tools/flexcat/src/scanpo.c

   1 /*
   2  * $Id$
   3  *
   4  * Copyright (C) 1993-1999 Jochen Wiedmann and Marcin Orlowski
   5  * Copyright (C) 2002-2017 FlexCat Open Source Team
   6  *
   7  * This program is free software; you can redistribute it and/or modify
   8  * it under the terms of the GNU General Public License as published by
   9  * the Free Software Foundation; either version 2 of the License, or (at
  10  * your option) any later version.
  11  *
  12  * This program is distributed in the hope that it will be useful, but
  13  * WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  15  * General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU General Public License
  18  * along with this program; if not, write to the Free Software
  19  * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  20  *
  21  */
  22
  23 #ifdef AMIGA
  24   #include <proto/locale.h>  /* This is to get locale.library/IsAlpha() */
  25 #endif
  26
  27 #include <errno.h>
  28 #include <limits.h>
  29 #include <time.h>
  30
  31 #include "flexcat.h"
  32 #include "showfuncs.h"
  33 #include "readprefs.h"
  34 #include "globals.h"
  35 #include "utils.h"
  36 #include "createcat.h"
  37
  38 extern char *CatVersionString;
  39 extern char *CatLanguage;
  40 extern char *CatRcsId;
  41 extern char *CatName;
  42 extern int   CodeSet;
  43 extern int   CT_Scanned;
  44
  45 #define IS_NUMBER_OR_LETTER(c) (((c) >= '0' && (c) <= '9') || \
  46                                 ((c) >= 'a' && (c) <= 'z') || \
  47                                 ((c) >= 'A' && (c) <= 'Z'))
  48
  49 #if defined(__amigaos3__) || defined(__MORPHOS__) || defined(WIN32)
  50 char *strptime(const char *string, const char *fmt, struct tm *res);
  51 #endif
  52
  53 /// ScanCTFile
  54
  55 /* This function scans a PO-style format catalog description/translation file.
  56
  57    Inputs: pofile - name of the description/translation file to scan.
  58    Result: TRUE if successful, FALSE otherwise.
  59 */
  60 int ScanPOFile(char *pofile, int verwarning)
  61 {
  62   FILE *fp;
  63   char *readLineBuffer;
  64   int Result = TRUE;
  65   int CodeSet_checked = FALSE;
  66   int inHeader = TRUE;
  67   int NextID = 0;
  68   const char *PoSrcCharset = "utf-8";
  69   const char *CatDstCharset = "iso-8859-1";
  70   char CatVersionDate[255] = "";
  71   char CatProjectName[255] = "";
  72   struct CatString *cs = NULL;
  73   struct CatString **csptr = &FirstCatString;
  74   int inMsgID = FALSE;
  75   int inMsgSTR = FALSE;
  76
  77   ScanFile = pofile;
  78   ScanLine = 0;
  79   if((fp = fopen(pofile, "r")) == NULL)
  80     ShowErrorQuick(MSG_ERR_NOCATALOGTRANSLATION, pofile);
  81
  82   if(!NoBufferedIO)
  83     setvbuf(fp, NULL, _IOFBF, buffer_size);
  84
  85   // initialize "readLineBuffer" ahead of the loop
  86   // the loop will bail out early for empty files
  87   readLineBuffer = NULL;
  88   while(!feof(fp) && (readLineBuffer = ReadLine(fp, TRUE)) != NULL)
  89   {
  90     // the buffer pointer will be modified all the way down, so better work with a copy,
  91     // otherwise the free() call after the loop will free the wrong pointer
  92     char *line = readLineBuffer;
  93
  94     if(inHeader == TRUE)
  95     {
  96       if(line[0] == '\0')
  97       {
  98         inHeader = FALSE;
  99
 100         // check that we have a valid verson
 101         if(CatVersion == -1)
 102         {
 103           if(verwarning)
 104             ShowWarn(MSG_ERR_NO_CAT_VERSION);
 105
 106           CatVersion = 0;
 107         }
 108
 109         // we found the end of the header so lets check if we have all
 110         // we require to continue
 111         if(CatVersionDate[0] != '\0' && CatProjectName[0] != '\0' &&
 112            CatVersionString == NULL)
 113         {
 114           char buf[255];
 115
 116           // warn about missing revision information
 117           if(CatRevision == -1)
 118           {
