4 // Copyright (C) 2006 Ingo Ruhnke <grumbel@gmx.de>
6 // This program is free software; you can redistribute it and/or
7 // modify it under the terms of the GNU General Public License
8 // as published by the Free Software Foundation; either version 2
9 // of the License, or (at your option) any later version.
11 // This program is distributed in the hope that it will be useful,
12 // but WITHOUT ANY WARRANTY; without even the implied warranty of
13 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 // GNU General Public License for more details.
16 // You should have received a copy of the GNU General Public License
17 // along with this program; if not, write to the Free Software
18 // Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
22 #include <sys/types.h>
31 #include "tinygettext.hpp"
33 #include "physfs/physfs_stream.hpp"
35 #include "findlocale.hpp"
37 //#define TRANSLATION_DEBUG
39 namespace TinyGetText {
41 /** Convert \a which is in \a from_charset to \a to_charset and return it */
42 std::string convert(const std::string& text,
43 const std::string& from_charset,
44 const std::string& to_charset)
46 if (from_charset == to_charset)
49 char *in = new char[text.length() + 1];
50 strcpy(in, text.c_str());
51 char *out = SDL_iconv_string(to_charset.c_str(), from_charset.c_str(), in, text.length() + 1);
55 log_warning << "Error: conversion from " << from_charset << " to " << to_charset << " failed" << std::endl;
62 iconv_t cd = SDL_iconv_open(to_charset.c_str(), from_charset.c_str());
64 size_t in_len = text.length();
65 size_t out_len = text.length()*3; // FIXME: cross fingers that this is enough
67 char* out_orig = new char[out_len];
68 char* in_orig = new char[in_len+1];
69 strcpy(in_orig, text.c_str());
72 ICONV_CONST char* in = in_orig;
73 size_t out_len_temp = out_len; // iconv is counting down the bytes it has
74 // written from this...
76 size_t retval = SDL_iconv(cd, &in, &in_len, &out, &out_len_temp);
77 out_len -= out_len_temp; // see above
78 if (retval == (size_t) -1)
80 log_warning << strerror(errno) << std::endl;
81 log_warning << "Error: conversion from " << from_charset << " to " << to_charset << " went wrong: " << retval << std::endl;
86 std::string ret(out_orig, out_len);
93 bool has_suffix(const std::string& lhs, const std::string rhs)
95 if (lhs.length() < rhs.length())
98 return lhs.compare(lhs.length() - rhs.length(), rhs.length(), rhs) == 0;
101 bool has_prefix(const std::string& lhs, const std::string rhs)
103 if (lhs.length() < rhs.length())
106 return lhs.compare(0, rhs.length(), rhs) == 0;
109 int plural1(int ) { return 0; }
110 int plural2_1(int n) { return (n != 1); }
111 int plural2_2(int n) { return (n > 1); }
112 int plural3_lv(int n) { return (n%10==1 && n%100!=11 ? 0 : n != 0 ? 1 : 2); }
113 int plural3_ga(int n) { return n==1 ? 0 : n==2 ? 1 : 2; }
114 int plural3_lt(int n) { return (n%10==1 && n%100!=11 ? 0 : n%10>=2 && (n%100<10 || n%100>=20) ? 1 : 2); }
115 int plural3_1(int n) { return (n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2); }
116 int plural3_sk(int n) { return (n==1) ? 0 : (n>=2 && n<=4) ? 1 : 2; }
117 int plural3_pl(int n) { return (n==1 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2); }
118 int plural3_sl(int n) { return (n%100==1 ? 0 : n%100==2 ? 1 : n%100==3 || n%100==4 ? 2 : 3); }
120 /** Language Definitions */
122 LanguageDef lang_hu("hu", "Hungarian", 1, plural1); // "nplurals=1; plural=0;"
123 LanguageDef lang_ja("ja", "Japanese", 1, plural1); // "nplurals=1; plural=0;"
124 LanguageDef lang_ko("ko", "Korean", 1, plural1); // "nplurals=1; plural=0;"
125 LanguageDef lang_tr("tr", "Turkish", 1, plural1); // "nplurals=1; plural=0;"
126 LanguageDef lang_da("da", "Danish", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
127 LanguageDef lang_nl("nl", "Dutch", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
128 LanguageDef lang_en("en", "English", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
129 LanguageDef lang_fo("fo", "Faroese", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
130 LanguageDef lang_de("de", "German", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
