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"
34 #include "findlocale.hpp"
36 //#define TRANSLATION_DEBUG
38 namespace TinyGetText {
40 /** Convert \a which is in \a from_charset to \a to_charset and return it */
41 std::string convert(const std::string& text,
42 const std::string& from_charset,
43 const std::string& to_charset)
45 if (from_charset == to_charset)
48 char *in = new char[text.length() + 1];
49 strcpy(in, text.c_str());
50 char *out = SDL_iconv_string(to_charset.c_str(), from_charset.c_str(), in, text.length() + 1);
54 log_warning << "Error: conversion from " << from_charset << " to " << to_charset << " failed" << std::endl;
61 iconv_t cd = SDL_iconv_open(to_charset.c_str(), from_charset.c_str());
63 size_t in_len = text.length();
64 size_t out_len = text.length()*3; // FIXME: cross fingers that this is enough
66 char* out_orig = new char[out_len];
67 char* in_orig = new char[in_len+1];
68 strcpy(in_orig, text.c_str());
71 ICONV_CONST char* in = in_orig;
72 size_t out_len_temp = out_len; // iconv is counting down the bytes it has
73 // written from this...
75 size_t retval = SDL_iconv(cd, &in, &in_len, &out, &out_len_temp);
76 out_len -= out_len_temp; // see above
77 if (retval == (size_t) -1)
79 log_warning << strerror(errno) << std::endl;
80 log_warning << "Error: conversion from " << from_charset << " to " << to_charset << " went wrong: " << retval << std::endl;
85 std::string ret(out_orig, out_len);
92 bool has_suffix(const std::string& lhs, const std::string rhs)
94 if (lhs.length() < rhs.length())
97 return lhs.compare(lhs.length() - rhs.length(), rhs.length(), rhs) == 0;
100 bool has_prefix(const std::string& lhs, const std::string rhs)
102 if (lhs.length() < rhs.length())
105 return lhs.compare(0, rhs.length(), rhs) == 0;
108 int plural1(int ) { return 0; }
109 int plural2_1(int n) { return (n != 1); }
110 int plural2_2(int n) { return (n > 1); }
111 int plural3_lv(int n) { return (n%10==1 && n%100!=11 ? 0 : n != 0 ? 1 : 2); }
112 int plural3_ga(int n) { return n==1 ? 0 : n==2 ? 1 : 2; }
113 int plural3_lt(int n) { return (n%10==1 && n%100!=11 ? 0 : n%10>=2 && (n%100<10 || n%100>=20) ? 1 : 2); }
114 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); }
115 int plural3_sk(int n) { return (n==1) ? 0 : (n>=2 && n<=4) ? 1 : 2; }
116 int plural3_pl(int n) { return (n==1 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2); }
117 int plural3_sl(int n) { return (n%100==1 ? 0 : n%100==2 ? 1 : n%100==3 || n%100==4 ? 2 : 3); }
119 /** Language Definitions */
121 LanguageDef lang_hu("hu", "Hungarian", 1, plural1); // "nplurals=1; plural=0;"
122 LanguageDef lang_ja("ja", "Japanese", 1, plural1); // "nplurals=1; plural=0;"
123 LanguageDef lang_ko("ko", "Korean", 1, plural1); // "nplurals=1; plural=0;"
124 LanguageDef lang_tr("tr", "Turkish", 1, plural1); // "nplurals=1; plural=0;"
125 LanguageDef lang_da("da", "Danish", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
126 LanguageDef lang_nl("nl", "Dutch", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
127 LanguageDef lang_en("en", "English", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
128 LanguageDef lang_fo("fo", "Faroese", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
129 LanguageDef lang_de("de", "German", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
130 LanguageDef lang_nb("nb", "Norwegian Bokmal", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
131 LanguageDef lang_no("no", "Norwegian", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
132 LanguageDef lang_nn("nn", "Norwegian Nynorsk", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
133 LanguageDef lang_sv("sv", "Swedish", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
134 LanguageDef lang_et("et", "Estonian", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
135 LanguageDef lang_fi("fi", "Finnish", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
136 LanguageDef lang_el("el", "Greek", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
137 LanguageDef lang_he("he", "Hebrew", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
138 LanguageDef lang_it("it", "Italian", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
139 LanguageDef lang_pt("pt", "Portuguese", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
140 LanguageDef lang_es("es", "Spanish", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