 119             if(verwarning)
 120               ShowWarn(MSG_ERR_NO_CAT_REVISION);
 121
 122             CatRevision = 0;
 123           }
 124
 125           if(strstr(CatProjectName, ".catalog") != NULL)
 126             snprintf(buf, sizeof(buf), "$VER: %s %d.%d (%s)", CatProjectName, CatVersion, CatRevision, CatVersionDate);
 127           else
 128             snprintf(buf, sizeof(buf), "$VER: %s.catalog %d.%d (%s)", CatProjectName, CatVersion, CatRevision, CatVersionDate);
 129
 130           CatVersionString = AllocString(buf);
 131         }
 132       }
 133       else switch(*line)
 134       {
 135         case '#':
 136         {
 137           // comment lines start with #
 138           // but they may contain some valueable information for catalog
 139           // file creation. So lets parse these lines as well
 140           while(*line == '#' || *line == ' ' || *line == '\t')
 141             ++line;
 142
 143           if(CatVersion == -1 && Strnicmp(line, "version", 7) == 0)
 144           {
 145             line += 8;
 146             OverSpace(&line);
 147             CatVersion = strtol(line, &line, 0);
 148           }
 149           else if(CatRevision == -1 && Strnicmp(line, "revision", 8) == 0)
 150           {
 151             line += 9;
 152             OverSpace(&line);
 153             CatRevision = strtol(line, &line, 0);
 154           }
 155           else if(CatRevision == -1 &&
 156                   Strnicmp(line, "$Id: ", 5) == 0)
 157           {
 158             char *p;
 159
 160             line += 6;
 161             p = line;
 162
 163             // search second space
 164             p = strchr(p, ' ');
 165             if(p != NULL)
 166             {
 167               p++;
 168               CatRevision = strtol(p, &p, 0);
 169             }
 170           }
 171           else if(CatRevision == -1 &&
 172                   Strnicmp(line, "$Revision: ", 11) == 0)
 173           {
 174             line += 12;
 175             CatRevision = strtol(line, &line, 0);
 176           }
 177         }
 178         break;
 179
 180         case '"':
 181         {
 182           if(Strnicmp(line, "\"Language: ", 11) == 0)
 183           {
 184             char *p;
 185             const char *language = NULL;
 186
 187             if(CatLanguage)
 188               ShowError(MSG_ERR_DOUBLECTLANGUAGE);
 189
 190             line += 11;
 191             p = strchr(line, '\\');
 192             if(p != NULL)
 193               *p = '\0';
 194
 195             if(Stricmp(line, "bs") == 0) // bosnian
 196             {
 197               language = "bosanski";
 198               CatDstCharset = "iso-8859-2";
 199             }
 200             else if(Stricmp(line, "ca") == 0) // catalan
 201             {
 202               language = "catal\xE0";
 203               CatDstCharset = "iso-8859-15";
 204             }
 205             else if(Stricmp(line, "hr") == 0) // croatian
 206             {
 207               language = "hrvatski";
 208               CatDstCharset = "iso-8859-16";
 209             }
 210             else if(Stricmp(line, "cs") == 0) // czech
 211             {
 212               language = "czech";
 213               CatDstCharset = "iso-8859-2";
 214             }
 215             else if(Stricmp(line, "da") == 0) // danish
 216             {
 217               language = "dansk";
 218               CatDstCharset = "iso-8859-15";
 219             }
 220             else if(Stricmp(line, "nl") == 0) // dutch
 221             {
 222               language = "nederlands";
 223               CatDstCharset = "iso-8859-15";
 224             }
 225             else if(Stricmp(line, "en_GB") == 0) // english-british
 226               language = "english-british";
 227             else if(Stricmp(line, "fi") == 0) // finnish
 228             {
 229               language = "suomi";
 230               CatDstCharset = "iso-8859-15";