131 LanguageDef lang_nb("nb", "Norwegian Bokmal", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
132 LanguageDef lang_no("no", "Norwegian", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
133 LanguageDef lang_nn("nn", "Norwegian Nynorsk", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
134 LanguageDef lang_sv("sv", "Swedish", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
135 LanguageDef lang_et("et", "Estonian", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
136 LanguageDef lang_fi("fi", "Finnish", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
137 LanguageDef lang_el("el", "Greek", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
138 LanguageDef lang_he("he", "Hebrew", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
139 LanguageDef lang_it("it", "Italian", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
140 LanguageDef lang_pt("pt", "Portuguese", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
141 LanguageDef lang_es("es", "Spanish", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
142 LanguageDef lang_eo("eo", "Esperanto", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
143 LanguageDef lang_fr("fr", "French", 2, plural2_2); // "nplurals=2; plural=(n > 1);"
144 LanguageDef lang_pt_BR("pt_BR", "Brazilian", 2, plural2_2); // "nplurals=2; plural=(n > 1);"
145 LanguageDef lang_lv("lv", "Latvian", 3, plural3_lv); // "nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n != 0 ? 1 : 2);"
146 LanguageDef lang_ga("ga", "Irish", 3, plural3_ga); // "nplurals=3; plural=n==1 ? 0 : n==2 ? 1 : 2;"
147 LanguageDef lang_lt("lt", "Lithuanian", 3, plural3_lt); // "nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && (n%100<10 || n%100>=20) ? 1 : 2);"
148 LanguageDef lang_hr("hr", "Croatian", 3, plural3_1); // "nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);"
149 LanguageDef lang_cs("cs", "Czech", 3, plural3_1); // "nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);"
150 LanguageDef lang_ru("ru", "Russian", 3, plural3_1); // "nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);"
151 LanguageDef lang_uk("uk", "Ukrainian", 3, plural3_1); // "nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);"
152 LanguageDef lang_sk("sk", "Slovak", 3, plural3_sk); // "nplurals=3; plural=(n==1) ? 0 : (n>=2 && n<=4) ? 1 : 2;"
153 LanguageDef lang_pl("pl", "Polish", 3, plural3_pl); // "nplurals=3; plural=(n==1 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2);
154 LanguageDef lang_sl("sl", "Slovenian", 3, plural3_sl); // "nplurals=4; plural=(n%100==1 ? 0 : n%100==2 ? 1 : n%100==3 || n%100==4 ? 2 : 3);"
158 get_language_def(const std::string& name)
160 if (name == "hu") return lang_hu;
161 else if (name == "ja") return lang_ja;
162 else if (name == "ko") return lang_ko;
163 else if (name == "tr") return lang_tr;
164 else if (name == "da") return lang_da;
165 else if (name == "nl") return lang_nl;
166 else if (name == "en") return lang_en;
167 else if (name == "fo") return lang_fo;
168 else if (name == "de") return lang_de;
169 else if (name == "nb") return lang_nb;
170 else if (name == "no") return lang_no;
171 else if (name == "nn") return lang_nn;
172 else if (name == "sv") return lang_sv;
173 else if (name == "et") return lang_et;
174 else if (name == "fi") return lang_fi;
175 else if (name == "el") return lang_el;
176 else if (name == "he") return lang_he;
177 else if (name == "it") return lang_it;
178 else if (name == "pt") return lang_pt;
179 else if (name == "es") return lang_es;
180 else if (name == "eo") return lang_eo;
181 else if (name == "fr") return lang_fr;
182 else if (name == "pt_BR") return lang_pt_BR;
183 else if (name == "lv") return lang_lv;
184 else if (name == "ga") return lang_ga;
185 else if (name == "lt") return lang_lt;
186 else if (name == "hr") return lang_hr;
187 else if (name == "cs") return lang_cs;
188 else if (name == "ru") return lang_ru;
189 else if (name == "uk") return lang_uk;
190 else if (name == "sk") return lang_sk;
191 else if (name == "pl") return lang_pl;
192 else if (name == "sl") return lang_sl;
196 DictionaryManager::DictionaryManager()
197 : current_dict(&empty_dict)
199 parseLocaleAliases();
200 // Environment variable SUPERTUX_LANG overrides language settings.
