/js/jsgettext/resources/Gettext.js
JavaScript | 1265 lines | 937 code | 41 blank | 287 comment | 81 complexity | 946e49b92105ab018e1aefb782122c42 MD5 | raw file
1/* 2Pure Javascript implementation of Uniforum message translation. 3Copyright (C) 2008 Joshua I. Miller <unrtst@cpan.org>, all rights reserved 4 5This program is free software; you can redistribute it and/or modify it 6under the terms of the GNU Library General Public License as published 7by the Free Software Foundation; either version 2, or (at your option) 8any later version. 9 10This program is distributed in the hope that it will be useful, 11but WITHOUT ANY WARRANTY; without even the implied warranty of 12MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 13Library General Public License for more details. 14 15You should have received a copy of the GNU Library General Public 16License along with this program; if not, write to the Free Software 17Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, 18USA. 19 20=head1 NAME 21 22Javascript Gettext - Javascript implemenation of GNU Gettext API. 23 24=head1 SYNOPSIS 25 26 // ////////////////////////////////////////////////////////// 27 // Optimum caching way 28 <script language="javascript" src="/path/LC_MESSAGES/myDomain.json"></script> 29 <script language="javascript" src="/path/Gettext.js'></script> 30 31 // assuming myDomain.json defines variable json_locale_data 32 var params = { "domain" : "myDomain", 33 "locale_data" : json_locale_data 34 }; 35 var gt = new Gettext(params); 36 // create a shortcut if you'd like 37 function _ (msgid) { return gt.gettext(msgid); } 38 alert(_("some string")); 39 // or use fully named method 40 alert(gt.gettext("some string")); 41 // change to use a different "domain" 42 gt.textdomain("anotherDomain"); 43 alert(gt.gettext("some string")); 44 45 46 // ////////////////////////////////////////////////////////// 47 // The other way to load the language lookup is a "link" tag 48 // Downside is that not all browsers cache XMLHttpRequests the 49 // same way, so caching of the language data isn't guarenteed 50 // across page loads. 51 // Upside is that it's easy to specify multiple files 52 <link rel="gettext" href="/path/LC_MESSAGES/myDomain.json" /> 53 <script language="javascript" src="/path/Gettext.js'></script> 54 55 var gt = new Gettext({ "domain" : "myDomain" }); 56 // rest is the same 57 58 59 // ////////////////////////////////////////////////////////// 60 // The reson the shortcuts aren't exported by default is because they'd be 61 // glued to the single domain you created. So, if you're adding i18n support 62 // to some js library, you should use it as so: 63 64 if (typeof(MyNamespace) == 'undefined') MyNamespace = {}; 65 MyNamespace.MyClass = function () { 66 var gtParms = { "domain" : 'MyNamespace_MyClass' }; 67 this.gt = new Gettext(gtParams); 68 return this; 69 }; 70 MyNamespace.MyClass.prototype._ = function (msgid) { 71 return this.gt.gettext(msgid); 72 }; 73 MyNamespace.MyClass.prototype.something = function () { 74 var myString = this._("this will get translated"); 75 }; 76 77 // ////////////////////////////////////////////////////////// 78 // Adding the shortcuts to a global scope is easier. If that's 79 // ok in your app, this is certainly easier. 80 var myGettext = new Gettext({ 'domain' : 'myDomain' }); 81 function _ (msgid) { 82 return myGettext.gettext(msgid); 83 } 84 alert( _("text") ); 85 86 // ////////////////////////////////////////////////////////// 87 // Data structure of the json data 88 // NOTE: if you're loading via the <script> tag, you can only 89 // load one file, but it can contain multiple domains. 90 var json_locale_data = { 91 "MyDomain" : { 92 "" : { 93 "header_key" : "header value", 94 "header_key" : "header value", 95 "msgid" : [ "msgid_plural", "msgstr", "msgstr_plural", "msgstr_pluralN" ], 96 "msgctxt\004msgid" : [ null, "msgstr" ], 97 }, 98 "AnotherDomain" : { 99 }, 100 } 101 102=head1 DESCRIPTION 103 104This is a javascript implementation of GNU Gettext, providing internationalization support for javascript. It differs from existing javascript implementations in that it will support all current Gettext features (ex. plural and context support), and will also support loading language catalogs from .mo, .po, or preprocessed json files (converter included). 105 106The locale initialization differs from that of GNU Gettext / POSIX. Rather than setting the category, domain, and paths, and letting the libs find the right file, you must explicitly load the file at some point. The "domain" will still be honored. Future versions may be expanded to include support for set_locale like features. 107 108 109=head1 INSTALL 110 111To install this module, simply copy the file lib/Gettext.js to a web accessable location, and reference it from your application. 112 113 114=head1 CONFIGURATION 115 116Configure in one of two ways: 117 118=over 119 120=item 1. Optimal. Load language definition from statically defined json data. 121 122 <script language="javascript" src="/path/locale/domain.json"></script> 123 124 // in domain.json 125 json_locale_data = { 126 "mydomain" : { 127 // po header fields 128 "" : { 129 "plural-forms" : "...", 130 "lang" : "en", 131 }, 132 // all the msgid strings and translations 133 "msgid" : [ "msgid_plural", "translation", "plural_translation" ], 134 }, 135 }; 136 // please see the included bin/po2json script for the details on this format 137 138This method also allows you to use unsupported file formats, so long as you can parse them into the above format. 