web/public_php/webtt/app/vendors/SheetParser.php

   1 <?php
   2 /*
   3         Ryzom Core Web-Based Translation Tool
   4         Copyright (C) 2011 Piotr Kaczmarek <p.kaczmarek@openlink.pl>
   5
   6         This program is free software: you can redistribute it and/or modify
   7         it under the terms of the GNU General Public License as published by
   8         the Free Software Foundation, either version 3 of the License, or
   9         (at your option) any later version.
  10
  11         This program is distributed in the hope that it will be useful,
  12         but WITHOUT ANY WARRANTY; without even the implied warranty of
  13         MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14         GNU General Public License for more details.
  15
  16         You should have received a copy of the GNU General Public License
  17         along with this program.  If not, see <http://www.gnu.org/licenses/>.
  18 */
  19 ?>
  20 <?php
  21 class SheetParser
  22 {
  23         var $debug = false;
  24
  25         function parseLine($str)
  26         {
  27                 $arr = str_getcsv($str, "\t");
  28                 return $arr;
  29         }
  30
  31         function parseFile($file)
  32         {
  33                 $parsedEnt = array();
  34                 $newEnt = false;
  35                 $prevStringLine = false;
  36                 $entities = array();
  37                 $diffFile = false;
  38
  39                 $file = mb_convert_encoding($file, 'UTF-8', 'UTF-16');
  40 //              $file = $this->removeBOM($file);
  41 //              $file = $this->removeComments($file);
  42                 $lines = explode("\n", $file);
  43                 if ($this->debug)
  44                 {
  45                         echo "<pre>\n\n";
  46                 }
  47                 $line_no=1;
  48
  49 /*              var_dump(setlocale(LC_ALL,NULL));
  50                 var_dump(setlocale(LC_ALL,'pl_PL.UTF-8'));*/
  51                 // Need to set UTF-8 locale to get str_getcsv to work with UTF-8 cyryllic
  52                 setlocale(LC_ALL,'pl_PL.UTF-8');
  53
  54                 foreach ($lines as $line)
  55                 {
  56                         if ($this->debug)
  57                         {
  58                                 echo "\n\t#################### LINE NUMBER " . $line_no . "\n\n";
  59                         }
  60
  61 //                      var_dump($line);
  62                         $line = rtrim($line,"\r\n");
  63
  64                         $parsedLine = $this->parseLine($line);
  65
  66                         if (!$line || mb_strpos($line, "DIFF NOT") === 0 || mb_strpos($line, "REMOVE THE") === 0)
  67                                 continue;
  68
  69                         if ($line_no == 1)
  70                         {
  71                                 $parsedEnt["type"] = "sheet_description";
  72                                 if ($parsedLine[0] == "DIFF_CMD")
  73                                         $diffFile = true;
  74                                 $parsedEnt["sheet_id_column"] = $parsedLine[2];
  75                         }
  76                         else
  77                         {
  78                                 if ($diffFile)
  79                                 {
  80                                         $parsedEnt["diff"] = $parsedEnt["command"] = rtrim($parsedLine[0]);
  81                                         $parsedEnt["hash_value"] = $parsedLine[1];
  82                                         $parsedEnt["identifier"] = $parsedLine[2];
  83                                 }
  84                                 else
  85                                 {
  86                                         $parsedEnt["hash_value"] = $parsedLine[0];
  87                                         $parsedEnt["identifier"] = $parsedLine[1];
  88                                 }
  89                                 $parsedEnt["type"] = "sheet";
  90                         }
  91
  92                         if ($diffFile)
  93                                 $columns = array_slice($parsedLine, 3);
  94                         else
  95                                 $columns = array_slice($parsedLine, 2);
  96
  97                         if (!isset($columnsCount))
  98                                 $columnsCount = count($columns);
  99
 100                         $parsedEnt["columns"] = $columns;
 101
 102                         if ($this->debug)
 103                         {
 104                                 echo "%%%% parsedEnt %%%%%\n";
 105                                 var_dump($parsedEnt);
 106                         }
 107
 108 /*                      if ($parsedLine["type"] == "internal_index")
 109                                         $parsedEnt["internal_index"] = $parsedLine["index"];
 110
 111                         if ($parsedLine["type"] == "diff")
 112                         {
 113                                 $parsedEnt["diff"] = $parsedLine["command"];
 114                                 $parsedEnt["index"] = $parsedLine["index"];
 115                         }*/
 116
 117                         $newEnt = true;
 118
 119                         if ($newEnt)
 120                         {
 121                                 if ($this->debug && 0)
 122                                 {
 123                                         echo "\t%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n";
 124                                         echo "\t%%%% newEnt %%%%%%%%% newEnt %%%%%%%%% newEnt %%%%%%%%% newEnt %%%%%\n";
 125                                         echo "\t%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\n\n";
 126                                         var_dump($parsedEnt);
 127                                 }
 128 /*                              if (!isset($parsedEnt["diff"]) && !isset($parsedEnt["index"]))
 129                                         $parsedEnt["index"] = $parsedEnt["internal_index"];*/
 130
 131                                 $entities[] = $parsedEnt;
 132                                 $parsedEnt =array();
 133                                 $newEnt = false;
 134                         }
 135                         $line_no++;
 136                 }
 137
 138                 if ($this->debug && 0)
 139                 {
 140                         echo "<pre>";
 141                         var_dump($entities);
 142                         echo "</pre>\n";
 143                 }
 144                 return $entities;
 145         }
 146
 147         function CRLF($s)
 148         {
 149                 $s = str_replace("\r\n", "\n", $s);
 150                 $s = str_replace("\n", "\r\n", $s);
 151                 return $s;
 152         }
 153
 154         function addBOM($str)
 155         {
 156                 if(($bom = substr($str, 0,3)) != pack("CCC",0xef,0xbb,0xbf))
 157                         return pack("CCC",0xef,0xbb,0xbf) . $str;
 158                 else
 159                         return $str;
 160         }
 161
 162         function buildFile($entities)
 163         {
 164                 $content = '';
 165                 foreach ($entities as $ent)
 166                 {
 167                         if ($ent['type'] == 'sheet_description')
 168                         {
 169                                 $_columns = $ent['columns'];
 170                                 $_sheet_id_column = $ent['sheet_id_column'];
 171
 172                                 if (isset($ent['diff']))
 173                                 {
 174                                         $content .= 'DIFF_CMD' . "\t" . '*HASH_VALUE' . "\t" . $_sheet_id_column . "\t";
 175                                         foreach ($ent['columns'] as $value)
 176                                         {
 177                                                 $content .= $value . "\t";
 178                                         }
 179                                         $content = mb_substr($content, 0, -1);
 180                                         $content .= "\n";
 181                                 }
 182                                 continue;
 183                         }
 184
 185                         if (isset($ent['command']))
 186                                 $content .= $ent['command'] . "\t";
 187                         if (isset($ent['hash_value']))
 188                                 $content .= $ent['hash_value'] . "\t";
 189                         else
 190                                 $content .= '_0000000000000000' . "\t";
 191                         $content .= $ent['identifier'] . "\t";
 192                         foreach ($ent['columns'] as $value)
 193                         {
 194                                 $content .= $value . "\t";
 195                         }
 196                         $content = mb_substr($content, 0, -1);
 197                         $content .= "\n";
 198                 }
 199                 return mb_convert_encoding($this->addBOM($this->CRLF($content)), 'UTF-16LE', 'UTF-8');
 200 //              return mb_convert_encoding($this->CRLF($content), 'UTF-16LE', 'UTF-8');
 201         }
 202 }
 203 ?>