201 const char* lang = getenv( "SUPERTUX_LANG" );
203 set_language( lang );
206 // use findlocale to setup language
208 FL_FindLocale( &locale, FL_MESSAGES );
210 if (locale->country) {
211 set_language( std::string(locale->lang)+"_"+std::string(locale->country) );
213 set_language( std::string(locale->lang) );
216 FL_FreeLocale( &locale );
220 DictionaryManager::parseLocaleAliases()
222 // try to parse language alias list
223 std::ifstream in("/usr/share/locale/locale.alias");
226 while(in.good() && !in.eof()) {
227 while(isspace(c) && !in.eof())
230 if(c == '#') { // skip comments
231 while(c != '\n' && !in.eof())
237 while(!isspace(c) && !in.eof()) {
241 while(isspace(c) && !in.eof())
243 std::string language;
244 while(!isspace(c) && !in.eof()) {
251 set_language_alias(alias, language);
256 DictionaryManager::get_dictionary(const std::string& spec)
259 //log_debug << "Dictionary for language \"" << spec << "\" requested" << std::endl;
261 std::string lang = get_language_from_spec(spec);
263 //log_debug << "...normalized as \"" << lang << "\"" << std::endl;
265 Dictionaries::iterator i = dictionaries.find(get_language_from_spec(lang));
266 if (i != dictionaries.end())
270 else // Dictionary for languages lang isn't loaded, so we load it
272 //log_debug << "get_dictionary: " << lang << std::endl;
273 Dictionary& dict = dictionaries[lang];
275 dict.set_language(get_language_def(lang));
277 dict.set_charset(charset);
279 for (SearchPath::iterator p = search_path.begin(); p != search_path.end(); ++p)
281 char** files = PHYSFS_enumerateFiles(p->c_str());
284 log_warning << "Error: enumerateFiles() failed on " << *p << std::endl;
288 for(const char* const* filename = files;
289 *filename != 0; filename++) {
291 // check if filename matches requested language
292 std::string fname = std::string(*filename);
293 std::string load_from_file = "";
294 if(fname == lang + ".po") {
295 load_from_file = fname;
297 std::string::size_type s = lang.find("_");
298 if(s != std::string::npos) {
299 std::string lang_short = std::string(lang, 0, s);
300 if (fname == lang_short + ".po") {
301 load_from_file = lang_short;
306 // if it matched, load dictionary
307 if (load_from_file != "") {
308 //log_debug << "Loading dictionary for language \"" << lang << "\" from \"" << filename << "\"" << std::endl;
309 std::string pofile = *p + "/" + *filename;
311 IFileStream in(pofile);
312 read_po_file(dict, in);
313 } catch(std::exception& e) {
314 log_warning << "Error: Failure file opening: " << pofile << std::endl;
315 log_warning << e.what() << "" << std::endl;
320 PHYSFS_freeList(files);
328 std::set<std::string>
329 DictionaryManager::get_languages()
331 std::set<std::string> languages;
333 for (SearchPath::iterator p = search_path.begin(); p != search_path.end(); ++p)
335 char** files = PHYSFS_enumerateFiles(p->c_str());
338 log_warning << "Error: opendir() failed on " << *p << std::endl;
342 for(const char* const* file = files; *file != 0; file++) {
343 if(has_suffix(*file, ".po")) {
344 std::string filename = *file;
345 languages.insert(filename.substr(0, filename.length()-3));
348 PHYSFS_freeList(files);
355 DictionaryManager::set_language(const std::string& lang)
357 //log_debug << "set_language \"" << lang << "\"" << std::endl;
358 language = get_language_from_spec(lang);
359 //log_debug << "==> \"" << language << "\"" << std::endl;
360 current_dict = & (get_dictionary(language));
364 DictionaryManager::get_language() const