139 140=item 2. Use AJAX to load language file. 141 142Use XMLHttpRequest (actually, SJAX - syncronous) to load an external resource. 143 144Supported external formats are: 145 146=over 147 148=item * Javascript Object Notation (.json) 149 150(see bin/po2json) 151 152 type=application/json 153 154=item * Uniforum Portable Object (.po) 155 156(see GNU Gettext's xgettext) 157 158 type=application/x-po 159 160=item * Machine Object (compiled .po) (.mo) 161 162NOTE: .mo format isn't actually supported just yet, but support is planned. 163 164(see GNU Gettext's msgfmt) 165 166 type=application/x-mo 167 168=back 169 170=back 171 172=head1 METHODS 173 174The following methods are implemented: 175 176 new Gettext(args) 177 textdomain (domain) 178 gettext (msgid) 179 dgettext (domainname, msgid) 180 dcgettext (domainname, msgid, LC_MESSAGES) 181 ngettext (msgid, msgid_plural, count) 182 dngettext (domainname, msgid, msgid_plural, count) 183 dcngettext (domainname, msgid, msgid_plural, count, LC_MESSAGES) 184 pgettext (msgctxt, msgid) 185 dpgettext (domainname, msgctxt, msgid) 186 dcpgettext (domainname, msgctxt, msgid, LC_MESSAGES) 187 npgettext (msgctxt, msgid, msgid_plural, count) 188 dnpgettext (domainname, msgctxt, msgid, msgid_plural, count) 189 dcnpgettext (domainname, msgctxt, msgid, msgid_plural, count, LC_MESSAGES) 190 strargs (string, args_array) 191 192 193=head2 new Gettext (args) 194 195Several methods of loading locale data are included. You may specify a plugin or alternative method of loading data by passing the data in as the "locale_data" option. For example: 196 197 var get_locale_data = function () { 198 // plugin does whatever to populate locale_data 199 return locale_data; 200 }; 201 var gt = new Gettext( 'domain' : 'messages', 202 'locale_data' : get_locale_data() ); 203 204The above can also be used if locale data is specified in a statically included <SCRIPT> tag. Just specify the variable name in the call to new. Ex: 205 206 var gt = new Gettext( 'domain' : 'messages', 207 'locale_data' : json_locale_data_variable ); 208 209Finally, you may load the locale data by referencing it in a <LINK> tag. Simply exclude the 'locale_data' option, and all <LINK rel="gettext" ...> items will be tried. The <LINK> should be specified as: 210 211 <link rel="gettext" type="application/json" href="/path/to/file.json"> 212 <link rel="gettext" type="text/javascript" href="/path/to/file.json"> 213 <link rel="gettext" type="application/x-po" href="/path/to/file.po"> 214 <link rel="gettext" type="application/x-mo" href="/path/to/file.mo"> 215 216args: 217 218=over 219 220=item domain 221 222The Gettext domain, not www.whatev.com. It's usually your applications basename. If the .po file was "myapp.po", this would be "myapp". 223 224=item locale_data 225 226Raw locale data (in json structure). If specified, from_link data will be ignored. 227 228=back 229 230=cut 231 232*/ 233 234Gettext = function (args) { 235 this.domain = 'messages'; 236 // locale_data will be populated from <link...> if not specified in args 237 this.locale_data = undefined; 238 239 // set options 240 var options = [ "domain", "locale_data" ]; 241 if (this.isValidObject(args)) { 242 for (var i in args) { 243 for (var j=0; j<options.length; j++) { 244 if (i == options[j]) { 245 // don't set it if it's null or undefined 246 if (this.isValidObject(args[i])) 247 this[i] = args[i]; 248 } 249 } 250 } 251 } 252 253 254 // try to load the lang file from somewhere 255 this.try_load_lang(); 256 257 return this; 258} 259 260Gettext.context_glue = "\004"; 261Gettext._locale_data = {}; 262 263Gettext.prototype.try_load_lang = function() { 264 // check to see if language is statically included 265 if (typeof(this.locale_data) != 'undefined') { 266 // we're going to reformat it, and overwrite the variable 267 var locale_copy = this.locale_data; 268 this.locale_data = undefined; 269 this.parse_locale_data(locale_copy); 270 271 if (typeof(Gettext._locale_data[this.domain]) == 'undefined') { 272 throw new Error("Error: Gettext 'locale_data' does not contain the domain '"+this.domain+"'"); 273 } 274 } 275 276 277 // try loading from JSON 278 // get lang links 279 var lang_link = this.get_lang_refs(); 280 281 if (typeof(lang_link) == 'object' && lang_link.length > 0) { 282 // NOTE: there will be a delay here, as this is async. 283 // So, any i18n calls made right after page load may not 284 // get translated. 285 // XXX: we may want to see if we can "fix" this behavior 286 for (var i=0; i<lang_link.length; i++) { 287 var link = lang_link[i]; 288 if (link.type == 'application/json') { 289 if (! this.try_load_lang_json(link.href) ) { 290 throw new Error("Error: Gettext 'try_load_lang_json' failed. Unable to exec xmlhttprequest for link ["+link.href+"]"); 291 } 292 } else if (link.type == 'application/x-po') { 293 if (! this.try_load_lang_po(link.href) ) { 294 throw new Error("Error: Gettext 'try_load_lang_po' failed. Unable to exec xmlhttprequest for link ["+link.href+"]"); 295 } 296 } else { 297 // TODO: implement the other types (.mo) 298 throw new Error("TODO: link type ["+link.type+"] found, and support is planned, but not implemented at this time."); 299 } 300 } 301 } 302}; 303 304// This takes the bin/po2json'd data, and moves it