1*b0844812SAndrew Rist /**************************************************************
2cdf0e10cSrcweir *
3*b0844812SAndrew Rist * Licensed to the Apache Software Foundation (ASF) under one
4*b0844812SAndrew Rist * or more contributor license agreements. See the NOTICE file
5*b0844812SAndrew Rist * distributed with this work for additional information
6*b0844812SAndrew Rist * regarding copyright ownership. The ASF licenses this file
7*b0844812SAndrew Rist * to you under the Apache License, Version 2.0 (the
8*b0844812SAndrew Rist * "License"); you may not use this file except in compliance
9*b0844812SAndrew Rist * with the License. You may obtain a copy of the License at
10*b0844812SAndrew Rist *
11*b0844812SAndrew Rist * http://www.apache.org/licenses/LICENSE-2.0
12*b0844812SAndrew Rist *
13*b0844812SAndrew Rist * Unless required by applicable law or agreed to in writing,
14*b0844812SAndrew Rist * software distributed under the License is distributed on an
15*b0844812SAndrew Rist * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
16*b0844812SAndrew Rist * KIND, either express or implied. See the License for the
17*b0844812SAndrew Rist * specific language governing permissions and limitations
18*b0844812SAndrew Rist * under the License.
19*b0844812SAndrew Rist *
20*b0844812SAndrew Rist *************************************************************/
21*b0844812SAndrew Rist
22*b0844812SAndrew Rist
23cdf0e10cSrcweir
24cdf0e10cSrcweir // MARKER(update_precomp.py): autogen include statement, do not remove
25cdf0e10cSrcweir #include "precompiled_lingucomponent.hxx"
26cdf0e10cSrcweir
27cdf0e10cSrcweir #if defined(WNT)
28cdf0e10cSrcweir #include <tools/prewin.h>
29cdf0e10cSrcweir #endif
30cdf0e10cSrcweir
31cdf0e10cSrcweir #if defined(WNT)
32cdf0e10cSrcweir #include <Windows.h>
33cdf0e10cSrcweir #endif
34cdf0e10cSrcweir
35cdf0e10cSrcweir #if defined(WNT)
36cdf0e10cSrcweir #include <tools/postwin.h>
37cdf0e10cSrcweir #endif
38cdf0e10cSrcweir
39cdf0e10cSrcweir
40cdf0e10cSrcweir #include <osl/thread.h>
41cdf0e10cSrcweir #include <osl/file.hxx>
42cdf0e10cSrcweir #include <tools/debug.hxx>
43cdf0e10cSrcweir #include <tools/urlobj.hxx>
44cdf0e10cSrcweir #include <i18npool/mslangid.hxx>
45cdf0e10cSrcweir #include <unotools/lingucfg.hxx>
46cdf0e10cSrcweir #include <unotools/pathoptions.hxx>
47cdf0e10cSrcweir #include <rtl/ustring.hxx>
48cdf0e10cSrcweir #include <rtl/string.hxx>
49cdf0e10cSrcweir #include <rtl/tencinfo.h>
50cdf0e10cSrcweir #include <linguistic/misc.hxx>
51cdf0e10cSrcweir
52cdf0e10cSrcweir #include <set>
53cdf0e10cSrcweir #include <vector>