141 LanguageDef lang_eo("eo", "Esperanto", 2, plural2_1); // "nplurals=2; plural=(n != 1);"
142 LanguageDef lang_fr("fr", "French", 2, plural2_2); // "nplurals=2; plural=(n > 1);"
143 LanguageDef lang_pt_BR("pt_BR", "Brazilian", 2, plural2_2); // "nplurals=2; plural=(n > 1);"
144 LanguageDef lang_lv("lv", "Latvian", 3, plural3_lv); // "nplurals=3; plural=(n%10==1 && n%100!=11 ? 0 : n != 0 ? 1 : 2);"
145 LanguageDef lang_ga("ga", "Irish", 3, plural3_ga); // "nplurals=3; plural=n==1 ? 0 : n==2 ? 1 : 2;"
146 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);"
147 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);"
148 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);"
149 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);"
150 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);"
151 LanguageDef lang_sk("sk", "Slovak", 3, plural3_sk); // "nplurals=3; plural=(n==1) ? 0 : (n>=2 && n<=4) ? 1 : 2;"
152 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);
153 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);"
157 get_language_def(const std::string& name)
159 if (name == "hu") return lang_hu;
160 else if (name == "ja") return lang_ja;
161 else if (name == "ko") return lang_ko;
162 else if (name == "tr") return lang_tr;
163 else if (name == "da") return lang_da;
164 else if (name == "nl") return lang_nl;
165 else if (name == "en") return lang_en;
166 else if (name == "fo") return lang_fo;
167 else if (name == "de") return lang_de;
168 else if (name == "nb") return lang_nb;
169 else if (name == "no") return lang_no;
170 else if (name == "nn") return lang_nn;
171 else if (name == "sv") return lang_sv;
172 else if (name == "et") return lang_et;
173 else if (name == "fi") return lang_fi;
174 else if (name == "el") return lang_el;
175 else if (name == "he") return lang_he;
176 else if (name == "it") return lang_it;
177 else if (name == "pt") return lang_pt;
178 else if (name == "es") return lang_es;
179 else if (name == "eo") return lang_eo;
180 else if (name == "fr") return lang_fr;
181 else if (name == "pt_BR") return lang_pt_BR;
182 else if (name == "lv") return lang_lv;
183 else if (name == "ga") return lang_ga;
184 else if (name == "lt") return lang_lt;
185 else if (name == "hr") return lang_hr;
186 else if (name == "cs") return lang_cs;
187 else if (name == "ru") return lang_ru;
188 else if (name == "uk") return lang_uk;
189 else if (name == "sk") return lang_sk;
190 else if (name == "pl") return lang_pl;
191 else if (name == "sl") return lang_sl;
195 DictionaryManager::DictionaryManager()
196 : current_dict(&empty_dict)
198 parseLocaleAliases();
199 // Environment variable SUPERTUX_LANG overrides language settings.
200 const char* lang = getenv( "SUPERTUX_LANG" );
202 set_language( lang );
205 // use findlocale to setup language
207 FL_FindLocale( &locale, FL_MESSAGES );
209 if (locale->country) {
210 set_language( std::string(locale->lang)+"_"+std::string(locale->country) );
212 set_language( std::string(locale->lang) );
215 FL_FreeLocale( &locale );
219 DictionaryManager::parseLocaleAliases()
221 // try to parse language alias list
222 std::ifstream in("/usr/share/locale/locale.alias");
225 while(in.good() && !in.eof()) {
226 while(isspace(static_cast<unsigned char>(c)) && !in.eof())
229 if(c == '#') { // skip comments
230 while(c != '\n' && !in.eof())
236 while(!isspace(static_cast<unsigned char>(c)) && !in.eof()) {
240 while(isspace(static_cast<unsigned char>(c)) && !in.eof())
242 std::string language;
243 while(!isspace(static_cast<unsigned char>(c)) && !in.eof()) {
250 set_language_alias(alias, language);
255 DictionaryManager::get_dictionary(const std::string& spec)
258 //log_debug << "Dictionary for language \"" << spec << "\" requested" << std::endl;
260 std::string lang = get_language_from_spec(spec);
262 //log_debug << "...normalized as \"" << lang << "\"" << std::endl;
264 Dictionaries::iterator i = dictionaries.find(get_language_from_spec(lang));
265 if (i != dictionaries.end())
269 else // Dictionary for languages lang isn't loaded, so we load it
271 //log_debug << "get_dictionary: " << lang << std::endl;
272 Dictionary& dict = dictionaries[lang];
274 dict.set_language(get_language_def(lang));
276 dict.set_charset(charset);
278 for (SearchPath::iterator p = search_path.begin(); p != search_path.end(); ++p)
280 char** files = PHYSFS_enumerateFiles(p->c_str());
283 log_warning << "Error: enumerateFiles() failed on " << *p << std::endl;
