tools/flexcat/contrib/cd2po/cd2po.sh

   1 #!/bin/bash
   2 #
   3 # a shell script that allows to convert between Amiga-style catalog
   4 # description/translation files (.cd/.ct) and gettext-style translation
   5 # files (.pot/.po).
   6 #
   7 # Copyright 2013-2014 Jens Maus <mail@jens-maus.de>
   8 #
   9 # This program is free software; you can redistribute it and/or modify
  10 # it under the terms of the GNU General Public License as published by
  11 # the Free Software Foundation; either version 2 of the License, or
  12 # (at your option) any later version.
  13 #
  14 # This program is distributed in the hope that it will be useful,
  15 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  16 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  17 # GNU General Public License for more details.
  18 #
  19 # You should have received a copy of the GNU General Public License
  20 # along with this program; if not, write to the Free Software
  21 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  22 #
  23 # $Id$
  24 #
  25
  26 VERSION="1.6"
  27
  28 ########################################################
  29 # Script starts here
  30 #
  31
  32 displayUsage()
  33 {
  34   echo >&2 "cd2po.sh v${VERSION} - convert between Amiga-style and gettext translation files"
  35   echo >&2 "Copyright (c) 2013-2014 Jens Maus <mail@jens-maus.de>"
  36   echo >&2
  37   echo >&2 "Usage: $0 <options> [inputfile (.cd/.ct/.pot/.po)]"
  38   echo >&2 "Options:"
  39   echo >&2 "  -c <charset> : use <charset> when converting the input file"
  40   echo >&2 "                 default: iso-8859-1"
  41 }
  42
  43 ################################
  44 # AWK scripts                  #
  45 ################################
  46
  47 # the following is an awk script that converts an
  48 # Amiga-style catalog description file (.cd) to a gettext
  49 # PO-style translation template file (.pot).
  50 read -d '' cd2pot << 'EOF'
  51 BEGIN {
  52   tagfound=0
  53   firsttag=0
  54   multiline=0
  55
  56   # get current date/time
  57   cmd="date +'%Y-%m-%d %H:%M%z'"
  58   cmd | getline date
  59
  60   print "# Translation catalog description file (pot-style)"
  61   print "# $Id$"
  62   print "#"
  63   print "# WARNING: This file was automatically generated by cd2po.sh"
  64   print "#"
  65 }
  66 {
  67   if(firsttag == 0)
  68   {
  69     if($0 ~ /^#version .*/)
  70     {
  71       version=$2
  72       next
  73     }
  74     else if($0 ~ /^#language .*/)
  75     {
  76       language=$2
  77       next
  78     }
  79   }
  80
  81   if($0 ~ /^MSG_.*\(.*\)/)
  82   {
  83     if(tagfound == 1)
  84     {
  85       # this is the end of the current
  86       # tag so lets output it in PO-format
  87       print ""
  88       #print "#: " msgcomment
  89       if(length(comment) > 0)
  90       {
  91         print comment
  92       }
  93       print "msgctxt \\"" msgctxt "\\""
  94       print "msgid " msgid
  95       print "msgstr \\"\\""
  96     }
  97
  98     tagfound=1
  99
 100     if(firsttag == 0)
 101     {
 102       print "# version " version
 103       print "# language " language
 104       print "#"
 105       print "#, fuzzy"
 106       print "msgid \\"\\""
 107       print "msgstr \\"\\""
 108       print "\\"Project-Id-Version: " version "\\\\n\\""