 231             }
 232             else if(Stricmp(line, "fr") == 0) // french
 233             {
 234               language = "fran""\xE7""ais";
 235               CatDstCharset = "iso-8859-15";
 236             }
 237             else if(Stricmp(line, "de") == 0) // german
 238             {
 239               language = "deutsch";
 240               CatDstCharset = "iso-8859-15";
 241             }
 242             else if(Stricmp(line, "el") == 0) // greek
 243             {
 244               language = "greek";
 245               CatDstCharset = "iso-8859-7";
 246             }
 247             else if(Stricmp(line, "hu") == 0) // hungarian
 248             {
 249               language = "magyar";
 250               CatDstCharset = "iso-8859-16";
 251             }
 252             else if(Stricmp(line, "it") == 0) // italian
 253             {
 254               language = "italiano";
 255               CatDstCharset = "iso-8859-15";
 256             }
 257             else if(Stricmp(line, "ja") == 0) // japanese
 258             {
 259               language = "nihongo";
 260               CatDstCharset = "euc-jp";
 261             }
 262             else if(Stricmp(line, "ko") == 0) // korean
 263             {
 264               language = "hangul";
 265               CatDstCharset = "euc-kr";
 266             }
 267             else if(Stricmp(line, "no") == 0) // norwegian
 268             {
 269               language = "norsk";
 270               CatDstCharset = "iso-8859-15";
 271             }
 272             else if(Stricmp(line, "fa") == 0) // persian
 273             {
 274               language = "farsi";
 275               CatDstCharset = "utf-8";
 276             }
 277             else if(Stricmp(line, "pl") == 0) // polish
 278             {
 279               language = "polski";
 280               CatDstCharset = "iso-8859-2";
 281             }
 282             else if(Stricmp(line, "pt") == 0) // portuguese
 283               language = "portugu""\xEA""s";
 284             else if(Stricmp(line, "pt_BR") == 0) // portuguese-brazil
 285               language = "portugu""\xEA""s-brasil";
 286             else if(Stricmp(line, "ru") == 0) // russian
 287             {
 288               language = "russian";
 289               #if defined(AMIGA)
 290               CatDstCharset = "Amiga-1251";
 291               #else
 292               CatDstCharset = "windows-1251"; // iconv doesn't know anything about Amiga-1251 :(
 293               #endif
 294             }
 295             else if(Stricmp(line, "sr") == 0) // serbian
 296             {
 297               language = "srpski";
 298               CatDstCharset = "iso-8859-16";
 299             }
 300             else if(Stricmp(line, "sk") == 0) // slovakian
 301             {
 302               language = "slovak";
 303               CatDstCharset = "iso-8859-2";
 304             }
 305             else if(Stricmp(line, "sl") == 0) // slovenian
 306             {
 307               language = "slovensko";
 308               CatDstCharset = "iso-8859-2";
 309             }
 310             else if(Stricmp(line, "es") == 0) // spanish
 311             {
 312               language = "espa""\xF1""ol";
 313               CatDstCharset = "iso-8859-15";
 314             }
 315             else if(Stricmp(line, "sv") == 0) // swedish
 316             {
 317               language = "svenska";
 318               CatDstCharset = "iso-8859-15";
 319             }
 320             else if(Stricmp(line, "tr") == 0) // turkish
 321             {
 322               language = "t""\xFC""rk""\xE7""e";
 323               CatDstCharset = "iso-8859-9";
 324             }
 325
 326             if(language != NULL)
 327               CatLanguage = AddCatalogChunk(strdup("LANG"), language);
 328           }
 329           else if(Strnicmp(line, "\"Language-Team: ", 16) == 0)
 330           {
 331             char *p;
 332
 333             line += 16;