370 DictionaryManager::set_charset(const std::string& charset)
372 dictionaries.clear(); // changing charset invalidates cache
373 this->charset = charset;
374 set_language(language);
378 DictionaryManager::set_language_alias(const std::string& alias,
379 const std::string& language)
381 language_aliases.insert(std::make_pair(alias, language));
385 DictionaryManager::get_language_from_spec(const std::string& spec)
387 std::string lang = spec;
388 Aliases::iterator i = language_aliases.find(lang);
389 if(i != language_aliases.end()) {
393 std::string::size_type s = lang.find(".");
394 if(s != std::string::npos) {
395 lang = std::string(lang, 0, s);
399 if(s == std::string::npos) {
400 std::string lang_big = lang;
401 std::transform (lang_big.begin(), lang_big.end(), lang_big.begin(), toupper);
402 lang += "_" + lang_big;
410 DictionaryManager::add_directory(const std::string& pathname)
412 dictionaries.clear(); // adding directories invalidates cache
413 search_path.push_back(pathname);
414 set_language(language);
417 //---------------------------------------------------------------------------
419 Dictionary::Dictionary(const LanguageDef& language_, const std::string& charset_)
420 : language(language_), charset(charset_)
424 Dictionary::Dictionary()
430 Dictionary::get_charset() const
436 Dictionary::set_charset(const std::string& charset_)
442 Dictionary::set_language(const LanguageDef& lang)
448 Dictionary::translate(const std::string& msgid, const std::string& msgid2, int num)
450 PluralEntries::iterator i = plural_entries.find(msgid);
451 std::map<int, std::string>& msgstrs = i->second;
453 if (i != plural_entries.end() && !msgstrs.empty())
455 int g = language.plural(num);
456 std::map<int, std::string>::iterator j = msgstrs.find(g);
457 if (j != msgstrs.end())
463 // Return the first translation, in case we can't translate the specific number
464 return msgstrs.begin()->second;
469 #ifdef TRANSLATION_DEBUG
470 log_warning << "Couldn't translate: " << msgid << std::endl;
471 log_warning << "Candidates: " << std::endl;
472 for (PluralEntries::iterator i = plural_entries.begin(); i != plural_entries.end(); ++i)
473 log_debug << "'" << i->first << "'" << std::endl;
476 if (plural2_1(num)) // default to english rules
484 Dictionary::translate(const char* msgid)
486 Entries::iterator i = entries.find(msgid);
487 if (i != entries.end() && !i->second.empty())
489 return i->second.c_str();
493 #ifdef TRANSLATION_DBEUG
494 log_warning << "Couldn't translate: " << msgid << std::endl;
501 Dictionary::translate(const std::string& msgid)
503 Entries::iterator i = entries.find(msgid);
504 if (i != entries.end() && !i->second.empty())
510 #ifdef TRANSLATION_DBEUG
511 log_warning << "Couldn't translate: " << msgid << std::endl;
518 Dictionary::add_translation(const std::string& msgid, const std::string& ,
519 const std::map<int, std::string>& msgstrs)
521 // Do we need msgid2 for anything? its after all supplied to the
522 // translate call, so we just throw it away
523 plural_entries[msgid] = msgstrs;
527 Dictionary::add_translation(const std::string& msgid, const std::string& msgstr)
529 entries[msgid] = msgstr;
543 std::string from_charset;
544 std::string to_charset;
546 std::string current_msgid;
547 std::string current_msgid_plural;
548 std::map<int, std::string> msgstr_plural;
552 enum { WANT_MSGID, WANT_MSGSTR, WANT_MSGSTR_PLURAL, WANT_MSGID_PLURAL } state;
555 POFileReader(std::istream& in, Dictionary& dict_)