 109       print "\\"Report-Msgid-Bugs-To: http://URL/\\\\n\\""
 110       print "\\"POT-Creation-Date: " date "\\\\n\\""
 111       print "\\"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\\\\n\\""
 112       print "\\"Last-Translator: FULL NAME <EMAIL@ADDRESS>\\\\n\\""
 113       print "\\"Language-Team: LANGUAGE <LL@li.org>\\\\n\\""
 114       print "\\"MIME-Version: 1.0\\\\n\\""
 115       print "\\"Content-Type: text/plain; charset=UTF-8\\\\n\\""
 116       print "\\"Content-Transfer-Encoding: 8bit\\\\n\\""
 117       print "\\"Language: " language "\\\\n\\""
 118
 119       firsttag=1
 120     }
 121
 122     msgctxt=$0
 123     msgcomment=$1
 124
 125     # proceed with next word
 126     next
 127   }
 128   else if($0 ~ /^;$/)
 129   {
 130     if(tagfound == 1)
 131     {
 132       # this is the end of the current
 133       # tag so lets output it in PO-format
 134       print ""
 135       #print "#: " msgcomment
 136       if(length(comment) > 0)
 137       {
 138         print comment
 139       }
 140       print "msgctxt \\"" msgctxt "\\""
 141       print "msgid " msgid
 142       print "msgstr \\"\\""
 143     }
 144
 145     tagfound=0
 146     multiline=0
 147     comment=""
 148
 149     # proceed with next word
 150     next
 151   }
 152   else if($0 ~ /^;.+/)
 153   {
 154     if(length(comment) > 0)
 155     {
 156       comment = comment "\\n"
 157     }
 158
 159     tmp=substr($0, 2)
 160
 161     # remove any leading white space
 162     gsub(/^ /, "", tmp)
 163
 164     # replace \\\\ by \\
 165     gsub(/\\\\\\\\/, "\\\\", tmp)
 166
 167     comment = comment "#. " tmp
 168     multiline=0
 169
 170     # proceed with next word
 171     next
 172   }
 173
 174   if(tagfound == 1)
 175   {
 176     // remove any backslash at the end of line
 177     gsub(/\\\\$/, "")
 178
 179     # replace \e with \033
 180     gsub(/\\\\\\e/, "\\\\033")
 181
 182     # replace plain " with \" but make
 183     # sure to check if \" is already there
 184     gsub(/\\\\"/, "\\"") # replace \" with "
 185     gsub(/"/, "\\\\\\"") # replace " with \"
 186
 187     # replace \\\\ by \\
 188     gsub(/\\\\\\\\/, "\\\\")
 189
 190     # we have to escape the \033 and other escape
 191     # sequences
 192     gsub(/\\\\0/,  "\\\\\\\\\\\\0")
 193     gsub(/\\\\33/, "\\\\\\\\\\\\033")
 194
 195     if(multiline == 0)
 196     {
 197       # the .po format doesn't allow empty msgid
 198       # strings, thus lets escape them with <EMPTY>
 199       if(length($0) == 0)
 200       {
 201         msgid="\\"<EMPTY>\\""
 202       }
 203       else
 204       {
 205         msgid="\\"" $0 "\\""
 206       }
 207
 208       multiline=1
 209     }
 210     else
 211     {
 212       msgid=msgid "\\n" "\\"" $0 "\\""
 213     }
 214   }
 215 }
 216 EOF
 217
 218 # the following is an awk script that converts a
 219 # gettext PO-style translation template file (.pot)
 220 # to an Amiga-style catalog description file (.cd)
 221 read -d '' pot2cd << 'EOF'
 222 BEGIN {
 223   tagfound=0
 224   firsttag=0
 225   msgidfound=0
 226   print "; Catalog description file (Amiga-cd format)"
 227   print "; $Id$"
 228   print ";"
 229   print "; WARNING: This file was automatically generated by cd2po.sh"
 230   print ";"
 231 }
 232 {
 233   if(firsttag == 0)
 234   {
 235     if($0 ~ /^# version .*/)