into an internal form 305// for use in our lib, and puts it in our object as: 306// Gettext._locale_data = { 307// domain : { 308// head : { headfield : headvalue }, 309// msgs : { 310// msgid : [ msgid_plural, msgstr, msgstr_plural ], 311// }, 312Gettext.prototype.parse_locale_data = function(locale_data) { 313 if (typeof(Gettext._locale_data) == 'undefined') { 314 Gettext._locale_data = { }; 315 } 316 317 // suck in every domain defined in the supplied data 318 for (var domain in locale_data) { 319 // skip empty specs (flexibly) 320 if ((! locale_data.hasOwnProperty(domain)) || (! this.isValidObject(locale_data[domain]))) 321 continue; 322 // skip if it has no msgid's 323 var has_msgids = false; 324 for (var msgid in locale_data[domain]) { 325 has_msgids = true; 326 break; 327 } 328 if (! has_msgids) continue; 329 330 // grab shortcut to data 331 var data = locale_data[domain]; 332 333 // if they specifcy a blank domain, default to "messages" 334 if (domain == "") domain = "messages"; 335 // init the data structure 336 if (! this.isValidObject(Gettext._locale_data[domain]) ) 337 Gettext._locale_data[domain] = { }; 338 if (! this.isValidObject(Gettext._locale_data[domain].head) ) 339 Gettext._locale_data[domain].head = { }; 340 if (! this.isValidObject(Gettext._locale_data[domain].msgs) ) 341 Gettext._locale_data[domain].msgs = { }; 342 343 for (var key in data) { 344 if (key == "") { 345 var header = data[key]; 346 for (var head in header) { 347 var h = head.toLowerCase(); 348 Gettext._locale_data[domain].head[h] = header[head]; 349 } 350 } else { 351 Gettext._locale_data[domain].msgs[key] = data[key]; 352 } 353 } 354 } 355 356 // build the plural forms function 357 for (var domain in Gettext._locale_data) { 358 if (this.isValidObject(Gettext._locale_data[domain].head['plural-forms']) && 359 typeof(Gettext._locale_data[domain].head.plural_func) == 'undefined') { 360 // untaint data 361 var plural_forms = Gettext._locale_data[domain].head['plural-forms']; 362 var pf_re = new RegExp('^(\\s*nplurals\\s*=\\s*[0-9]+\\s*;\\s*plural\\s*=\\s*(?:\\s|[-\\?\\|&=!<>+*/%:;a-zA-Z0-9_\(\)])+)', 'm'); 363 if (pf_re.test(plural_forms)) { 364 //ex english: "Plural-Forms: nplurals=2; plural=(n != 1);\n" 365 //pf = "nplurals=2; plural=(n != 1);"; 366 //ex russian: nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10< =4 && (n%100<10 or n%100>=20) ? 1 : 2) 367 //pf = "nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2)"; 368 369 var pf = Gettext._locale_data[domain].head['plural-forms']; 370 if (! /;\s*$/.test(pf)) pf = pf.concat(';'); 371 /* We used to use eval, but it seems IE has issues with it. 372 * We now use "new Function", though it carries a slightly 373 * bigger performance hit. 374 var code = 'function (n) { var plural; var nplurals; '+pf+' return { "nplural" : nplurals, "plural" : (plural === true ? 1 : plural ? plural : 0) }; };'; 375 Gettext._locale_data[domain].head.plural_func = eval("("+code+")"); 376 */ 377 var code = 'var plural; var nplurals; '+pf+' return { "nplural" : nplurals, "plural" : (plural === true ? 1 : plural ? plural : 0) };'; 378 Gettext._locale_data[domain].head.plural_func = new Function("n", code); 379 } else { 380 throw new Error("Syntax error in language file. Plural-Forms header is invalid ["+plural_forms+"]"); 381 } 382 383 // default to english plural form 384 } else if (typeof(Gettext._locale_data[domain].head.plural_func) == 'undefined') { 385 Gettext._locale_data[domain].head.plural_func = function (n) { 386 var p = (n != 1) ? 1 : 0; 387 return { 'nplural' : 2, 'plural' : p }; 388 }; 389 } // else, plural_func already created 390 } 391 392 return; 393}; 394 395 396// try_load_lang_po : do an ajaxy call to load in the .po lang defs 397Gettext.prototype.try_load_lang_po = function(uri) { 398 var data = this.sjax(uri); 399 if (! data) return; 400 401 var domain = this.uri_basename(uri); 402 var parsed = this.parse_po(data); 403 404 var rv = {}; 405 // munge domain into/outof header 406 if (parsed) { 407 if (! parsed[""]) parsed[""] = {}; 408 if (! parsed[""]["domain"]) parsed[""]["domain"] = domain; 409 domain = parsed[""]["domain"]; 410 rv[domain] = parsed; 411 412 this.parse_locale_data(rv); 413 } 414 415 return 1; 416}; 417 418Gettext.prototype.uri_basename = function(uri) { 419 var rv; 420 if (rv = uri.match(/^(.*\/)?(.*)/)) { 421 var ext_strip; 422 if (ext_strip = rv[2].match(/^(.*)\..+$/)) 423 return ext_strip[1]; 424 else 425 return rv[2]; 426 } else { 427 return ""; 428 } 429}; 430 431Gettext.prototype.parse_po = function(data) { 432 var rv = {}; 433 var buffer = {}; 434 var lastbuffer = ""; 435 var errors = []; 436 var lines = data.split("\n"); 437 for (var i=0; i<lines.length; i++) { 438 // chomp 439 lines[i] = lines[i].replace(/(\n|\r)+$/, ''); 440 441 var match; 442 443 // Empty line / End of an entry. 444 if (/^$/.test(lines[i])) { 445 if (typeof(buffer['msgid']) != 'undefined') { 446 var msg_ctxt_id = (typeof(buffer['msgctxt']) != 'undefined' && 447 buffer['msgctxt'].length) ? 448 buffer['msgctxt']+Gettext.context_glue+buffer['msgid'] : 449 buffer['msgid']; 450 var msgid_plural = (typeof(buffer['msgid_plural']) != 'undefined' && 451 buffer['msgid_plural'].length) ? 