561 if(c == (char) 0xef) { // skip UTF-8 intro that some texteditors produce
570 void parse_header(const std::string& header)
572 // Seperate the header in lines
573 typedef std::vector<std::string> Lines;
576 std::string::size_type start = 0;
577 for(std::string::size_type i = 0; i < header.length(); ++i)
579 if (header[i] == '\n')
581 lines.push_back(header.substr(start, i - start));
586 for(Lines::iterator i = lines.begin(); i != lines.end(); ++i)
588 if (has_prefix(*i, "Content-Type: text/plain; charset=")) {
589 from_charset = i->substr(strlen("Content-Type: text/plain; charset="));
593 if (from_charset.empty() || from_charset == "CHARSET")
595 log_warning << "Error: Charset not specified for .po, fallback to ISO-8859-1" << std::endl;
596 from_charset = "ISO-8859-1";
599 to_charset = dict.get_charset();
600 if (to_charset.empty())
601 { // No charset requested from the dict, use utf-8
602 to_charset = "utf-8";
603 dict.set_charset(from_charset);
607 void add_token(const Token& token)
612 if (token.keyword == "msgid")
614 current_msgid = token.content;
615 state = WANT_MSGID_PLURAL;
617 else if (token.keyword.empty())
619 //log_warning << "Got EOF, everything looks ok." << std::endl;
623 log_warning << "tinygettext: expected 'msgid' keyword, got " << token.keyword << " at line " << line_num << std::endl;
627 case WANT_MSGID_PLURAL:
628 if (token.keyword == "msgid_plural")
630 current_msgid_plural = token.content;
631 state = WANT_MSGSTR_PLURAL;
641 if (token.keyword == "msgstr")
643 if (current_msgid == "")
644 { // .po Header is hidden in the msgid with the empty string
645 parse_header(token.content);
649 dict.add_translation(current_msgid, convert(token.content, from_charset, to_charset));
655 log_warning << "tinygettext: expected 'msgstr' keyword, got " << token.keyword << " at line " << line_num << std::endl;
659 case WANT_MSGSTR_PLURAL:
660 if (has_prefix(token.keyword, "msgstr["))
663 if (sscanf(token.keyword.c_str(), "msgstr[%d]", &num) != 1)
665 log_warning << "Error: Couldn't parse: " << token.keyword << std::endl;
669 msgstr_plural[num] = convert(token.content, from_charset, to_charset);
674 dict.add_translation(current_msgid, current_msgid_plural, msgstr_plural);
683 inline int getchar(std::istream& in)
691 void tokenize_po(std::istream& in)
693 enum State { READ_KEYWORD,
695 READ_CONTENT_IN_STRING,
698 State state = READ_KEYWORD;
702 while((c = getchar(in)) != EOF)
704 //log_debug << "Lexing char: " << char(c) << " " << state << std::endl;
710 state = SKIP_COMMENT;
719 } while((c = getchar(in)) != EOF && !isspace(c));
722 state = READ_CONTENT;
727 while((c = getchar(in)) != EOF)
730 // Found start of content
731 state = READ_CONTENT_IN_STRING;
733 } else if (isspace(c)) {
735 } else { // Read something that may be a keyword
737 state = READ_KEYWORD;
744 case READ_CONTENT_IN_STRING:
749 if (c == 'n') token.content += '\n';
750 else if (c == 't') token.content += '\t';
751 else if (c == 'r') token.content += '\r';
752 else if (c == '"') token.content += '"';
753 else if (c == '\\') token.content += '\\';
756 log_warning << "Unhandled escape character: " << char(c) << std::endl;
761 log_warning << "Unterminated string" << std::endl;
763 } else if (c == '"') { // Content string is terminated
764 state = READ_CONTENT;
772 state = READ_KEYWORD;
780 void read_po_file(Dictionary& dict_, std::istream& in)
782 POFileReader reader(in, dict_);
785 } // namespace TinyGetText