 236     {
 237       version=$3
 238       next
 239     }
 240     else if($0 ~ /^# language .*/)
 241     {
 242       language=$3
 243       next
 244     }
 245   }
 246
 247   if($0 ~ /^msgctxt "MSG_.*/)
 248   {
 249     tagfound=1
 250     msgidfound=0
 251
 252     if(firsttag == 0)
 253     {
 254       print "#version " version
 255       print "#language " language
 256       print ";"
 257
 258       firsttag=1
 259     }
 260
 261     # extract the tag "MSG_XXXXX (X//)" as tag
 262     tag=substr($0, length($1)+2)
 263
 264     # strip quotes (") from start&end
 265     gsub(/^"/, "", tag)
 266     gsub(/"$/, "", tag)
 267   }
 268   else if($0 ~ /^#\. .*/)
 269   {
 270     if(length(comment) > 0)
 271     {
 272       comment = comment "\\n"
 273     }
 274
 275     # replace \\033 with \033
 276     gsub(/\\\\\\\\0/, "\\\\0")
 277     gsub(/\\\\\\\\33/, "\\\\033")
 278
 279     # replace \\ by \\\\
 280     gsub(/\\\\\\\\/, "\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\")
 281
 282     comment = comment "; " substr($0, length($1)+2)
 283   }
 284   else if(length($0) == 0 && length(tag) != 0)
 285   {
 286     tagfound=0
 287     msgidfound=0
 288
 289     print tag
 290     print msgid
 291     if(length(comment) > 0)
 292     {
 293       print comment
 294     }
 295     print ";"
 296
 297     tag=""
 298     comment=""
 299   }
 300
 301   if(tagfound == 1)
 302   {
 303     if($0 ~ /^msgid ".*/)
 304     {
 305       # get the msgid text only
 306       tmp=substr($0, length($1)+2)
 307
 308       # strip quotes (") from start&end
 309       gsub(/^"/, "", tmp)
 310       gsub(/"$/, "", tmp)
 311
 312       # replace \\033 with \033
 313       gsub(/\\\\\\\\0/, "\\\\0", tmp)
 314       gsub(/\\\\\\\\33/, "\\\\033", tmp)
 315
 316       # replace \\ by \\\\
 317       gsub(/\\\\\\\\/, "\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\", tmp)
 318
 319       if(length(tmp) > 0)
 320       {
 321         # replace "<EMPTY>" with ""
 322         gsub(/<EMPTY>.*/, "", tmp)
 323         msgid = tmp
 324       }
 325       else
 326       {
 327         msgid=""
 328       }
 329
 330       msgidfound=1
 331     }
 332     else if($0 ~ /^msgstr ".*/)
 333     {
 334       # ignore msgstr
 335       msgidfound=0
 336     }
 337     else if(msgidfound == 1)
 338     {
 339       # strip quotes (") from start&end
 340       gsub(/^"/, "")
 341       gsub(/"$/, "")
 342
 343       # replace \\033 with \033
 344       gsub(/\\\\\\\\0/, "\\\\0")
 345       gsub(/\\\\\\\\33/, "\\\\033")
 346
 347       # replace \\ by \\\\
 348       gsub(/\\\\\\\\/, "\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\")
 349
 350       if(length(msgid) > 0)
 351       {
 352         msgid = msgid "\\\\\\n" $0
 353       }
 354       else
 355       {
 356         msgid = $0
 357       }
 358     }
 359   }
 360 }
 361 END {
 362   if(length(tag) != 0)
 363   {
 364     print tag
 365     print msgid
 366     if(length(comment) > 0)
 367     {
 368       print comment
 369     }
 370     print ";"
 371   }
 372 }
 373 EOF
 374
 375 # the following is an awk script that converts an
 376 # Amiga-style catalog translation file (.ct) to a
 377 # gettext PO-style translation file (.po)
 378 read -d '' ct2po << 'EOF'
 379 BEGIN {
 380   tagfound=0
 381   firsttag=0