452 buffer['msgid_plural'] : 453 null; 454 455 // find msgstr_* translations and push them on 456 var trans = []; 457 for (var str in buffer) { 458 var match; 459 if (match = str.match(/^msgstr_(\d+)/)) 460 trans[parseInt(match[1])] = buffer[str]; 461 } 462 trans.unshift(msgid_plural); 463 464 // only add it if we've got a translation 465 // NOTE: this doesn't conform to msgfmt specs 466 if (trans.length > 1) rv[msg_ctxt_id] = trans; 467 468 buffer = {}; 469 lastbuffer = ""; 470 } 471 472 // comments 473 } else if (/^#/.test(lines[i])) { 474 continue; 475 476 // msgctxt 477 } else if (match = lines[i].match(/^msgctxt\s+(.*)/)) { 478 lastbuffer = 'msgctxt'; 479 buffer[lastbuffer] = this.parse_po_dequote(match[1]); 480 481 // msgid 482 } else if (match = lines[i].match(/^msgid\s+(.*)/)) { 483 lastbuffer = 'msgid'; 484 buffer[lastbuffer] = this.parse_po_dequote(match[1]); 485 486 // msgid_plural 487 } else if (match = lines[i].match(/^msgid_plural\s+(.*)/)) { 488 lastbuffer = 'msgid_plural'; 489 buffer[lastbuffer] = this.parse_po_dequote(match[1]); 490 491 // msgstr 492 } else if (match = lines[i].match(/^msgstr\s+(.*)/)) { 493 lastbuffer = 'msgstr_0'; 494 buffer[lastbuffer] = this.parse_po_dequote(match[1]); 495 496 // msgstr[0] (treak like msgstr) 497 } else if (match = lines[i].match(/^msgstr\[0\]\s+(.*)/)) { 498 lastbuffer = 'msgstr_0'; 499 buffer[lastbuffer] = this.parse_po_dequote(match[1]); 500 501 // msgstr[n] 502 } else if (match = lines[i].match(/^msgstr\[(\d+)\]\s+(.*)/)) { 503 lastbuffer = 'msgstr_'+match[1]; 504 buffer[lastbuffer] = this.parse_po_dequote(match[2]); 505 506 // continued string 507 } else if (/^"/.test(lines[i])) { 508 buffer[lastbuffer] += this.parse_po_dequote(lines[i]); 509 510 // something strange 511 } else { 512 errors.push("Strange line ["+i+"] : "+lines[i]); 513 } 514 } 515 516 517 // handle the final entry 518 if (typeof(buffer['msgid']) != 'undefined') { 519 var msg_ctxt_id = (typeof(buffer['msgctxt']) != 'undefined' && 520 buffer['msgctxt'].length) ? 521 buffer['msgctxt']+Gettext.context_glue+buffer['msgid'] : 522 buffer['msgid']; 523 var msgid_plural = (typeof(buffer['msgid_plural']) != 'undefined' && 524 buffer['msgid_plural'].length) ? 525 buffer['msgid_plural'] : 526 null; 527 528 // find msgstr_* translations and push them on 529 var trans = []; 530 for (var str in buffer) { 531 var match; 532 if (match = str.match(/^msgstr_(\d+)/)) 533 trans[parseInt(match[1])] = buffer[str]; 534 } 535 trans.unshift(msgid_plural); 536 537 // only add it if we've got a translation 538 // NOTE: this doesn't conform to msgfmt specs 539 if (trans.length > 1) rv[msg_ctxt_id] = trans; 540 541 buffer = {}; 542 lastbuffer = ""; 543 } 544 545 546 // parse out the header 547 if (rv[""] && rv[""][1]) { 548 var cur = {}; 549 var hlines = rv[""][1].split(/\\n/); 550 for (var i=0; i<hlines.length; i++) { 551 if (! hlines.length) continue; 552 553 var pos = hlines[i].indexOf(':', 0); 554 if (pos != -1) { 555 var key = hlines[i].substring(0, pos); 556 var val = hlines[i].substring(pos +1); 557 var keylow = key.toLowerCase(); 558 559 if (cur[keylow] && cur[keylow].length) { 560 errors.push("SKIPPING DUPLICATE HEADER LINE: "+hlines[i]); 561 } else if (/#-#-#-#-#/.test(keylow)) { 562 errors.push("SKIPPING ERROR MARKER IN HEADER: "+hlines[i]); 563 } else { 564 // remove begining spaces if any 565 val = val.replace(/^\s+/, ''); 566 cur[keylow] = val; 567 } 568 569 } else { 570 errors.push("PROBLEM LINE IN HEADER: "+hlines[i]); 571 cur[hlines[i]] = ''; 572 } 573 } 574 575 // replace header string with assoc array 576 rv[""] = cur; 577 } else { 578 rv[""] = {}; 579 } 580 581 // TODO: XXX: if there are errors parsing, what do we want to do? 582 // GNU Gettext silently ignores errors. So will we. 583 // alert( "Errors parsing po file:\n" + errors.join("\n") ); 584 585 return rv; 586}; 587 588 589Gettext.prototype.parse_po_dequote = function(str) { 590 var match; 591 if (match = str.match(/^"(.*)"/)) { 592 str = match[1]; 593 } 594 // unescale all embedded quotes (fixes bug #17504) 595 str = str.replace(/\\"/g, "\""); 596 return str; 597}; 598 599 600// try_load_lang_json : do an ajaxy call to load in the lang defs 601Gettext.prototype.try_load_lang_json = function(uri) { 602 var data = this.sjax(uri); 603 if (! data) return; 604 605 var rv = this.JSON(data); 606 this.parse_locale_data(rv); 607 608 return 1; 609}; 610 611// this finds all <link> tags, filters out ones that match our 612// specs, and returns a list of hashes of those 613Gettext.prototype.get_lang_refs = function() { 614 var langs = new Array(); 615 var links = document.getElementsByTagName("link"); 616 // find all <link> tags in dom; filter ours 617 for (var i=0; i<links.length; i++) { 618 if (links[i].rel == 'gettext' && links[i].href) { 619 if (typeof(links[i].type) == 'undefined' || 620 links[i].type == '') { 621 if (/\.json$/i.test(links[i].href)) { 622 links[i].type = 'application/json'; 623 } else if (/\.js$/i.test(links[i].href)) { 624 links[i].type = 'application/json'; 625 } else if (/\.po$/i.test(links[i].href)) { 626 links[i].type = 'application/x-po'; 627 } else if (/\.mo$/i.test(links[i].href)) { 628 links[i].type = 'application/x-mo'; 629 } else { 630 throw new Error("LINK tag with rel=gettext found, but the type and extension are unrecognized."); 631 } 632 } 633 634 links[i].type = links[i].type.toLowerCase(); 635 if (links[i].type == 'application/json') { 636 links[i].type = 'application/json'; 637 } else if (links[i].type == 'text/javascript') { 638 links[i].type = 'application/json'; 639 } else if (links[i].type == 'application/x-po') { 640 links[i].type = 'application/x-po'; 641 } else if (links[i].type == 'application/x-mo') { 642 links[i].type = 'application/x-mo'; 643 } else { 644 throw new Error("LINK tag with rel=gettext found, but the type attribute ["+links[i].type+"] is unrecognized."); 645 } 646 647 langs.push(links[i]); 648 } 649 } 650 return langs; 651}; 652 653 654/* 655 656=head2 textdomain( domain ) 657 658Set domain for future gettext() calls 659 660A message domain is a set of translatable msgid messages. Usually, 661every software package has its own message domain. The domain name is 662used to determine the message catalog where a translation is looked up; 663it must be a non-empty string. 