 334             p = strchr(line, '\\');
 335             if(p != NULL)
 336               *p = '\0';
 337
 338             AddCatalogChunk(strdup("AUTH"), line);
 339           }
 340           else if(CodeSet_checked == FALSE &&
 341                   Strnicmp(line, "\"Content-Type: ", 15) == 0)
 342           {
 343             char *p;
 344
 345             line += 15;
 346             p = strstr(line, "charset=");
 347             if(p != NULL)
 348             {
 349               char *q;
 350
 351               p += 8;
 352
 353               q = strchr(p, '\\');
 354               if(q != NULL)
 355                 *q = '\0';
 356
 357               PoSrcCharset = strdup(p);
 358             }
 359
 360             CodeSet_checked = TRUE;
 361           }
 362           else if(Strnicmp(line, "\"PO-Revision-Date: ", 19) == 0)
 363           {
 364             struct tm tm;
 365
 366             line += 19;
 367             memset(&tm, 0, sizeof(tm));
 368             strptime(line, "%Y-%m-%d", &tm);
 369             strftime(CatVersionDate, sizeof(CatVersionDate), "%d.%m.%Y", &tm);
 370           }
 371           else if(Strnicmp(line, "\"Catalog-Name: ", 15) == 0)
 372           {
 373             char *p;
 374
 375             line += 15;
 376             p = strchr(line, '\\');
 377             if(p != NULL)
 378               *p = '\0';
 379
 380             strcpy(CatProjectName, line);
 381           }
 382           else if(Strnicmp(line, "\"Project-Id-Version: ", 21) == 0 && CatProjectName[0] == '\0')
 383           {
 384             // fall back to the project ID as catalog name if it is not yet defined
 385             char *p;
 386
 387             line += 21;
 388             p = strchr(line, '\\');
 389             if(p != NULL)
 390               *p = '\0';
 391
 392             strcpy(CatProjectName, line);
 393           }
 394         }
 395         break;
 396       }
 397     }
 398     else
 399     {
 400       // check if we found a line starting with "msgctxt" as that signals us
 401       // a new catalog string should be added
 402       if(Strnicmp(line, "msgctxt \"", 9) == 0)
 403       {
 404         char *idstr;
 405
 406         // we found a new 'msgctxt' lets clear cs
 407         cs = NULL;
 408         inMsgID = FALSE;
 409         inMsgSTR = FALSE;
 410
 411         line += 9;
 412
 413         /* Check for blanks at the start of line. */
 414         if(*line == ' ' || *line == '\t')
 415         {
 416           ShowError(MSG_ERR_UNEXPECTEDBLANKS);
 417           OverSpace(&line);
 418         }
 419
 420         idstr = line;
 421         while(IS_NUMBER_OR_LETTER(*line) || *line == '_')
 422           ++line;
 423
 424         if(idstr == line)
 425         {
 426           ShowError(MSG_ERR_NOIDENTIFIER);
 427           Result = FALSE;
 428         }
 429         else
 430         {
 431           int found;
 432
 433           if((cs = malloc(sizeof(*cs))) == NULL)
 434             MemError();
 435
 436           // search for the next catstring ID in case the ID
 437           // specifier is missing "(//)" in the msgctxt
 438           do
 439           {
 440             struct CatString *scs;
 441
 442             found = TRUE;
 443             for(scs = FirstCatString; scs != NULL; scs = scs->Next)
 444             {
 445               if(scs->ID == NextID)
 446               {
 447                 found = FALSE;
 448                 ++NextID;
 449                 break;
 450               }
 451             }
 452           }
 453           while(found == FALSE);
 454
 455           cs->Next = NULL;
 456           cs->ID = NextID;
 457           cs->MinLen = 0;
 458           cs->MaxLen = -1;
 459           cs->CD_Str = (char *)"";
 460           cs->CT_Str = NULL;
 461           cs->NotInCT = TRUE;
 462           cs->POformat = TRUE;
 463
 464           if((cs->ID_Str = malloc((line - idstr) + 1)) == NULL)
 465             MemError();
