3 # Open \______ \ ____ ____ | | _\_ |__ _______ ___
4 # Source | _// _ \_/ ___\| |/ /| __ \ / _ \ \/ /
5 # Jukebox | | ( <_> ) \___| < | \_\ ( <_> > < <
6 # Firmware |____|_ /\____/ \___ >__|_ \|___ /\____/__/\_ \
10 # Copyright (C) 2006 by Daniel Stenberg
13 # binary version for the binary lang file
14 my $langversion = 3; # 3 was the latest one used in the v1 format
16 # A note for future users and readers: The original v1 language system allowed
17 # the build to create and use a different language than english built-in. We
18 # removed that feature from our build-system, but the build scripts still had
19 # the ability. But, starting now, this ability is no longer provided since I
20 # figured it was boring and unnecessary to write support for now since we
21 # don't use it anymore.
25 Usage: genlang [options] <langv2 file>
28 Make the tool create a [prefix].c and [prefix].h file.
31 Make the tool create a binary language (.lng) file namaed [outfile].
32 The use of this option requires that you also use -e.
35 Update language file. Given the translated file and the most recent english
36 file, you\'ll get an updated version sent to stdout. Suitable action to do
37 when you intend to update a translation.
39 -e=<english lang file>
40 Point out the english (original source) file, to use that as master
41 language template. Used in combination with -b or -u.
44 Specify which target you want the translations/phrases for. Required when
47 The target can in fact be specified as numerous different strings,
48 separated with colons. This will make genlang to use all the specified
49 strings when searching for a matching phrase.
52 Voice mode output. Outputs all id: and voice: lines for the given target!
55 Enables verbose (debug) output.
63 # 1) scan the english file, keep the whole <phrase> for each phrase.
64 # 2) read the translated file, for each end of phrase, compare:
65 # A) all source strings, if there's any change there should be a comment about
69 # 3) output the phrase with the comments from above
70 # 4) check which phrases that the translated version didn't have, and spit out
71 # the english version of those
81 my $check = ($binary?
1:0) + ($prefix?
1:0) + ($update?
1:0) + ($voiceout?
1:0);
84 print "Please use only one of -p, -u, -o and -b\n";
88 print "Please use at least one of -p, -u, -o and -b\n";
91 if(($binary || $update || $voiceout) && !$english) {
92 print "Please use -e too when you use -b, -o or -u\n";
97 if(!$target && !$update) {
98 print "Please specify a target (with -t)!\n";
103 my %id; # string to num hash
104 my @idnum; # num to string array
106 my %source; # id string to source phrase hash
107 my %dest; # id string to dest phrase hash
108 my %voice; # id string to voice phrase hash
110 my $input = $ARGV[0];
116 my ($string, $pattern)=@_;
118 $pattern =~ s/\*/.?*/g;
119 $pattern =~ s/\?/./g;
121 return ($string =~ $pattern);
130 my ($full, $n, $v)=@_;
136 my ($full, $n, $v)=@_;
141 my ($debug, $strref, $full, $n, $v)=@_;
143 my @all= split(" *, *", $n);
146 # print "TEST ($debug) $target for $test\n";
147 for my $part (split(":", $target)) {
148 if(match
($part, $test)) {
150 # print "MATCH: $test => $v\n";
160 parsetarget
("src", \
$src, @_);
165 parsetarget
("dest", \
$dest, @_);
170 parsetarget
("voice", \
$voice, @_);
176 # For the cases where the english file needs to be scanned/read, we do
177 # it before we read the translated file. For -b it isn't necessary, but for
178 # -u it is convenient.