 382   multiline=0
 383
 384   # get current date/time
 385   cmd="date +'%Y-%m-%d %H:%M%z'"
 386   cmd | getline date
 387
 388   print "# Catalog translation file (po-style)"
 389   print "# $Id$"
 390   print "#"
 391   print "# WARNING: This file was automatically generated by cd2po.sh"
 392   print "#"
 393 }
 394 {
 395   if(firsttag == 0)
 396   {
 397     if($0 ~ /^## version .*/)
 398     {
 399       version=substr($0, length($1)+length($2)+3)
 400       next
 401     }
 402     else if($0 ~ /^## language .*/)
 403     {
 404       language=$3
 405       next
 406     }
 407     else if($0 ~ /^## codeset .*/)
 408     {
 409       codeset=$3
 410       next
 411     }
 412     else if($0 ~ /^## chunk AUTH .*/)
 413     {
 414       auth=substr($0, length($1)+length($2)+length($3)+4)
 415       next
 416     }
 417   }
 418
 419   if($0 ~ /^MSG_.*$/)
 420   {
 421     tagfound=1
 422     multiline=0
 423     msgctxt=""
 424     msgid=""
 425     comment=""
 426
 427     if(firsttag == 0)
 428     {
 429       print "# version " version
 430       print "# language " language
 431       print "# codeset " codeset
 432       print "# chunk AUTH " auth
 433       print "#"
 434       print "# Translators:"
 435       print "msgid \\"\\""
 436       print "msgstr \\"\\""
 437       print "\\"Project-Id-Version: " version "\\\\n\\""
 438       print "\\"Report-Msgid-Bugs-To: http://URL/\\\\n\\""
 439       print "\\"POT-Creation-Date: " date "\\\\n\\""
 440       print "\\"PO-Revision-Date: " date "\\\\n\\""
 441       print "\\"Last-Translator: " auth "\\\\n\\""
 442       print "\\"Language-Team: " language "\\\\n\\""
 443       print "\\"MIME-Version: 1.0\\\\n\\""
 444       print "\\"Content-Type: text/plain; charset=UTF-8\\\\n\\""
 445       print "\\"Content-Transfer-Encoding: 8bit\\\\n\\""
 446       print "\\"Language: " language "\\\\n\\""
 447
 448       firsttag=1
 449     }
 450
 451     # now we have to search in the CD file for the same string
 452     cmd="sed -n '/^" $1 " (/,/^;$/p' *.cd"
 453     while((cmd | getline output) > 0)
 454     {
 455       if(output ~ /^MSG_.*$/)
 456       {
 457         msgctxt=output
 458       }
 459       else if(output ~ /^;.+$/)
 460       {
 461         # replace \\\\ by \\
 462         gsub(/\\\\\\\\/, "\\\\", output)
 463
 464         if(length(comment) > 0)
 465         {
 466           comment = comment "\\n"
 467         }
 468
 469         tmp=substr(output, 2)
 470         gsub(/^ /, "", tmp)
 471         comment = comment "#. " tmp
 472       }
 473       else if(output ~ /^;$/)
 474       {
 475         # nothing
 476       }
 477       else if(length(msgctxt) > 0)
 478       {
 479         # remove any backslash at the end of line
 480         gsub(/\\\\$/, "", output)
 481
 482         # replace \e with \033
 483         gsub(/\\\\\\e/, "\\\\033", output)
 484
 485         # replace plain " with \" but make
 486         # sure to check if \" is already there
 487         gsub(/\\\\"/, "\\"", output) # replace \" with "
 488         gsub(/"/, "\\\\\\"", output) # replace " with \"
 489
 490         # replace \\\\ by \\
 491         gsub(/\\\\\\\\/, "\\\\", output)
 492
 493         # replace \033 with \\033
 494         gsub(/\\\\0/, "\\\\\\\\\\\\0", output)
 495         gsub(/\\\\33/, "\\\\\\\\\\\\033", output)
 496