 466
 467           strncpy(cs->ID_Str, idstr, line - idstr);
 468           cs->ID_Str[line - idstr] = '\0';
 469           OverSpace(&line);
 470
 471           /* Check if next char in line is '('? (//) */
 472           if(*line != '(')
 473           {
 474             ShowError(MSG_ERR_NO_LEADING_BRACKET, cs->ID_Str);
 475             Result = FALSE;
 476           }
 477           else
 478           {
 479             struct CatString *scs;
 480
 481             ++line;
 482             OverSpace(&line);
 483
 484             /* Check for default config of line (//) */
 485             if(*line != '/')
 486             {
 487               if(*line == '+')
 488                 NextID = cs->ID = NextID + strtol(line, &line, 0);
 489               else if(*line == '$')
 490               {
 491                 line++;
 492                 cs->ID = NextID = strtol(line, &line, 16);
 493               }
 494               else
 495                 cs->ID = NextID = strtol(line, &line, 0);
 496
 497               OverSpace(&line);
 498             }
 499
 500             /* Check for already used identifier. */
 501             for(scs = FirstCatString; scs != NULL; scs = scs->Next)
 502             {
 503               if(scs->ID == cs->ID)
 504               {
 505                 ShowError(MSG_ERR_DOUBLE_ID, cs->ID_Str);
 506                 Result = FALSE;
 507               }
 508               if(strcmp(cs->ID_Str, scs->ID_Str) == 0)
 509               {
 510                 ShowError(MSG_ERR_DOUBLE_IDENTIFIER, cs->ID_Str);
 511                 Result = FALSE;
 512               }
 513             }
 514
 515             /* Check for min/len values (//) */
 516             if(*line != '/')
 517             {
 518               ShowWarn(MSG_ERR_NO_MIN_LEN, cs->ID_Str);
 519               Result = FALSE;
 520             }
 521             else
 522             {
 523               ++line;
 524               OverSpace(&line);
 525               if(*line != '/')
 526               {
 527                 cs->MinLen = strtol(line, &line, 0);
 528                 OverSpace(&line);
 529               }
 530               if(*line != '/')
 531               {
 532                 ShowWarn(MSG_ERR_NO_MAX_LEN, cs->ID_Str);
 533                 Result = FALSE;
 534               }
 535               else
 536               {
 537                 ++line;
 538                 OverSpace(&line);
 539                 if(*line != ')')
 540                 {
 541                   cs->MaxLen = strtol(line, &line, 0);
 542                   OverSpace(&line);
 543                 }
 544                 if(*line != ')')
 545                 {
 546                   ShowError(MSG_ERR_NO_TRAILING_BRACKET, cs->ID_Str);
 547                   Result = FALSE;
 548                 }
 549                 else
 550                 {
 551                   ++line;
 552                   OverSpace(&line);
 553                   if(*line && *line != '\"')
 554                     ShowError(MSG_ERR_EXTRA_CHARACTERS_ID, cs->ID_Str);
 555                 }
 556               }
 557             }
 558           }
 559
 560           //printf("ID_Str: '%s' (%d)\n", cs->ID_Str, cs->ID);
 561
 562           cs->Nr = NumStrings;
 563           cs->LenBytes = 0;
 564           *csptr = cs;
 565           csptr = &cs->Next;
 566           ++NumStrings;
 567         }
 568       }
 569       else if(cs != NULL)
 570       {
 571         char *p;
 572
 573         // if the user want to force a certain output (destination)
 574         // codeset we set it here.
 575         if(DestCodeset[0] != '\0')
 576           CatDstCharset = DestCodeset;
 577
 578         // Make sure double backslashes end up in a single backslash.
 579         // We catch any double backslash followed by a zero character,
 580         // which covers strings like "\\0" and "\\033" or "\\33" as these are
 581         // common strings in MUI applications.