287 for(const char* const* filename = files;
288 *filename != 0; filename++) {
290 // check if filename matches requested language
291 std::string fname = std::string(*filename);
292 std::string load_from_file = "";
293 if(fname == lang + ".po") {
294 load_from_file = fname;
296 std::string::size_type s = lang.find("_");
297 if(s != std::string::npos) {
298 std::string lang_short = std::string(lang, 0, s);
299 if (fname == lang_short + ".po") {
300 load_from_file = lang_short;
305 // if it matched, load dictionary
306 if (load_from_file != "") {
307 //log_debug << "Loading dictionary for language \"" << lang << "\" from \"" << filename << "\"" << std::endl;
308 std::string pofile = *p + "/" + *filename;
310 IFileStream in(pofile);
311 read_po_file(dict, in);
312 } catch(std::exception& e) {
313 log_warning << "Error: Failure file opening: " << pofile << std::endl;
314 log_warning << e.what() << "" << std::endl;
319 PHYSFS_freeList(files);
327 std::set<std::string>
328 DictionaryManager::get_languages()
330 std::set<std::string> languages;
332 for (SearchPath::iterator p = search_path.begin(); p != search_path.end(); ++p)
334 char** files = PHYSFS_enumerateFiles(p->c_str());
337 log_warning << "Error: opendir() failed on " << *p << std::endl;
341 for(const char* const* file = files; *file != 0; file++) {
342 if(has_suffix(*file, ".po")) {
343 std::string filename = *file;
344 languages.insert(filename.substr(0, filename.length()-3));
347 PHYSFS_freeList(files);
354 DictionaryManager::set_language(const std::string& lang)
356 //log_debug << "set_language \"" << lang << "\"" << std::endl;
357 language = get_language_from_spec(lang);
358 //log_debug << "==> \"" << language << "\"" << std::endl;
359 current_dict = & (get_dictionary(language));
363 DictionaryManager::get_language() const
369 DictionaryManager::set_charset(const std::string& charset)
371 dictionaries.clear(); // changing charset invalidates cache
372 this->charset = charset;
373 set_language(language);
377 DictionaryManager::set_language_alias(const std::string& alias,
378 const std::string& language)
380 language_aliases.insert(std::make_pair(alias, language));
384 DictionaryManager::get_language_from_spec(const std::string& spec)
386 std::string lang = spec;
387 Aliases::iterator i = language_aliases.find(lang);
388 if(i != language_aliases.end()) {
392 std::string::size_type s = lang.find(".");
393 if(s != std::string::npos) {
394 lang = std::string(lang, 0, s);
398 if(s == std::string::npos) {
399 std::string lang_big = lang;
400 std::transform (lang_big.begin(), lang_big.end(), lang_big.begin(), toupper);
401 lang += "_" + lang_big;
409 DictionaryManager::add_directory(const std::string& pathname)
411 dictionaries.clear(); // adding directories invalidates cache
412 search_path.push_back(pathname);
413 set_language(language);
416 //---------------------------------------------------------------------------
418 Dictionary::Dictionary(const LanguageDef& language_, const std::string& charset_)
419 : language(language_), charset(charset_)
423 Dictionary::Dictionary()
429 Dictionary::get_charset() const
435 Dictionary::set_charset(const std::string& charset_)
441 Dictionary::set_language(const LanguageDef& lang)
447 Dictionary::translate(const std::string& msgid, const std::string& msgid2, int num)
449 PluralEntries::iterator i = plural_entries.find(msgid);
450 std::map<int, std::string>& msgstrs = i->second;
452 if (i != plural_entries.end() && !msgstrs.empty())
454 int g = language.plural(num);
455 std::map<int, std::string>::iterator j = msgstrs.find(g);
456 if (j != msgstrs.end())
462 // Return the first translation, in case we can't translate the specific number
463 return msgstrs.begin()->second;
468 #ifdef TRANSLATION_DEBUG
469 log_warning << "Couldn't translate: " << msgid << std::endl;
470 log_warning << "Candidates: " << std::endl;
471 for (PluralEntries::iterator i = plural_entries.begin(); i != plural_entries.end(); ++i)
472 log_debug << "'" << i->first << "'" << std::endl;
475 if (plural2_1(num)) // default to english rules
483 Dictionary::translate(const char* msgid)
485 Entries::iterator i = entries.find(msgid);
486 if (i != entries.end() && !i->second.empty())
488 return i->second.c_str();
492 #ifdef TRANSLATION_DBEUG
493 log_warning << "Couldn't translate: " << msgid << std::endl;
500 Dictionary::translate(const std::string& msgid)