180 my $idnum=0; # start with a true number
181 my $vidnum=0x8000; # first voice id
182 open(ENG
, "<$english") || die "can't open $english";
187 # get rid of DOS newlines
190 if($_ =~ /^ *\<phrase\>/) {
191 # this is the start of a phrase
193 elsif($_ =~ /^ *\<\/phrase\
>/) {
194 # this is the end of a phrase, add it to the english hash
195 $english{$id}=join("", @phrase);
199 # gather everything related to this phrase
203 if($_ =~ /^ *id: ([^ \t\n]+)/i) {
205 # voice-only entries get a difference range
206 if($id =~ /^VOICE_/) {
207 # Assign an ID number to this entry
212 # Assign an ID number to this entry
221 # a function that compares the english phrase with the translated one.
222 # compare source strings and desc
224 # Then output the updated version!
226 my ($idstr, $engref, $locref)=@_;
228 my ($esource, $lsource);
231 for my $l (@
$engref) {
232 if($l =~ /^ *desc: (.*)/) {
235 elsif($l =~ / *\<source\>/i) {
239 if($l =~ / *\<\/source\
>/i
) {
250 for my $l (@
$locref) {
251 if($l =~ /^ *desc: (.*)/) {
253 if($edesc ne $ldesc) {
254 $l = "### The 'desc' field differs from the english!\n### the previously used desc is commented below:\n### desc: $ldesc\n desc: $edesc\n";
258 elsif($l =~ / *\<source\>/i) {
263 if($l =~ / *\<\/source\
>/i
) {
266 if($esource ne $lsource) {
267 print "### The <source> section differs from the english!\n",
268 "### the previously used one is commented below:\n";
269 for(split("\n", $lsource)) {
277 undef @show; # start over
294 my $idcount; # counter for lang ID numbers
295 my $voiceid=0x8000; # counter for voice-only ID numbers
298 # Now start the scanning of the selected language string
301 open(LANG
, "<$input") || die "couldn't read language file named $input\n";
307 # get rid of DOS newlines
310 if($_ =~ /^( *\#|[ \t\n\r]*\z)/) {
311 # comment or empty line
321 # this is an XML-lookalike tag
322 if (/^(<|[^\"<]+<)([^>]*)>/) {
324 # print "P: $part\n";
327 # this was a closing tag
329 if($part eq "/phrase") {
332 my $idstr = $phrase{'id'};
335 if($dest =~ /^none\z/i) {
336 # "none" as dest (without quotes) means that this entire
337 # phrase is to be ignored
340 # allow the keyword 'deprecated' to be used on dest and
341 # voice strings to mark that as deprecated. It will then
342 # be replaced with "".
344 $dest =~ s/^deprecate(|d)\z/\"\"/i;
345 $voice =~ s/^deprecate(|d)\z/\"\"/i;
347 # Use the ID name to figure out which id number range we
348 # should use for this phrase. Voice-only strings are
351 # basic syntax error alerts, if there are no quotes we
352 # will assume an empty string was intended
354 print STDERR
"Warning: dest before line $line lacks quotes!\n";
358 print STDERR
"Warning: source before line $line lacks quotes!\n";
361 if($voice !~ /^\"/) {
362 print STDERR
"Warning: voice before line $line lacks quotes!\n";
366 if($idstr =~ /^VOICE/) {
373 $id{$idstr} = $idnum;
374 $idnum[$idnum]=$idstr;
376 $source{$idstr}=$src;
378 $voice{$idstr}=$voice;
381 print "id: $phrase{id} ($idnum)\n";
382 print "source: $src\n";
383 print "dest: $dest\n";
384 print "voice: $voice\n";
394 my $e = $english{$idstr};
397 # compare original english with this!
398 my @eng = split("\n", $english{$idstr});
400 compare
($idstr, \
@eng, \
@phrase);
402 $english{$idstr}=""; # clear it
405 print "### $idstr: The phrase is not used. Skipped\n";
412 # starts with a slash, this _ends_ this section
413 $m = pop @m; # get back old value, the previous level's tag
417 # This is an opening (sub) tag
419 push @m, $m; # store old value
424 if(/^ *([^:]+): *(.*)/) {
425 my ($name, $val)=($1, $2);
426 &$m($_, $name, $val);
436 "### This phrase below was not present in the translated file\n",
445 # We create a .c and .h file
447 open(HFILE
, ">$prefix.h") ||
448 die "couldn't create file $prefix.h\n";
449 open(CFILE
, ">$prefix.c") ||
450 die "couldn't create file $prefix.c\n";
453 /* This file was automatically generated using genlang */
455 * The str() macro/functions is how to access strings that might be
456 * translated. Use it like str(MACRO) and expect a string to be
459 #define str(x) language_strings[x]
461 /* this is the array for holding the string pointers.