664 665The current message domain is used by the gettext, ngettext, pgettext, 666npgettext functions, and by the dgettext, dcgettext, dngettext, dcngettext, 667dpgettext, dcpgettext, dnpgettext and dcnpgettext functions when called 668with a NULL domainname argument. 669 670If domainname is not NULL, the current message domain is set to 671domainname. 672 673If domainname is undefined, null, or empty string, the function returns 674the current message domain. 675 676If successful, the textdomain function returns the current message 677domain, after possibly changing it. (ie. if you set a new domain, the 678value returned will NOT be the previous domain). 679 680=cut 681 682*/ 683Gettext.prototype.textdomain = function (domain) { 684 if (domain && domain.length) this.domain = domain; 685 return this.domain; 686} 687 688/* 689 690=head2 gettext( MSGID ) 691 692Returns the translation for B<MSGID>. Example: 693 694 alert( gt.gettext("Hello World!\n") ); 695 696If no translation can be found, the unmodified B<MSGID> is returned, 697i. e. the function can I<never> fail, and will I<never> mess up your 698original message. 699 700One common mistake is to interpolate a variable into the string like this: 701 702 var translated = gt.gettext("Hello " + full_name); 703 704The interpolation will happen before it's passed to gettext, and it's 705unlikely you'll have a translation for every "Hello Tom" and "Hello Dick" 706and "Hellow Harry" that may arise. 707 708Use C<strargs()> (see below) to solve this problem: 709 710 var translated = Gettext.strargs( gt.gettext("Hello %1"), [full_name] ); 711 712This is espeically useful when multiple replacements are needed, as they 713may not appear in the same order within the translation. As an English to 714French example: 715 716 Expected result: "This is the red ball" 717 English: "This is the %1 %2" 718 French: "C'est le %2 %1" 719 Code: Gettext.strargs( gt.gettext("This is the %1 %2"), ["red", "ball"] ); 720 721(The example is stupid because neither color nor thing will get 722translated here ...). 723 724=head2 dgettext( TEXTDOMAIN, MSGID ) 725 726Like gettext(), but retrieves the message for the specified 727B<TEXTDOMAIN> instead of the default domain. In case you wonder what 728a textdomain is, see above section on the textdomain() call. 729 730=head2 dcgettext( TEXTDOMAIN, MSGID, CATEGORY ) 731 732Like dgettext() but retrieves the message from the specified B<CATEGORY> 733instead of the default category C<LC_MESSAGES>. 734 735NOTE: the categories are really useless in javascript context. This is 736here for GNU Gettext API compatability. In practice, you'll never need 737to use this. This applies to all the calls including the B<CATEGORY>. 738 739 740=head2 ngettext( MSGID, MSGID_PLURAL, COUNT ) 741 742Retrieves the correct translation for B<COUNT> items. In legacy software 743you will often find something like: 744 745 alert( count + " file(s) deleted.\n" ); 746 747or 748 749 printf(count + " file%s deleted.\n", $count == 1 ? '' : 's'); 750 751I<NOTE: javascript lacks a builtin printf, so the above isn't a working example> 752 753The first example looks awkward, the second will only work in English 754and languages with similar plural rules. Before ngettext() was introduced, 755the best practice for internationalized programs was: 756 757 if (count == 1) { 758 alert( gettext("One file deleted.\n") ); 759 } else { 760 printf( gettext("%d files deleted.\n"), count ); 761 } 762 763This is a nuisance for the programmer and often still not sufficient 764for an adequate translation. Many languages have completely different 765ideas on numerals. Some (French, Italian, ...) treat 0 and 1 alike, 766others make no distinction at all (Japanese, Korean, Chinese, ...), 767others have two or more plural forms (Russian, Latvian, Czech, 768Polish, ...). The solution is: 769 770 printf( ngettext("One file deleted.\n", 771 "%d files deleted.\n", 772 count), // argument to ngettext! 773 count); // argument to printf! 774 775In English, or if no translation can be found, the first argument 776(B<MSGID>) is picked if C<count> is one, the second one otherwise. 777For other languages, the correct plural form (of 1, 2, 3, 4, ...) 778is automatically picked, too. You don't have to know anything about 779the plural rules in the target language, ngettext() will take care 780of that. 781 782This is most of the time sufficient but you will have to prove your 783creativity in cases like 784 785 "%d file(s) deleted, and %d file(s) created.\n" 786 787That said, javascript lacks C<printf()> support. Supplied with Gettext.js 788is the C<strargs()> method, which can be used for these cases: 789 790 Gettext.strargs( gt.ngettext( "One file deleted.