54cdf0e10cSrcweir #include <string.h>
55cdf0e10cSrcweir
56cdf0e10cSrcweir #include <lingutil.hxx>
57cdf0e10cSrcweir
58cdf0e10cSrcweir
59cdf0e10cSrcweir
60cdf0e10cSrcweir
61cdf0e10cSrcweir using ::com::sun::star::lang::Locale;
62cdf0e10cSrcweir using namespace ::com::sun::star;
63cdf0e10cSrcweir
64cdf0e10cSrcweir #if 0
65cdf0e10cSrcweir //////////////////////////////////////////////////////////////////////
66cdf0e10cSrcweir
67cdf0e10cSrcweir String GetDirectoryPathFromFileURL( const String &rFileURL )
68cdf0e10cSrcweir {
69cdf0e10cSrcweir // get file URL
70cdf0e10cSrcweir INetURLObject aURLObj;
71cdf0e10cSrcweir aURLObj.SetSmartProtocol( INET_PROT_FILE );
72cdf0e10cSrcweir aURLObj.SetSmartURL( rFileURL );
73cdf0e10cSrcweir aURLObj.removeSegment();
74cdf0e10cSrcweir DBG_ASSERT( !aURLObj.HasError(), "invalid URL" );
75cdf0e10cSrcweir String aRes = aURLObj.GetMainURL( INetURLObject::DECODE_TO_IURI );
76cdf0e10cSrcweir return aRes;
77cdf0e10cSrcweir }
78cdf0e10cSrcweir #endif
79cdf0e10cSrcweir
80cdf0e10cSrcweir #if defined(WNT)
Win_GetShortPathName(const rtl::OUString & rLongPathName)81cdf0e10cSrcweir rtl::OString Win_GetShortPathName( const rtl::OUString &rLongPathName )
82cdf0e10cSrcweir {
83cdf0e10cSrcweir rtl::OString aRes;
84cdf0e10cSrcweir
85cdf0e10cSrcweir sal_Unicode aShortBuffer[1024] = {0};
86cdf0e10cSrcweir sal_Int32 nShortBufSize = sizeof( aShortBuffer ) / sizeof( aShortBuffer[0] );
87cdf0e10cSrcweir
88cdf0e10cSrcweir // use the version of 'GetShortPathName' that can deal with Unicode...
89cdf0e10cSrcweir sal_Int32 nShortLen = GetShortPathNameW(
90cdf0e10cSrcweir reinterpret_cast<LPCWSTR>( rLongPathName.getStr() ),
91cdf0e10cSrcweir reinterpret_cast<LPWSTR>( aShortBuffer ),
92cdf0e10cSrcweir nShortBufSize );
93cdf0e10cSrcweir
94cdf0e10cSrcweir if (nShortLen < nShortBufSize) // conversion successful?
95cdf0e10cSrcweir aRes = rtl::OString( OU2ENC( rtl::OUString( aShortBuffer, nShortLen ), osl_getThreadTextEncoding()) );
96cdf0e10cSrcweir else
97cdf0e10cSrcweir DBG_ERROR( "Win_GetShortPathName: buffer to short" );
98cdf0e10cSrcweir
99cdf0e10cSrcweir return aRes;
100cdf0e10cSrcweir }
101cdf0e10cSrcweir #endif //defined(WNT)
102cdf0e10cSrcweir
103cdf0e10cSrcweir //////////////////////////////////////////////////////////////////////
104cdf0e10cSrcweir
105cdf0e10cSrcweir // build list of old style diuctionaries (not as extensions) to use.
106cdf0e10cSrcweir // User installed dictionaries (the ones residing in the user paths)
107cdf0e10cSrcweir // will get precedence over system installed ones for the same language.