 582         while((p = strstr(line, "\\\\0")) != NULL || (p = strstr(line, "\\\\33")) != NULL)
 583           memmove(p, p+1, strlen(p));
 584
 585         // unquote the string
 586         if(line[0] != '\0' && line[strlen(line)-1] == '"')
 587           line[strlen(line)-1] = '\0';
 588
 589         if(Strnicmp(line, "msgid \"", 7) == 0)
 590         {
 591           line += 7;
 592
 593           // if the string starts with <EMPTY> we out to remove
 594           // the rest of the string!
 595           if(strncmp(line, "<EMPTY>", 7) == 0)
 596             line[0] = '\0';
 597
 598           if(strlen(line) > 0)
 599           {
 600             if((cs->CD_Str = ConvertString(line, PoSrcCharset, CatDstCharset)) == NULL)
 601               ShowWarn(MSG_ERR_CONVERSION_FAILED, cs->ID_Str);
 602           }
 603           else
 604           {
 605             cs->CD_Str = malloc(1);
 606             cs->CD_Str[0] = '\0';
 607           }
 608
 609           //printf("CD_Str: '%s' '%s'\n", cs->CD_Str, line);
 610
 611           inMsgID = TRUE;
 612           inMsgSTR = FALSE;
 613         }
 614         else if(Strnicmp(line, "msgstr \"", 8) == 0)
 615         {
 616           line += 8;
 617
 618           // don't search for "<EMPTY>" here, this will be done later
 619           // during all other checks. If the string would be erased here
 620           // we would be no longer able to tell it apart from really
 621           // missing translations.
 622           if(strlen(line) > 0)
 623           {
 624             if((cs->CT_Str = ConvertString(line, PoSrcCharset, CatDstCharset)) == NULL)
 625               ShowWarn(MSG_ERR_CONVERSION_FAILED, cs->ID_Str);
 626           }
 627           else
 628           {
 629             cs->CT_Str = malloc(1);
 630             cs->CT_Str[0] = '\0';
 631           }
 632
 633           cs->NotInCT = FALSE;
 634
 635           //printf("CT_Str: '%s'\n", cs->CT_Str);
 636
 637           inMsgSTR = TRUE;
 638           inMsgID = FALSE;
 639         }
 640         else if(*line == '"') // line starts with "
 641         {
 642           line++;
 643
 644           if(inMsgID == TRUE)
 645           {
 646             char *t;
 647
 648             if((t = ConvertString(line, PoSrcCharset, CatDstCharset)) == NULL)
 649               ShowWarn(MSG_ERR_CONVERSION_FAILED, cs->ID_Str);
 650             else
 651               cs->CD_Str = AddString(cs->CD_Str, t);
 652
 653             //printf("CD_Str2: '%s' '%s'\n", cs->CD_Str, line);
 654
 655             free(t);
 656           }
 657           else if(inMsgSTR == TRUE)
 658           {
 659             char *t;
 660
 661             if((t = ConvertString(line, PoSrcCharset, CatDstCharset)) == NULL)
 662               ShowWarn(MSG_ERR_CONVERSION_FAILED, cs->ID_Str);
 663             else
 664               cs->CT_Str = AddString(cs->CT_Str, t);
 665
 666             //printf("CT_Str2: '%s' '%s'\n", cs->CT_Str, line);
 667
 668             free(t);
 669           }
 670         }
 671       }
 672     }
 673   }
 674
 675   free(readLineBuffer);
 676   fclose(fp);
 677
 678   /*
 679   printf("CatVersion: %d.%d\n", CatVersion, CatRevision);
 680   printf("CatVersionDate: '%s'\n", CatVersionDate);
 681   printf("CatVersionString: '%s'\n", CatVersionString);
 682   printf("CatLanguage: '%s'\n", CatLanguage);
 683   printf("PoSrcCharset: '%s'\n", PoSrcCharset);
 684   printf("CatDstCharset: '%s'\n", CatDstCharset);
 685   */
 686
 687   if(!CodeSet_checked)