\n", 791 "%d files deleted.\n", 792 count), // argument to ngettext! 793 count); // argument to strargs! 794 795NOTE: the variable replacement isn't done for you, so you must 796do it yourself as in the above. 797 798=head2 dngettext( TEXTDOMAIN, MSGID, MSGID_PLURAL, COUNT ) 799 800Like ngettext() but retrieves the translation from the specified 801textdomain instead of the default domain. 802 803=head2 dcngettext( TEXTDOMAIN, MSGID, MSGID_PLURAL, COUNT, CATEGORY ) 804 805Like dngettext() but retrieves the translation from the specified 806category, instead of the default category C<LC_MESSAGES>. 807 808 809=head2 pgettext( MSGCTXT, MSGID ) 810 811Returns the translation of MSGID, given the context of MSGCTXT. 812 813Both items are used as a unique key into the message catalog. 814 815This allows the translator to have two entries for words that may 816translate to different foreign words based on their context. For 817example, the word "View" may be a noun or a verb, which may be 818used in a menu as File->View or View->Source. 819 820 alert( pgettext( "Verb: To View", "View" ) ); 821 alert( pgettext( "Noun: A View", "View" ) ); 822 823The above will both lookup different entries in the message catalog. 824 825In English, or if no translation can be found, the second argument 826(B<MSGID>) is returned. 827 828=head2 dpgettext( TEXTDOMAIN, MSGCTXT, MSGID ) 829 830Like pgettext(), but retrieves the message for the specified 831B<TEXTDOMAIN> instead of the default domain. 832 833=head2 dcpgettext( TEXTDOMAIN, MSGCTXT, MSGID, CATEGORY ) 834 835Like dpgettext() but retrieves the message from the specified B<CATEGORY> 836instead of the default category C<LC_MESSAGES>. 837 838 839=head2 npgettext( MSGCTXT, MSGID, MSGID_PLURAL, COUNT ) 840 841Like ngettext() with the addition of context as in pgettext(). 842 843In English, or if no translation can be found, the second argument 844(MSGID) is picked if B<COUNT> is one, the third one otherwise. 845 846=head2 dnpgettext( TEXTDOMAIN, MSGCTXT, MSGID, MSGID_PLURAL, COUNT ) 847 848Like npgettext() but retrieves the translation from the specified 849textdomain instead of the default domain. 850 851=head2 dcnpgettext( TEXTDOMAIN, MSGCTXT, MSGID, MSGID_PLURAL, COUNT, CATEGORY ) 852 853Like dnpgettext() but retrieves the translation from the specified 854category, instead of the default category C<LC_MESSAGES>. 855 856=cut 857 858*/ 859 860// gettext 861Gettext.prototype.gettext = function (msgid) { 862 var msgctxt; 863 var msgid_plural; 864 var n; 865 var category; 866 return this.dcnpgettext(null, msgctxt, msgid, msgid_plural, n, category); 867}; 868 869Gettext.prototype.dgettext = function (domain, msgid) { 870 var msgctxt; 871 var msgid_plural; 872 var n; 873 var category; 874 return this.dcnpgettext(domain, msgctxt, msgid, msgid_plural, n, category); 875}; 876 877Gettext.prototype.dcgettext = function (domain, msgid, category) { 878 var msgctxt; 879 var msgid_plural; 880 var n; 881 return this.dcnpgettext(domain, msgctxt, msgid, msgid_plural, n, category); 882}; 883 884// ngettext 885Gettext.prototype.ngettext = function (msgid, msgid_plural, n) { 886 var msgctxt; 887 var category; 888 return this.dcnpgettext(null, msgctxt, msgid, msgid_plural, n, category); 889}; 890 891Gettext.prototype.dngettext = function (domain, msgid, msgid_plural, n) { 892 var msgctxt; 893 var category; 894 return this.dcnpgettext(domain, msgctxt, msgid, msgid_plural, n, category); 895}; 896 897Gettext.prototype.dcngettext = function (domain, msgid, msgid_plural, n, category) { 898 var msgctxt; 899 return this.dcnpgettext(domain, msgctxt, msgid, msgid_plural, n, category, category); 900}; 901 902// pgettext 903Gettext.prototype.pgettext = function (msgctxt, msgid) { 904 var msgid_plural; 905 var n; 906 var category; 907 return this.dcnpgettext(null, msgctxt, msgid, msgid_plural, n, category); 908}; 909 910Gettext.prototype.dpgettext = function (domain, msgctxt, msgid) { 911 var msgid_plural; 912 var n; 913 var category; 914 return this.dcnpgettext(domain, msgctxt, msgid, msgid_plural, n, category); 915}; 916 917Gettext.prototype.dcpgettext = function (domain, msgctxt, msgid, category) { 918 var msgid_plural; 919 var n; 920 return this.dcnpgettext(domain, msgctxt, msgid, msgid_plural, n, category); 921}; 922 923// npgettext 924Gettext.prototype.npgettext = function (msgctxt, msgid, msgid_plural, n) { 925 var category; 926 return this.dcnpgettext(null, msgctxt, msgid, msgid_plural, n, category); 927}; 928 929Gettext.prototype.dnpgettext = function (domain, msgctxt, msgid, msgid_plural, n) { 930 var category; 931 return this.dcnpgettext(domain, msgctxt, msgid, msgid_plural, n, category); 932}; 933 934// this has all the options, so we use it for all of them. 935Gettext.prototype.dcnpgettext = function (domain, msgctxt, msgid, msgid_plural, n, category) { 936 if (! this.isValidObject(msgid)) return ''; 937 938 var plural = this.isValidObject(msgid_plural); 939 var msg_ctxt_id = this.isValidObject(msgctxt) ? msgctxt+Gettext.context_glue+msgid : msgid; 940 941 var domainname = this.isValidObject(domain) ? domain : 942 this.isValidObject(this.domain) ? this.domain : 943 'messages'; 944 945 // category is always LC_MESSAGES. We ignore all else 946 var category_name = 'LC_MESSAGES'; 947 var category = 5; 948 949 var locale_data = new Array(); 950 if (typeof(Gettext._locale_data) != 'undefined' && 951 