197   {
 198     wcerr << L"Error (" << xmlTextReaderGetParserLineNumber(reader);
 199     wcerr << L"): '<" << elemname;
 200     wcerr << L"' element must specify non-void '";
 201     wcerr << attrname << L"' attribute." << endl;
 202     exit(EXIT_FAILURE);
 203   }
 204 }
 205
 206 void
 207 TMXPreprocessor::procNode()
 208 {
 209   xmlChar const *xnombre = xmlTextReaderConstName(reader);
 210   wstring nombre = XMLParseUtil::towstring(xnombre);
 211
 212   wcout << nombre << endl;
 213
 214
 215   if(nombre == TMX_TMX_ELEM)
 216   {
 217     wcout << nombre << endl;
 218   }
 219   else
 220   {
 221     wcerr << L"Error (" << xmlTextReaderGetParserLineNumber(reader);
 222     wcerr << L"): Invalid node '<" << nombre << L">'." << endl;
 223     exit(EXIT_FAILURE);
 224   }
 225 }
 226
 227 int main(int argc, char *argv[])
 228 {
 229   TMXPreprocessor tmxp;
 230   if(argc == 3)
 231   {
 232     tmxp.parse(argv[1], L"LR");
 233   }
 234   return 0;
 235 }
 236