462 It will be initialized at runtime. */
463 extern unsigned char *language_strings[];
464 /* this contains the concatenation of all strings, separated by \\0 chars */
465 extern const unsigned char language_builtin[];
467 /* The enum below contains all available strings */
473 /* This file was automaticly generated using genlang, the strings come
478 unsigned char *language_strings[LANG_LAST_INDEX_IN_ARRAY];
479 const unsigned char language_builtin[] =
483 # Output the ID names for the enum in the header file
485 for $i (1 .. $idcount) {
486 my $name=$idnum[$i - 1]; # get the ID name
488 $name =~ s/\"//g; # cut off the quotes
490 printf HFILE
(" %s,\n", $name);
493 # Output separation marker for last string ID and the upcoming voice IDs
496 LANG_LAST_INDEX_IN_ARRAY, /* this is not a string, this is a marker */
497 /* --- below this follows voice-only strings --- */
498 VOICEONLY_DELIMITER = 0x8000,
502 # Output the ID names for the enum in the header file
503 for $i (0x8000 .. ($voiceid-1)) {
504 my $name=$idnum[$i]; # get the ID name
506 $name =~ s/\"//g; # cut off the quotes
508 printf HFILE
(" %s,\n", $name);
512 print HFILE
"\n};\n/* end of generated enum list */\n";
514 # Output the target phrases for the source file
515 for $i (1 .. $idcount) {
516 my $name=$idnum[$i - 1]; # get the ID
517 my $dest = $dest{$name}; # get the destination phrase
519 $dest =~ s
:\"$:\\0\":; # insert a \0 before the second quote
522 # this is just to be on the safe side
526 printf CFILE
(" %s\n", $dest);
529 # Output end of string chunk
532 /* end of generated string list */
538 } # end of the c/h file generation
540 # Creation of a binary lang file was requested
542 # We must first scan the english file to get the correct order of the id
543 # numbers used there, as that is what sets the id order for all language
544 # files. The english file is scanned before the translated file was
547 open(OUTF
, ">$binary") or die "Can't create $binary";
549 printf OUTF
("\x1a%c", $langversion); # magic lang file header
551 # loop over the target phrases
552 for $i (1 .. $idcount) {
553 my $name=$idnum[$i - 1]; # get the ID
554 my $dest = $dest{$name}; # get the destination phrase
557 $dest =~ s/^\"(.*)\"\s*$/$1/g; # cut off quotes
559 # Now, make sure we get the number from the english sort order:
560 $idnum = $idmap{$name};
562 printf OUTF
("%c%c%s\x00", ($idnum>>8), ($idnum&0xff), $dest);
567 # voice output requested, display id: and voice: strings in a v1-like
572 # This loops over the strings in the translated language file order
573 my @ids = ((0 .. ($idcount-1)));
574 push @ids, (0x8000 .. ($voiceid-1));
581 my $name=$idnum[$i]; # get the ID
582 my $dest = $voice{$name}; # get the destination voice string
585 $dest =~ s/^\"(.*)\"\s*$/$1/g; # cut off quotes
587 # Now, make sure we get the number from the english sort order:
588 $idnum = $idmap{$name};
592 # print "Input index $i output index $idnum\n";
600 my $name=$idnum[$o]; # get the ID
601 my $dest = $voice{$name}; # get the destination voice string
603 print "#$i\nid: $name\nvoice: $dest\n";
610 printf("%d ID strings scanned\n", $idcount);
614 printf "$_: %s\n", $head{$_};