GetOldStyleDics(const char * pDicType)108cdf0e10cSrcweir std::vector< SvtLinguConfigDictionaryEntry > GetOldStyleDics( const char *pDicType )
109cdf0e10cSrcweir {
110cdf0e10cSrcweir std::vector< SvtLinguConfigDictionaryEntry > aRes;
111cdf0e10cSrcweir
112cdf0e10cSrcweir if (!pDicType)
113cdf0e10cSrcweir return aRes;
114cdf0e10cSrcweir
115cdf0e10cSrcweir rtl::OUString aFormatName;
116cdf0e10cSrcweir String aDicExtension;
117cdf0e10cSrcweir #ifdef SYSTEM_DICTS
118cdf0e10cSrcweir rtl::OUString aSystemDir;
119cdf0e10cSrcweir rtl::OUString aSystemPrefix;
120cdf0e10cSrcweir rtl::OUString aSystemSuffix;
121cdf0e10cSrcweir #endif
122cdf0e10cSrcweir if (strcmp( pDicType, "DICT" ) == 0)
123cdf0e10cSrcweir {
124cdf0e10cSrcweir aFormatName = A2OU("DICT_SPELL");
125cdf0e10cSrcweir aDicExtension = String::CreateFromAscii( ".dic" );
126cdf0e10cSrcweir #ifdef SYSTEM_DICTS
127cdf0e10cSrcweir aSystemDir = A2OU( DICT_SYSTEM_DIR );
128cdf0e10cSrcweir aSystemSuffix = aDicExtension;
129cdf0e10cSrcweir #endif
130cdf0e10cSrcweir }
131cdf0e10cSrcweir else if (strcmp( pDicType, "HYPH" ) == 0)
132cdf0e10cSrcweir {
133cdf0e10cSrcweir aFormatName = A2OU("DICT_HYPH");
134cdf0e10cSrcweir aDicExtension = String::CreateFromAscii( ".dic" );
135cdf0e10cSrcweir #ifdef SYSTEM_DICTS
136cdf0e10cSrcweir aSystemDir = A2OU( HYPH_SYSTEM_DIR );
137cdf0e10cSrcweir aSystemPrefix = A2OU( "hyph_" );
138cdf0e10cSrcweir aSystemSuffix = aDicExtension;
139cdf0e10cSrcweir #endif
140cdf0e10cSrcweir }
141cdf0e10cSrcweir else if (strcmp( pDicType, "THES" ) == 0)
142cdf0e10cSrcweir {
143cdf0e10cSrcweir aFormatName = A2OU("DICT_THES");
144cdf0e10cSrcweir aDicExtension = String::CreateFromAscii( ".dat" );
145cdf0e10cSrcweir #ifdef SYSTEM_DICTS
146cdf0e10cSrcweir aSystemDir = A2OU( THES_SYSTEM_DIR );
147cdf0e10cSrcweir aSystemPrefix = A2OU( "th_" );
148cdf0e10cSrcweir aSystemSuffix = A2OU( "_v2.dat" );
149cdf0e10cSrcweir #endif
150cdf0e10cSrcweir }
151cdf0e10cSrcweir
152cdf0e10cSrcweir
153cdf0e10cSrcweir if (aFormatName.getLength() == 0 || aDicExtension.Len() == 0)
154cdf0e10cSrcweir return aRes;
155cdf0e10cSrcweir
156cdf0e10cSrcweir // set of languages to remember the language where it is already
157cdf0e10cSrcweir // decided to make use of the dictionary.
158cdf0e10cSrcweir std::set< LanguageType > aDicLangInUse;
159cdf0e10cSrcweir
160cdf0e10cSrcweir #ifdef SYSTEM_DICTS
161cdf0e10cSrcweir osl::Directory aSystemDicts(aSystemDir);
162cdf0e10cSrcweir if (aSystemDicts.open() == osl::FileBase::E_None)
163cdf0e10cSrcweir {
164cdf0e10cSrcweir osl::DirectoryItem aItem;
165cdf0e10cSrcweir osl::FileStatus aFileStatus(FileStatusMask_FileURL);
166cdf0e10cSrcweir while (aSystemDicts.getNextItem(aItem) == osl::FileBase::E_None)
167cdf0e10cSrcweir {
168cdf0e10cSrcweir aItem.getFileStatus(aFileStatus);
169cdf0e10cSrcweir rtl::OUString sPath = aFileStatus.getFileURL();
170cdf0e10cSrcweir if (sPath.lastIndexOf(aSystemSuffix) == sPath.getLength()-aSystemSuffix.getLength())
171cdf0e10cSrcweir {
172cdf0e10cSrcweir sal_Int32 nStartIndex = sPath.lastIndexOf(sal_Unicode('/')) + 1;
173cdf0e10cSrcweir if (!sPath.match(aSystemPrefix, nStartIndex))
174cdf0e10cSrcweir continue;
175cdf0e10cSrcweir rtl::OUString sChunk = sPath.copy(0, sPath.getLength() - aSystemSuffix.getLength());