 497         if(length(msgid) > 0)
 498         {
 499           msgid = msgid "\\n"
 500         }
 501
 502         msgid = msgid "\\"" output "\\""
 503       }
 504     }
 505     close(cmd)
 506
 507     if(length(msgctxt) == 0)
 508     {
 509       tagfound=0
 510     }
 511
 512     next
 513   }
 514   else if($0 ~ /^;.*$/)
 515   {
 516     if(tagfound == 1)
 517     {
 518       # output the stuff
 519       print ""
 520       if(length(comment) > 0)
 521       {
 522         print comment
 523       }
 524       print "msgctxt \\"" msgctxt "\\""
 525
 526       # the .po format doesn't allow empty msgid
 527       # strings, thus lets escape them with <EMPTY>
 528       if(length(msgid) <= 2)
 529       {
 530         print "msgid \\"<EMPTY>\\""
 531       }
 532       else
 533       {
 534         # find out if this msgid is a multiline msgid or not
 535         if(msgid ~ /\\n/)
 536         {
 537           print "msgid \\"\\""
 538           print msgid
 539         }
 540         else
 541         {
 542           print "msgid " msgid
 543         }
 544       }
 545
 546       print "msgstr \\"" msgstr "\\""
 547
 548       tagfound=0
 549       multiline=0
 550     }
 551
 552     next
 553   }
 554
 555   if(tagfound == 1)
 556   {
 557     # remove any backslash at the end of line
 558     gsub(/\\\\$/, "")
 559
 560     # replace \e with \033
 561     gsub(/\\\\\\e/, "\\\\033")
 562
 563     # replace plain " with \" but make
 564     # sure to check if \" is already there
 565     gsub(/\\\\"/, "\\"") # replace \" with "
 566     gsub(/"/, "\\\\\\"") # replace " with \"
 567
 568     # replace \033 with \\033
 569     gsub(/\\\\0/, "\\\\\\\\\\\\0")
 570     gsub(/\\\\33/, "\\\\\\\\\\\\033")
 571
 572     if(multiline == 0)
 573     {
 574       msgstr = $0
 575       multiline=1
 576     }
 577     else
 578     {
 579       msgstr = msgstr $0
 580     }
 581   }
 582 }
 583 EOF
 584
 585 # the following is an awk script that converts a
 586 # gettext PO-style translation file (.po) to an
 587 # Amiga-style catalog translation file (.ct).
 588 read -d '' po2ct << 'EOF'
 589 BEGIN {
 590   tagfound=0
 591   firsttag=0
 592   msgidfound=0
 593   msgstrfound=0
 594 }
 595 {
 596   if(firsttag == 0)
 597   {
 598     if($0 ~ /^#? version .*/)
 599     {
 600       version=substr($0, length($1)+length($2)+3)
 601       next
 602     }
 603     else if($0 ~ /^# \\$Id: .* \\$$/)
 604     {
 605       revision=$4 # get revision out of $Id$ SVN keyword
 606       next
 607     }
 608     else if($0 ~ /^## language .*/)
 609     {
 610       language=$3
 611       next
 612     }
 613     else if($0 ~ /^## codeset .*/)
 614     {
 615       codeset=$3
 616       next
 617     }
 618     else if($0 ~ /^## chunk AUTH .*/)
 619     {
 620       auth=substr($0, length($1)+length($2)+length($3)+4)
 621       next
 622     }
 623     else if($0 ~ /^"PO-Revision-Date: .*"/)
 624     {
 625       revdate=substr($0, length($1)+2)
 626       gsub(/\\\\n"/, "", revdate);
 627
 628       # parse the revision date
 629       cmd="date +'%d.%m.%Y' -d \\"" revdate "\\""
 630       cmd | getline revdate
 631
 632       next
 633     }
 634     else if($0 ~ /^"Language: .*"/)
 635     {
 636       language=substr($0, length($1)+2)
 637       gsub(/\\\\n"/, "", language);
 638       next