 688     ShowErrorQuick(MSG_ERR_NOCTCODESET);
 689
 690   if(!(CatVersionString || (CatRcsId && CatName)))
 691     ShowErrorQuick(MSG_ERR_NOCTVERSION);
 692
 693   // lets translate CatDstCharset to CodeSet number
 694   if(Stricmp(CatDstCharset, "iso-8859-1") == 0)
 695     CodeSet = 4;
 696   else if(Stricmp(CatDstCharset, "iso-8859-2") == 0)
 697     CodeSet = 5;
 698   else if(Stricmp(CatDstCharset, "iso-8859-7") == 0)
 699     CodeSet = 10;
 700   else if(Stricmp(CatDstCharset, "iso-8859-9") == 0)
 701     CodeSet = 12;
 702   else if(Stricmp(CatDstCharset, "utf-8") == 0 || Stricmp(CatDstCharset, "utf8") == 0)
 703     CodeSet = 106;
 704   else if(Stricmp(CatDstCharset, "iso-8859-15") == 0)
 705     CodeSet = 111;
 706   else if(Stricmp(CatDstCharset, "iso-8859-16") == 0)
 707     CodeSet = 112;
 708   else if(Stricmp(CatDstCharset, "amiga-1251") == 0 || Stricmp(CatDstCharset, "windows-1251") == 0)
 709     CodeSet = 2104;
 710   else
 711     CodeSet = 0;
 712
 713   // check consistenty of translations found
 714   for(cs = FirstCatString; cs != NULL; cs = cs->Next)
 715   {
 716     if(cs->CT_Str == NULL)
 717       ShowWarnQuick(MSG_ERR_MISSINGTRANSLATION, cs->ID_Str);
 718     else
 719     {
 720       size_t reallen;
 721       size_t reallen_utf8;
 722       size_t cd_len;
 723
 724       // get string length for both ASCII and UTF8 encoding
 725       // the length check must be done against the UTF8 length,
 726       // which might be less than the ASCII length due to certain
 727       // UTF8 characters which are encoded with up to 3 ASCII
 728       // characters
 729       reallen = strlen(cs->CT_Str);
 730       reallen_utf8 = utf8_strlen(cs->CT_Str);
 731       cd_len = strlen(cs->CD_Str);
 732
 733       // check for empty translations
 734       if(cd_len > 0)
 735       {
 736         if(reallen == 0)
 737         {
 738           // for .po files empty strings are really missing translations
 739           ShowWarnQuick(MSG_ERR_MISSINGTRANSLATION, cs->ID_Str);
 740
 741           // now remove the cs from the list
 742           cs->NotInCT = TRUE;
 743           continue;
 744         }
 745       }
 746
 747       // check for intentionally empty translations
 748       if(strncmp(cs->CT_Str, "<EMPTY>", 7) == 0)
 749         cs->CT_Str[0] = '\0';
 750
 751       if(cs->MinLen > 0 && reallen_utf8 < (size_t)cs->MinLen)
 752         ShowWarnQuick(MSG_ERR_STRING_TOO_SHORT, cs->ID_Str);
 753
 754       if(cs->MaxLen > 0 && reallen_utf8 > (size_t)cs->MaxLen)
 755         ShowWarnQuick(MSG_ERR_STRING_TOO_LONG, cs->ID_Str);
 756
 757       // check for trailing ellipsis
 758       if(reallen >= 3 && cd_len >= 3)
 759       {
 760         if(strcmp(&cs->CD_Str[cd_len - 3], "...") == 0 &&
 761            strcmp(&cs->CT_Str[reallen - 3], "...") != 0)
 762         {
 763           ShowWarnQuick(MSG_ERR_TRAILING_ELLIPSIS, cs->ID_Str);
 764         }
 765
 766         if(strcmp(&cs->CD_Str[cd_len - 3], "...") != 0 &&
 767            strcmp(&cs->CT_Str[reallen - 3], "...") == 0)
 768         {
 769           ShowWarnQuick(MSG_ERR_NO_TRAILING_ELLIPSIS, cs->ID_Str);
 770         }
 771       }
 772
 773       // check for trailing spaces
 774       if(reallen >= 1 && cd_len >= 1)
 775       {
 776         if(strcmp(&cs->CD_Str[cd_len - 1], " ") == 0 &&
 777            strcmp(&cs->CT_Str[reallen - 1], " ") != 0)
 778
 779         {
 780           ShowWarnQuick(MSG_ERR_TRAILING_BLANKS, cs->ID_Str);
 781         }
 782