176cdf0e10cSrcweir sal_Int32 nIndex = nStartIndex + aSystemPrefix.getLength();
177cdf0e10cSrcweir rtl::OUString sLang = sChunk.getToken( 0, '_', nIndex );
178cdf0e10cSrcweir if (!sLang.getLength())
179cdf0e10cSrcweir continue;
180cdf0e10cSrcweir rtl::OUString sRegion;
181cdf0e10cSrcweir if (nIndex != -1)
182cdf0e10cSrcweir sRegion = sChunk.copy( nIndex, sChunk.getLength() - nIndex );
183cdf0e10cSrcweir
184cdf0e10cSrcweir // Thus we first get the language of the dictionary
185cdf0e10cSrcweir LanguageType nLang = MsLangId::convertIsoNamesToLanguage(
186cdf0e10cSrcweir sLang, sRegion );
187cdf0e10cSrcweir
188cdf0e10cSrcweir if (aDicLangInUse.count( nLang ) == 0)
189cdf0e10cSrcweir {
190cdf0e10cSrcweir // remember the new language in use
191cdf0e10cSrcweir aDicLangInUse.insert( nLang );
192cdf0e10cSrcweir
193cdf0e10cSrcweir // add the dictionary to the resulting vector
194cdf0e10cSrcweir SvtLinguConfigDictionaryEntry aDicEntry;
195cdf0e10cSrcweir aDicEntry.aLocations.realloc(1);
196cdf0e10cSrcweir aDicEntry.aLocaleNames.realloc(1);
197cdf0e10cSrcweir rtl::OUString aLocaleName( MsLangId::convertLanguageToIsoString( nLang ) );
198cdf0e10cSrcweir aDicEntry.aLocations[0] = sPath;
199cdf0e10cSrcweir aDicEntry.aFormatName = aFormatName;
200cdf0e10cSrcweir aDicEntry.aLocaleNames[0] = aLocaleName;
201cdf0e10cSrcweir aRes.push_back( aDicEntry );
202cdf0e10cSrcweir }
203cdf0e10cSrcweir }
204cdf0e10cSrcweir }
205cdf0e10cSrcweir }
206cdf0e10cSrcweir
207cdf0e10cSrcweir #endif
208cdf0e10cSrcweir
209cdf0e10cSrcweir return aRes;
210cdf0e10cSrcweir }
211cdf0e10cSrcweir
212cdf0e10cSrcweir
MergeNewStyleDicsAndOldStyleDics(std::list<SvtLinguConfigDictionaryEntry> & rNewStyleDics,const std::vector<SvtLinguConfigDictionaryEntry> & rOldStyleDics)213cdf0e10cSrcweir void MergeNewStyleDicsAndOldStyleDics(
214cdf0e10cSrcweir std::list< SvtLinguConfigDictionaryEntry > &rNewStyleDics,
215cdf0e10cSrcweir const std::vector< SvtLinguConfigDictionaryEntry > &rOldStyleDics )
216cdf0e10cSrcweir {
217cdf0e10cSrcweir // get list of languages supported by new style dictionaries
218cdf0e10cSrcweir std::set< LanguageType > aNewStyleLanguages;
219cdf0e10cSrcweir std::list< SvtLinguConfigDictionaryEntry >::const_iterator aIt;
220cdf0e10cSrcweir for (aIt = rNewStyleDics.begin() ; aIt != rNewStyleDics.end(); ++aIt)
221cdf0e10cSrcweir {
222cdf0e10cSrcweir const uno::Sequence< rtl::OUString > aLocaleNames( aIt->aLocaleNames );
223cdf0e10cSrcweir sal_Int32 nLocaleNames = aLocaleNames.getLength();
224cdf0e10cSrcweir for (sal_Int32 k = 0; k < nLocaleNames; ++k)
225cdf0e10cSrcweir {
226cdf0e10cSrcweir LanguageType nLang = MsLangId::convertIsoStringToLanguage( aLocaleNames[k] );
227cdf0e10cSrcweir aNewStyleLanguages.insert( nLang );
228cdf0e10cSrcweir }
229cdf0e10cSrcweir }
230cdf0e10cSrcweir
231cdf0e10cSrcweir // now check all old style dictionaries if they will add a not yet
232cdf0e10cSrcweir // added language. If so add them to the resulting vector
233cdf0e10cSrcweir std::vector< SvtLinguConfigDictionaryEntry >::const_iterator aIt2;
234cdf0e10cSrcweir for (aIt2 = rOldStyleDics.begin(); aIt2 != rOldStyleDics.end(); ++aIt2)
235cdf0e10cSrcweir {
236cdf0e10cSrcweir sal_Int32 nOldStyleDics = aIt2->aLocaleNames.getLength();
237cdf0e10cSrcweir
238cdf0e10cSrcweir // old style dics should only have one language listed...