 639     }
 640     else if($0 ~ /^"Language-Team: .*"/)
 641     {
 642       auth=substr($0, length($1)+2)
 643       gsub(/\\\\n"/, "", auth);
 644       next
 645     }
 646   }
 647
 648   if($0 ~ /^msgctxt "MSG_.*/)
 649   {
 650     tagfound=1
 651     msgidfound=0
 652     msgstrfound=0
 653     msgid=""
 654     msgstr=""
 655
 656     if(firsttag == 0)
 657     {
 658       print "## version $VER: XXXX.catalog " version "." revision " (" revdate ")"
 659       print "## language " lang
 660       print "## codeset " cset
 661       print "## chunk AUTH " auth
 662       print ";"
 663       print "; $Id$"
 664       print ";"
 665
 666       firsttag=1
 667     }
 668
 669     # strip quotes (") so that we get the plain MSG_XXXX
 670     # tag names
 671     gsub(/"/, "", $2);
 672     tag=$2
 673   }
 674   else if(length($0) == 0 && length(tag) != 0)
 675   {
 676     tagfound=0
 677     msgidfound=0
 678     msgstrfound=0
 679
 680     if(length(msgstr) > 0)
 681     {
 682       print tag
 683       print msgstr
 684       print msgid
 685       print ";"
 686     }
 687
 688     tag=""
 689   }
 690
 691   if(tagfound == 1)
 692   {
 693     if($0 ~ /^msgid ".*/)
 694     {
 695       # get the msgid text only
 696       tmp=substr($0, length($1)+2)
 697
 698       # strip quotes (") from start&end
 699       gsub(/^"/, "", tmp)
 700       gsub(/"$/, "", tmp)
 701
 702       # replace \\033 with \033
 703       gsub(/\\\\\\\\0/, "\\\\0", tmp)
 704       gsub(/\\\\\\\\33/, "\\\\033", tmp)
 705
 706       if(length(tmp) > 0)
 707       {
 708         if(length(msgid) > 0)
 709         {
 710           msgid = msgid "\\\\\\n; " tmp
 711         }
 712         else
 713         {
 714           msgid = "; " tmp
 715         }
 716       }
 717
 718       msgstrfound=0
 719       msgidfound=1
 720     }
 721     else if($0 ~ /^msgstr ".*/)
 722     {
 723       # get the msgid text only
 724       tmp=substr($0, length($1)+2)
 725
 726       # strip quotes (") from start&end
 727       gsub(/^"/, "", tmp)
 728       gsub(/"$/, "", tmp)
 729
 730       # replace \\033 with \033
 731       gsub(/\\\\\\\\0/, "\\\\0", tmp)
 732       gsub(/\\\\\\\\33/, "\\\\033", tmp)
 733
 734       if(length(tmp) > 0)
 735       {
 736         # replace "<EMPTY>" with ""
 737         gsub(/<EMPTY>.*/, "", tmp)
 738
 739         if(length(msgstr) > 0)
 740         {
 741           msgstr = msgstr "\\\\\\n" tmp
 742         }
 743         else
 744         {
 745           msgstr = tmp
 746         }
 747       }
 748
 749       msgstrfound=1
 750       msgidfound=0
 751     }
 752     else if(msgidfound == 1)
 753     {
 754       # strip quotes (") from start&end
 755       gsub(/^"/, "")
 756       gsub(/"$/, "")
 757
 758       # replace \\033 with \033
 759       gsub(/\\\\\\\\0/, "\\\\0")
 760       gsub(/\\\\\\\\33/, "\\\\033")
 761
 762       if(length($0) > 0)
 763       {
 764         if(length(msgid) > 0)
 765         {
 766           msgid = msgid "\\\\\\n; " $0
 767         }
 768         else
 769         {
 770           msgid = "; " $0
 771         }
 772       }
 773     }
 774     else if(msgstrfound == 1)
 775     {
 776       # strip quotes (") from start&end
 777       gsub(/^"/, "")
 778       gsub(/"$/, "")
 779
 780       # replace \\033 with \033
 781       gsub(/\\\\\\\\0/, "\\\\0")