this.isValidObject(Gettext._locale_data[domainname])) { 952 locale_data.push( Gettext._locale_data[domainname] ); 953 954 } else if (typeof(Gettext._locale_data) != 'undefined') { 955 // didn't find domain we're looking for. Search all of them. 956 for (var dom in Gettext._locale_data) { 957 locale_data.push( Gettext._locale_data[dom] ); 958 } 959 } 960 961 var trans = []; 962 var found = false; 963 var domain_used; // so we can find plural-forms if needed 964 if (locale_data.length) { 965 for (var i=0; i<locale_data.length; i++) { 966 var locale = locale_data[i]; 967 if (this.isValidObject(locale.msgs[msg_ctxt_id])) { 968 // make copy of that array (cause we'll be destructive) 969 for (var j=0; j<locale.msgs[msg_ctxt_id].length; j++) { 970 trans[j] = locale.msgs[msg_ctxt_id][j]; 971 } 972 trans.shift(); // throw away the msgid_plural 973 domain_used = locale; 974 found = true; 975 // only break if found translation actually has a translation. 976 if ( trans.length > 0 && trans[0].length != 0 ) 977 break; 978 } 979 } 980 } 981 982 // default to english if we lack a match, or match has zero length 983 if ( trans.length == 0 || trans[0].length == 0 ) { 984 trans = [ msgid, msgid_plural ]; 985 } 986 987 var translation = trans[0]; 988 if (plural) { 989 var p; 990 if (found && this.isValidObject(domain_used.head.plural_func) ) { 991 var rv = domain_used.head.plural_func(n); 992 if (! rv.plural) rv.plural = 0; 993 if (! rv.nplural) rv.nplural = 0; 994 // if plurals returned is out of bound for total plural forms 995 if (rv.nplural <= rv.plural) rv.plural = 0; 996 p = rv.plural; 997 } else { 998 p = (n != 1) ? 1 : 0; 999 } 1000 if (this.isValidObject(trans[p])) 1001 translation = trans[p]; 1002 } 1003 1004 return translation; 1005}; 1006 1007 1008/* 1009 1010=head2 strargs (string, argument_array) 1011 1012 string : a string that potentially contains formatting characters. 1013 argument_array : an array of positional replacement values 1014 1015This is a utility method to provide some way to support positional parameters within a string, as javascript lacks a printf() method. 1016 1017The format is similar to printf(), but greatly simplified (ie. fewer features). 1018 1019Any percent signs followed by numbers are replaced with the corrosponding item from the B<argument_array>. 1020 1021Example: 1022 1023 var string = "%2 roses are red, %1 violets are blue"; 1024 var args = new Array("10", "15"); 1025 var result = Gettext.strargs(string, args); 1026 // result is "15 roses are red, 10 violets are blue" 1027 1028The format numbers are 1 based, so the first itme is %1. 1029 1030A lone percent sign may be escaped by preceeding it with another percent sign. 1031 1032A percent sign followed by anything other than a number or another percent sign will be passed through as is. 1033 1034Some more examples should clear up any abmiguity. The following were called with the orig string, and the array as Array("[one]", "[two]") : 1035 1036 orig string "blah" becomes "blah" 1037 orig string "" becomes "" 1038 orig string "%%" becomes "%" 1039 orig string "%%%" becomes "%%" 1040 orig string "%%%%" becomes "%%" 1041 orig string "%%%%%" becomes "%%%" 1042 orig string "tom%%dick" becomes "tom%dick" 1043 orig string "thing%1bob" becomes "thing[one]bob" 1044 orig string "thing%1%2bob" becomes "thing[one][two]bob" 1045 orig string "thing%1asdf%2asdf" becomes "thing[one]asdf[two]asdf" 1046 orig string "%1%2%3" becomes "[one][two]" 1047 orig string "tom%1%%2%aDick" becomes "tom[one]%2%aDick" 1048 1049This is especially useful when using plurals, as the string will nearly always contain the number. 1050 1051It's also useful in translated strings where the translator may have needed to move the position of the parameters. 1052 1053For example: 1054 1055 var count = 14; 1056 Gettext.strargs( gt.ngettext('one banana', '%1 bananas', count), [count] ); 1057 1058NOTE: this may be called as an instance method, or as a class method. 1059 1060 // instance method: 1061 var gt = new Gettext(params); 1062 gt.strargs(string, args); 1063 1064 // class method: 1065 Gettext.strargs(string, args); 1066 1067=cut 1068 1069*/ 1070/* utility method, since javascript lacks a printf */ 1071Gettext.strargs = function (str, args) { 1072 // make sure args is an array 1073 if ( null == args || 1074 'undefined' == typeof(args) ) { 1075 args = []; 1076 } else if (args.constructor != Array) { 1077 args = [args]; 1078 } 1079 1080 // NOTE: javascript lacks support for zero length negative look-behind 1081 // in regex, so we must step through w/ index. 1082 // The perl equiv would simply be: 1083 // $string =~ s/(?<!