239cdf0e10cSrcweir DBG_ASSERT( nOldStyleDics, "old style dictionary with more then one language found!");
240cdf0e10cSrcweir if (nOldStyleDics > 0)
241cdf0e10cSrcweir {
242cdf0e10cSrcweir LanguageType nLang = MsLangId::convertIsoStringToLanguage( aIt2->aLocaleNames[0] );
243cdf0e10cSrcweir
244cdf0e10cSrcweir if (nLang == LANGUAGE_DONTKNOW || nLang == LANGUAGE_NONE)
245cdf0e10cSrcweir {
246cdf0e10cSrcweir DBG_ERROR( "old style dictionary with invalid language found!" );
247cdf0e10cSrcweir continue;
248cdf0e10cSrcweir }
249cdf0e10cSrcweir
250cdf0e10cSrcweir // language not yet added?
251cdf0e10cSrcweir if (aNewStyleLanguages.count( nLang ) == 0)
252cdf0e10cSrcweir rNewStyleDics.push_back( *aIt2 );
253cdf0e10cSrcweir }
254cdf0e10cSrcweir else
255cdf0e10cSrcweir {
256cdf0e10cSrcweir DBG_ERROR( "old style dictionary with no language found!" );
257cdf0e10cSrcweir }
258cdf0e10cSrcweir }
259cdf0e10cSrcweir }
260cdf0e10cSrcweir
261cdf0e10cSrcweir
getTextEncodingFromCharset(const sal_Char * pCharset)262cdf0e10cSrcweir rtl_TextEncoding getTextEncodingFromCharset(const sal_Char* pCharset)
263cdf0e10cSrcweir {
264cdf0e10cSrcweir // default result: used to indicate that we failed to get the proper encoding
265cdf0e10cSrcweir rtl_TextEncoding eRet = RTL_TEXTENCODING_DONTKNOW;
266cdf0e10cSrcweir
267cdf0e10cSrcweir if (pCharset)
268cdf0e10cSrcweir {
269cdf0e10cSrcweir eRet = rtl_getTextEncodingFromMimeCharset(pCharset);
270cdf0e10cSrcweir if (eRet == RTL_TEXTENCODING_DONTKNOW)
271cdf0e10cSrcweir eRet = rtl_getTextEncodingFromUnixCharset(pCharset);
272cdf0e10cSrcweir if (eRet == RTL_TEXTENCODING_DONTKNOW)
273cdf0e10cSrcweir {
274cdf0e10cSrcweir if (strcmp("ISCII-DEVANAGARI", pCharset) == 0)
275cdf0e10cSrcweir eRet = RTL_TEXTENCODING_ISCII_DEVANAGARI;
276cdf0e10cSrcweir }
277cdf0e10cSrcweir }
278cdf0e10cSrcweir return eRet;
279cdf0e10cSrcweir }
280cdf0e10cSrcweir
281cdf0e10cSrcweir //////////////////////////////////////////////////////////////////////
282cdf0e10cSrcweir
283