 782       gsub(/\\\\\\\\33/, "\\\\033")
 783
 784       if(length($0) > 0)
 785       {
 786         if(length(msgstr) > 0)
 787         {
 788           msgstr = msgstr "\\\\\\n" $0
 789         }
 790         else
 791         {
 792           msgstr = $0
 793         }
 794       }
 795     }
 796   }
 797 }
 798 END {
 799   if(length(tag) != 0 && length(msgstr) > 0)
 800   {
 801     print tag
 802     print msgstr
 803     print msgid
 804     print ";"
 805   }
 806 }
 807 EOF
 808
 809 ###################################################
 810 identifyCharset()
 811 {
 812   file="$1"
 813   charset=""
 814
 815   case "${file}" in
 816     bosnian)
 817       charset="iso-8859-2"
 818     ;;
 819     catalan)
 820       charset="iso-8859-15"
 821     ;;
 822     croatian)
 823       charset="iso-8859-16"
 824     ;;
 825     czech)
 826       charset="iso-8859-2"
 827     ;;
 828     danish)
 829       charset="iso-8859-15"
 830     ;;
 831     dutch)
 832       charset="iso-8859-15"
 833     ;;
 834     finnish)
 835       charset="iso-8859-15"
 836     ;;
 837     french)
 838       charset="iso-8859-15"
 839     ;;
 840     german)
 841       charset="iso-8859-15"
 842     ;;
 843     greek)
 844       charset="iso-8859-7"
 845     ;;
 846     hungarian)
 847       charset="iso-8859-16"
 848     ;;
 849     italian)
 850       charset="iso-8859-15"
 851     ;;
 852     norwegian)
 853       charset="iso-8859-15"
 854     ;;
 855     polish)
 856       charset="iso-8859-16"
 857     ;;
 858     russian)
 859       charset="windows-1251" # this should be "Amiga-1251" but iconv doesn't support it :(
 860     ;;
 861     serbian)
 862       charset="iso-8859-16"
 863     ;;
 864     slovenian)
 865       charset="iso-8859-2"
 866     ;;
 867     spanish)
 868       charset="iso-8859-15"
 869     ;;
 870     swedish)
 871       charset="iso-8859-15"
 872     ;;
 873     turkish)
 874       charset="iso-8859-9"
 875     ;;
 876     *)
 877       charset="iso-8859-1"
 878     ;;
 879   esac
 880
 881   echo ${charset}
 882 }
 883
 884 identifyCodeset()
 885 {
 886   file="$1"
 887   codeset=""
 888
 889   case "${file}" in
 890     bosnian)
 891       codeset="5"
 892     ;;
 893     catalan)
 894       codeset="111"
 895     ;;
 896     croatian)
 897       codeset="112"
 898     ;;
 899     czech)
 900       codeset="5"
 901     ;;
 902     danish)
 903       codeset="111"
 904     ;;
 905     dutch)
 906       codeset="111"
 907     ;;
 908     finnish)
 909       codeset="111"
 910     ;;
 911     french)
 912       codeset="111"
 913     ;;
 914     german)
 915       codeset="111"
 916     ;;
 917     greek)
 918       codeset="10"
 919     ;;
 920     hungarian)
 921       codeset="112"
 922     ;;
 923     italian)
 924       codeset="111"
 925     ;;
 926     norwegian)
 927       codeset="111"
 928     ;;
 929     persian)
 930       codeset="0"
 931     ;;
 932     polish)
 933       codeset="112"
 934     ;;
 935     russian)
 936       codeset="2104" # 'Amiga-1251'
 937     ;;
 938     serbian)
 939       codeset="112"
 940     ;;
 941     slovenian)
 942       codeset="5"
 943     ;;
 944     spanish)
 945       codeset="111"
 946     ;;
 947     swedish)
 948       codeset="111"
 949     ;;
 950     turkish)
 951       codeset="12"
 952     ;;