\%)\%([0-9]+)/$args[$1]/g; 1084 // $string =~ s/\%\%/\%/g; # restore escaped percent signs 1085 1086 var newstr = ""; 1087 while (true) { 1088 var i = str.indexOf('%'); 1089 var match_n; 1090 1091 // no more found. Append whatever remains 1092 if (i == -1) { 1093 newstr += str; 1094 break; 1095 } 1096 1097 // we found it, append everything up to that 1098 newstr += str.substr(0, i); 1099 1100 // check for escpaed %% 1101 if (str.substr(i, 2) == '%%') { 1102 newstr += '%'; 1103 str = str.substr((i+2)); 1104 1105 // % followed by number 1106 } else if ( match_n = str.substr(i).match(/^%(\d+)/) ) { 1107 var arg_n = parseInt(match_n[1]); 1108 var length_n = match_n[1].length; 1109 if ( arg_n > 0 && args[arg_n -1] != null && typeof(args[arg_n -1]) != 'undefined' ) 1110 newstr += args[arg_n -1]; 1111 str = str.substr( (i + 1 + length_n) ); 1112 1113 // % followed by some other garbage - just remove the % 1114 } else { 1115 newstr += '%'; 1116 str = str.substr((i+1)); 1117 } 1118 } 1119 1120 return newstr; 1121} 1122 1123/* instance method wrapper of strargs */ 1124Gettext.prototype.strargs = function (str, args) { 1125 return Gettext.strargs(str, args); 1126} 1127 1128/* verify that something is an array */ 1129Gettext.prototype.isArray = function (thisObject) { 1130 return this.isValidObject(thisObject) && thisObject.constructor == Array; 1131}; 1132 1133/* verify that an object exists and is valid */ 1134Gettext.prototype.isValidObject = function (thisObject) { 1135 if (null == thisObject) { 1136 return false; 1137 } else if ('undefined' == typeof(thisObject) ) { 1138 return false; 1139 } else { 1140 return true; 1141 } 1142}; 1143 1144Gettext.prototype.sjax = function (uri) { 1145 var xmlhttp; 1146 if (window.XMLHttpRequest) { 1147 xmlhttp = new XMLHttpRequest(); 1148 } else if (navigator.userAgent.toLowerCase().indexOf('msie 5') != -1) { 1149 xmlhttp = new ActiveXObject("Microsoft.XMLHTTP"); 1150 } else { 1151 xmlhttp = new ActiveXObject("Msxml2.XMLHTTP"); 1152 } 1153 1154 if (! xmlhttp) 1155 throw new Error("Your browser doesn't do Ajax. Unable to support external language files."); 1156 1157 xmlhttp.open('GET', uri, false); 1158 try { xmlhttp.send(null); } 1159 catch (e) { return; } 1160 1161 // we consider status 200 and 0 as ok. 1162 // 0 happens when we request local file, allowing this to run on local files 1163 var sjax_status = xmlhttp.status; 1164 if (sjax_status == 200 || sjax_status == 0) { 1165 return xmlhttp.responseText; 1166 } else { 1167 var error = xmlhttp.statusText + " (Error " + xmlhttp.status + ")"; 1168 if (xmlhttp.responseText.length) { 1169 error += "\n" + xmlhttp.responseText; 1170 } 1171 alert(error); 1172 return; 1173 } 1174} 1175 1176Gettext.prototype.JSON = function (data) { 1177 return eval('(' + data + ')'); 1178} 1179 1180 1181/* 1182 1183=head1 NOTES 1184 1185These are some notes on the internals 1186 1187=over 1188 1189=item LOCALE CACHING 1190 1191Loaded locale data is currently cached class-wide. This means that if two scripts are both using Gettext.js, and both share the same gettext domain, that domain will only be loaded once. This will allow you to grab a new object many times from different places, utilize the same domain, and share a single translation file. The downside is that a domain won't be RE-loaded if a new object is instantiated on a domain that had already been instantiated. 1192 1193=back 1194 1195=head1 BUGS / TODO 1196 1197=over 1198 1199=item error handling 1200 1201Currently, there are several places that throw errors. In GNU Gettext, there are no fatal errors, which allows text to still be displayed regardless of how broken the environment becomes. We should evaluate and determine where we want to stand on that issue. 1202 1203=item syncronous only support (no ajax support) 1204 1205Currently, fetching language data is done purely syncronous, which means the page will halt while those files are fetched/loaded. 1206 1207This is often what you want, as then following translation requests will actually be translated. However, if all your calls are done dynamically (ie. error handling only or something), loading in the background may be more adventagous. 1208 1209It's still recommended to use the statically defined <script ...> method, which should have the same delay, but it will cache the result. 1210 1211=item domain support 1212 1213domain support while using shortcut methods like C<_('string')> or C<i18n('string')>. 1214 1215Under normal apps, the domain is usually set globally to the app, and a single language file is used. Under javascript, you may have multiple libraries or applications needing translation support, but the namespace is essentially global. 1216 1217It's recommended that your app initialize it's own shortcut with it's own domain. (See examples/wrapper/i18n.js for an example.) 1218 1219Basically, you'll want to accomplish something like this: 1220 1221 // in some other .js file that needs i18n 1222 this.i18nObj = new i18n; 1223 this.i18n = this.i18nObj.init('domain'); 1224 // do translation 1225 alert( this.i18n("string") ); 1226 1227If you use this raw Gettext object, then this is all handled for you, as you have your own object then, and will be calling C<myGettextObject.gettext('string')> and such. 1228 1229 1230=item encoding 1231 1232May want to add encoding/reencoding stuff. See GNU iconv, or the perl module Locale::Recode from libintl-perl. 1233 1234=back 1235 1236 1237=head1 COMPATABILITY 1238 1239This has been tested on the following browsers. It may work on others, but these are all those to which I have access. 1240 1241 FF1.5, FF2, FF3, IE6, IE7, Opera9, Opera10, Safari3.1, Chrome 1242 1243 *FF = Firefox 1244 *IE = Internet Explorer 1245 1246 1247=head1 REQUIRES 1248 1249bin/po2json requires perl, and the perl modules Locale::PO and JSON. 1250 1251=head1 SEE ALSO 1252 1253bin/po2json (included), 1254examples/normal/index.html, 1255examples/wrapper/i18n.html, examples/wrapper/i18n.js, 1256Locale::gettext_pp(3pm), POSIX(3pm), gettext(1), gettext(3) 1257 1258=head1 AUTHOR 1259 1260Copyright (C) 2008, Joshua I. Miller E<lt>unrtst@cpan.orgE<gt>, all rights reserved. See the source code for details. 1261 1262=cut 1263 1264*/ 1265