502 Entries::iterator i = entries.find(msgid);
503 if (i != entries.end() && !i->second.empty())
509 #ifdef TRANSLATION_DBEUG
510 log_warning << "Couldn't translate: " << msgid << std::endl;
517 Dictionary::add_translation(const std::string& msgid, const std::string& ,
518 const std::map<int, std::string>& msgstrs)
520 // Do we need msgid2 for anything? its after all supplied to the
521 // translate call, so we just throw it away
522 plural_entries[msgid] = msgstrs;
526 Dictionary::add_translation(const std::string& msgid, const std::string& msgstr)
528 entries[msgid] = msgstr;
542 std::string from_charset;
543 std::string to_charset;
545 std::string current_msgid;
546 std::string current_msgid_plural;
547 std::map<int, std::string> msgstr_plural;
551 enum { WANT_MSGID, WANT_MSGSTR, WANT_MSGSTR_PLURAL, WANT_MSGID_PLURAL } state;
554 POFileReader(std::istream& in, Dictionary& dict_)
560 if(c == (char) 0xef) { // skip UTF-8 intro that some texteditors produce
569 void parse_header(const std::string& header)
571 // Seperate the header in lines
572 typedef std::vector<std::string> Lines;
575 std::string::size_type start = 0;
576 for(std::string::size_type i = 0; i < header.length(); ++i)
578 if (header[i] == '\n')
580 lines.push_back(header.substr(start, i - start));
585 for(Lines::iterator i = lines.begin(); i != lines.end(); ++i)
587 if (has_prefix(*i, "Content-Type: text/plain; charset=")) {
588 from_charset = i->substr(strlen("Content-Type: text/plain; charset="));
592 if (from_charset.empty() || from_charset == "CHARSET")
594 log_warning << "Error: Charset not specified for .po, fallback to ISO-8859-1" << std::endl;
595 from_charset = "ISO-8859-1";
598 to_charset = dict.get_charset();
599 if (to_charset.empty())
600 { // No charset requested from the dict, use utf-8
601 to_charset = "utf-8";
602 dict.set_charset(from_charset);
606 void add_token(const Token& token)
611 if (token.keyword == "msgid")
613 current_msgid = token.content;
614 state = WANT_MSGID_PLURAL;
616 else if (token.keyword.empty())
618 //log_warning << "Got EOF, everything looks ok." << std::endl;
622 log_warning << "tinygettext: expected 'msgid' keyword, got " << token.keyword << " at line " << line_num << std::endl;
626 case WANT_MSGID_PLURAL:
627 if (token.keyword == "msgid_plural")
629 current_msgid_plural = token.content;
630 state = WANT_MSGSTR_PLURAL;
640 if (token.keyword == "msgstr")
642 if (current_msgid == "")
643 { // .po Header is hidden in the msgid with the empty string
644 parse_header(token.content);
648 dict.add_translation(current_msgid, convert(token.content, from_charset, to_charset));
654 log_warning << "tinygettext: expected 'msgstr' keyword, got " << token.keyword << " at line " << line_num << std::endl;
658 case WANT_MSGSTR_PLURAL:
659 if (has_prefix(token.keyword, "msgstr["))
662 if (sscanf(token.keyword.c_str(), "msgstr[%d]", &num) != 1)
664 log_warning << "Error: Couldn't parse: " << token.keyword << std::endl;
668 msgstr_plural[num] = convert(token.content, from_charset, to_charset);
673 dict.add_translation(current_msgid, current_msgid_plural, msgstr_plural);
682 inline int getchar(std::istream& in)
690 void tokenize_po(std::istream& in)
692 enum State { READ_KEYWORD,
694 READ_CONTENT_IN_STRING,
697 State state = READ_KEYWORD;
701 while((c = getchar(in)) != EOF)
703 //log_debug << "Lexing char: " << char(c) << " " << state << std::endl;
709 state = SKIP_COMMENT;
721 } while((c = getchar(in)) != EOF && !isspace(static_cast<unsigned char>(c)));
724 state = READ_CONTENT;
729 while((c = getchar(in)) != EOF)
732 // Found start of content
733 state = READ_CONTENT_IN_STRING;
735 } else if (isspace(static_cast<unsigned char>(c))) {
737 } else { // Read something that may be a keyword
739 state = READ_KEYWORD;
747 case READ_CONTENT_IN_STRING:
752 if (c == 'n') token.content += '\n';
753 else if (c == 't') token.content += '\t';
754 else if (c == 'r') token.content += '\r';
755 else if (c == '"') token.content += '"';
756 else if (c == '\\') token.content += '\\';
759 log_warning << "Unhandled escape character: " << char(c) << std::endl;
764 log_warning << "Unterminated string" << std::endl;
766 } else if (c == '"') { // Content string is terminated
767 state = READ_CONTENT;
775 state = READ_KEYWORD;
784 void read_po_file(Dictionary& dict_, std::istream& in)
786 POFileReader reader(in, dict_);
789 } // namespace TinyGetText