 953     *)
 954       codeset="4"
 955     ;;
 956   esac
 957
 958   echo ${codeset}
 959 }
 960
 961 identifyLanguage()
 962 {
 963   file="$1"
 964   language=""
 965
 966   case "${file}" in
 967     bosnian)
 968       language="bosanski"
 969     ;;
 970     catalan)
 971       language="català"
 972     ;;
 973     croatian)
 974       language="hrvatski"
 975     ;;
 976     danish)
 977       language="dansk"
 978     ;;
 979     dutch)
 980       language="nederlands"
 981     ;;
 982     english-british)
 983       language="english-british"
 984     ;;
 985     finnish)
 986       language="suomi"
 987     ;;
 988     french)
 989       language="français"
 990     ;;
 991     german)
 992       language="deutsch"
 993     ;;
 994     hungarian)
 995       language="magyar"
 996     ;;
 997     italian)
 998       language="italiano"
 999     ;;
1000     japanese)
1001       language="nihongo"
1002     ;;
1003     korean)
1004       language="hangul"
1005     ;;
1006     norwegian)
1007       language="norsk"
1008     ;;
1009     persian)
1010       language="farsi"
1011     ;;
1012     polish)
1013       language="polski"
1014     ;;
1015     portuguese)
1016       language="português"
1017     ;;
1018     portuguese-brazil)
1019       language="português-brasil"
1020     ;;
1021     serbian)
1022       language="srpski"
1023     ;;
1024     slovenian)
1025       language="slovensko"
1026     ;;
1027     spanish)
1028       language="español"
1029     ;;
1030     swedish)
1031       language="svenska"
1032     ;;
1033     turkish)
1034       language="türkçe"
1035     ;;
1036     *)
1037       language=${file}
1038     ;;
1039   esac
1040
1041   echo ${language}
1042 }
1043
1044
1045 ###################################################
1046 charset=""
1047 inputfile="$1"
1048
1049 # parse the command-line options
1050 while getopts "c:" opt
1051 do
1052   case "$opt" in
1053     c)  charset="$OPTARG";;
1054     \?)   # unknown flag
1055       displayUsage
1056       exit 2;;
1057   esac
1058 done
1059 shift `expr $OPTIND - 1`
1060
1061 if [ -z "${inputfile}" ]; then
1062   displayUsage
1063   exit 2
1064 fi
1065
1066 # lets identify by the file extension which operation to perform
1067 fname=$(basename "${inputfile}")
1068 filename="${fname%.*}"
1069 extension="${fname##*.}"
1070 case "${extension}" in
1071   cd) # convert from cd -> pot
1072     if [ -z "${charset}" ]; then
1073       charset="iso-8859-1"
1074     fi
1075     iconv -c -f "${charset}" -t utf8 ${inputfile} | awk "${cd2pot}"
1076   ;;
1077   ct) # convert from ct -> po
1078     if [ -z "${charset}" ]; then
1079       charset=$(identifyCharset ${filename})
1080     fi
1081     iconv -c -f ${charset} -t utf8 ${inputfile} | awk "${ct2po}"
1082   ;;
1083   po) # convert from po -> ct
1084     if [ -z "${charset}" ]; then
1085       charset=$(identifyCharset ${filename})
1086       codeset=$(identifyCodeset ${filename})
1087     fi
1088     lang=$(identifyLanguage ${filename})
1089     awk -v lang=${lang} -v cset=${codeset} "${po2ct}" ${inputfile} | iconv -c -f utf8 -t ${charset}
1090   ;;
1091   pot) # convert from pot -> cd
1092     if [ -z "${charset}" ]; then
1093       charset="iso-8859-1"
1094     fi
1095     awk "${pot2cd}" ${inputfile} | iconv -c -f utf8 -t ${charset}
1096   ;;
1097 esac
1098
1099 exit 0