X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=src%2FThesaurus.cpp;h=bf657762c1c2d4a423d3998a7dc8245734d3e132;hb=53a72a45266feb86c13f47615a4cbb15053f3c63;hp=91aa9f649ad1315a5625532c41f757e284824913;hpb=5a7524ac9bd67e1151afa342917b2d3b7941929b;p=lyx.git diff --git a/src/Thesaurus.cpp b/src/Thesaurus.cpp index 91aa9f649a..bf657762c1 100644 --- a/src/Thesaurus.cpp +++ b/src/Thesaurus.cpp @@ -4,6 +4,7 @@ * Licence details can be found in the file COPYING. * * \author John Levon + * \author Jürgen Spitzmüller * * Full author contact details are available in file CREDITS. */ @@ -12,84 +13,229 @@ #include "Thesaurus.h" -#include "gettext.h" +#include "LyXRC.h" +#include "support/FileNameList.h" +#include "support/Package.h" +#include "support/debug.h" +#include "support/filetools.h" +#include "support/gettext.h" #include "support/lstrings.h" +#include "support/os.h" + +#include "support/mythes/mythes.hxx" #include "frontends/alert.h" #include +#include +#include +using namespace std; +using namespace lyx::support; +using namespace lyx::support::os; namespace lyx { -#ifdef HAVE_LIBAIKSAURUS -using support::bformat; +namespace { -using std::sort; -using std::string; +typedef std::map Thesauri; +} // namespace anon -Thesaurus::Thesaurus() - : aik_(new Aiksaurus) -{} +struct Thesaurus::Private +{ + ~Private() + { + for (Thesauri::iterator it = thes_.begin(); + it != thes_.end(); ++it) { + delete it->second; + } + } + /// + bool thesaurusAvailable(docstring const & lang) const + { + for (Thesauri::const_iterator it = thes_.begin(); + it != thes_.end(); ++it) { + if (it->first == lang) + if (it->second) + return true; + } + return false; + } + /// + typedef std::pair ThesFiles; + /// + ThesFiles getThesaurus(string const & path, docstring const & lang); + ThesFiles getThesaurus(docstring const & lang); + /// add a thesaurus to the list + bool addThesaurus(docstring const & lang); -Thesaurus::~Thesaurus() + /// the thesauri + Thesauri thes_; + + /// the location below system/user directory + /// there the data+idx files lookup will happen + const string dataDirectory(void) { return "thes"; } + +}; + + +pair Thesaurus::Private::getThesaurus(string const & path, docstring const & lang) +{ + FileName base(path); + if (!base.isDirectory()) { + return make_pair(string(), string()); + } + FileNameList const idx_files = base.dirList("idx"); + FileNameList const data_files = base.dirList("dat"); + string idx; + string data; + string basename; + + LYXERR(Debug::FILES, "thesaurus path: " << path); + for (FileNameList::const_iterator it = idx_files.begin(); it != idx_files.end(); ++it) { + basename = it->onlyFileNameWithoutExt(); + if (contains(basename, to_ascii(lang))) { + ifstream ifs(it->absFileName().c_str()); + if (ifs) { + // check for appropriate version of index file + string encoding; // first line is encoding + int items = 0; // second line is no. of items + getline(ifs,encoding); + ifs >> items; + if (ifs.fail()) { + LYXERR(Debug::FILES, "ignore irregular thesaurus idx file: " << it->absFileName()); + continue; + } + if (encoding.length() == 0 || encoding.find_first_of(',') != string::npos) { + LYXERR(Debug::FILES, "ignore version1 thesaurus idx file: " << it->absFileName()); + continue; + } + } + idx = it->absFileName(); + LYXERR(Debug::FILES, "selected thesaurus idx file: " << idx); + break; + } + } + if (idx.empty()) { + return make_pair(string(), string()); + } + for (support::FileNameList::const_iterator it = data_files.begin(); it != data_files.end(); ++it) { + if (contains(it->onlyFileName(), basename)) { + data = it->absFileName(); + LYXERR(Debug::FILES, "selected thesaurus data file: " << data); + break; + } + } + return make_pair(idx, data); +} + + +pair Thesaurus::Private::getThesaurus(docstring const & lang) +{ + string const thes_path = external_path(lyxrc.thesaurusdir_path); + pair result ; + + if (thesaurusAvailable(lang)) + return make_pair(string(), string()); + + if (!thes_path.empty()) { + result = getThesaurus(thes_path, lang); + } + if (result.first.empty() || result.second.empty()) { + string const sys_path = external_path(addName(lyx::support::package().system_support().absFileName(),dataDirectory())) ; + result = getThesaurus(sys_path, lang); + } + if (result.first.empty() || result.second.empty()) { + string const user_path = external_path(addName(lyx::support::package().user_support().absFileName(),dataDirectory())) ; + result = getThesaurus(user_path, lang); + } + return result; +} + + +bool Thesaurus::Private::addThesaurus(docstring const & lang) +{ + if (thesaurusAvailable(lang)) + return true; + + ThesFiles files = getThesaurus(lang); + string const idx = files.first; + string const data = files.second; + + if (idx.empty() || data.empty()) + return false; + + char const * af = idx.c_str(); + char const * df = data.c_str(); + thes_[lang] = new MyThes(af, df); + return true; +} + + +bool Thesaurus::thesaurusAvailable(docstring const & lang) const +{ + return d->thesaurusAvailable(lang); +} + + +bool Thesaurus::thesaurusInstalled(docstring const & lang) const { - delete aik_; + if (thesaurusAvailable(lang)) + return true; + pair files = d->getThesaurus(lang); + return (!files.first.empty() && !files.second.empty()); } -Thesaurus::Meanings Thesaurus::lookup(docstring const & t) +Thesaurus::Meanings Thesaurus::lookup(docstring const & t, docstring const & lang) { Meanings meanings; + MyThes * mythes = 0; - // aiksaurus is for english text only, therefore it does not work - // with non-ascii strings. - // The interface of the Thesaurus class uses docstring because a - // non-english thesaurus is possible in theory. - if (!support::isAscii(t)) - // to_ascii() would assert + if (!d->addThesaurus(lang)) return meanings; - string const text = to_ascii(t); - - docstring error = from_ascii(aik_->error()); - if (!error.empty()) { - static bool sent_error = false; - if (!sent_error) { - frontend::Alert::error(_("Thesaurus failure"), - bformat(_("Aiksaurus returned the following error:\n\n%1$s."), - error)); - sent_error = true; + for (Thesauri::const_iterator it = d->thes_.begin(); + it != d->thes_.end(); ++it) { + if (it->first == lang) { + mythes = it->second; + break; } - return meanings; } - if (!aik_->find(text.c_str())) + + if (!mythes) return meanings; - // weird api, but ... + string const encoding = mythes->get_th_encoding(); + + mentry * pmean; + string const text = to_iconv_encoding(support::lowercase(t), encoding); + int len = strlen(text.c_str()); + int count = mythes->Lookup(text.c_str(), len, &pmean); + if (!count) + return meanings; - int prev_meaning = -1; - int cur_meaning; + // don't change value of pmean or count + // they are needed for the CleanUpAfterLookup routine + mentry * pm = pmean; docstring meaning; - - // correct, returns "" at the end - string ret = aik_->next(cur_meaning); - - while (!ret.empty()) { - if (cur_meaning != prev_meaning) { - meaning = from_ascii(ret); - ret = aik_->next(cur_meaning); - prev_meaning = cur_meaning; - } else { - if (ret != text) - meanings[meaning].push_back(from_ascii(ret)); + for (int i = 0; i < count; i++) { + vector ret; + meaning = from_iconv_encoding(string(pm->defn), encoding); + // remove silly item + if (support::prefixIs(meaning, '-')) + meaning = support::ltrim(meaning, "- "); + for (int j = 0; j < pm->count; j++) { + ret.push_back(from_iconv_encoding(string(pm->psyns[j]), encoding)); } - - ret = aik_->next(cur_meaning); + meanings[meaning] = ret; + ++pm; } + // now clean up all allocated memory + mythes->CleanUpAfterLookup(&pmean, count); for (Meanings::iterator it = meanings.begin(); it != meanings.end(); ++it) @@ -98,25 +244,17 @@ Thesaurus::Meanings Thesaurus::lookup(docstring const & t) return meanings; } -#else -Thesaurus::Thesaurus() +Thesaurus::Thesaurus() : d(new Thesaurus::Private) { } Thesaurus::~Thesaurus() { + delete d; } - -Thesaurus::Meanings Thesaurus::lookup(docstring const &) -{ - return Meanings(); -} - -#endif // HAVE_LIBAIKSAURUS - // Global instance Thesaurus thesaurus;