 783         if(strcmp(&cs->CD_Str[cd_len - 1], " ") != 0 &&
 784            strcmp(&cs->CT_Str[reallen - 1], " ") == 0)
 785
 786         {
 787           ShowWarnQuick(MSG_ERR_NO_TRAILING_BLANKS, cs->ID_Str);
 788         }
 789       }
 790
 791       // check for matching placeholders
 792       if(reallen >= 1 && cd_len >= 1)
 793       {
 794         char *cdP = cs->CD_Str;
 795         char *ctP = cs->CT_Str;
 796
 797         do
 798         {
 799           cdP = strchr(cdP, '%');
 800           ctP = strchr(ctP, '%');
 801
 802           if(cdP == NULL && ctP == NULL)
 803           {
 804             // no more placeholders, bail out
 805             break;
 806           }
 807           else if(cdP != NULL && ctP != NULL)
 808           {
 809             // skip the '%' sign
 810             cdP++;
 811             ctP++;
 812
 813             // check the placeholder only if the '%' is followed by an
 814             // alpha-numerical character or another percent sign
 815             if(IS_NUMBER_OR_LETTER(*cdP) || *cdP == '%')
 816             {
 817               if(*cdP != *ctP)
 818               {
 819                 ShowWarnQuick(MSG_ERR_MISMATCHING_PLACEHOLDERS, cs->ID_Str);
 820
 821                 break;
 822               }
 823
 824               // skip the second '%' sign
 825               if(*cdP == '%')
 826                   cdP++;
 827               if(*ctP == '%')
 828                   ctP++;
 829             }
 830             else if(IS_NUMBER_OR_LETTER(*ctP) || *ctP == '%')
 831             {
 832               // the translation uses a placeholder while the description
 833               // uses none.
 834               ShowWarnQuick(MSG_ERR_EXCESSIVE_PLACEHOLDERS, cs->ID_Str);
 835
 836               break;
 837             }
 838           }
 839           else if(cdP != NULL && ctP == NULL)
 840           {
 841             // skip the '%' sign
 842             cdP++;
 843
 844             // check if really a placeholder follows or just another percent sign
 845             // the original string is allowed to contain more single percent signs than the translated string
 846             if(IS_NUMBER_OR_LETTER(*cdP) || *cdP == '%')
 847             {
 848               // the description uses at least one more placeholder than the translation
 849               ShowWarnQuick(MSG_ERR_MISSING_PLACEHOLDERS, cs->ID_Str);
 850             }
 851
 852             break;
 853           }
 854           else if(cdP == NULL && ctP != NULL)
 855           {
 856             // skip the '%' sign
 857             ctP++;
 858
 859             // check if really a placeholder follows or just another percent sign
 860             // the translated string is allowed to contain more single percent signs than the original string
 861             if(IS_NUMBER_OR_LETTER(*ctP) || *ctP == '%')
 862             {
 863               // the translation uses at least one more placeholder than the description
 864               ShowWarnQuick(MSG_ERR_EXCESSIVE_PLACEHOLDERS, cs->ID_Str);
 865             }
 866
 867             break;
 868           }
 869         }
 870         while(TRUE);
 871       }
 872     }
 873   }
 874
 875   if(WarnCTGaps)
 876   {
 877     for(cs = FirstCatString; cs != NULL; cs = cs->Next)
 878     {
 879       if(cs->CT_Str == NULL)
 880       {
 881         ShowWarn(MSG_ERR_CTGAP, cs->ID_Str);
 882       }
 883     }
 884   }
 885
 886   if(Result)
 887     CT_Scanned = TRUE;
 888
 889   return(Result);
 890 }
 891
 892 ///
 893