Use the font language instead of guessing it

Similar to what we do with FontConfig.

Change-Id: Id01dabe0b52e4e3aea54073d42b719a924025920

1 files changed, 2 insertions, 158 deletions

diff --git a/vcl/win/gdi/salfont.cxx b/vcl/win/gdi/salfont.cxx
index b4341848d43d..c8c9893385de 100644
--- a/vcl/win/gdi/salfont.cxx
+++ b/vcl/win/gdi/salfont.cxx
@@ -134,150 +134,6 @@ RawFontData::RawFontData( HDC hDC, DWORD nTableTag )
 }
 
 // platform specific font substitution hooks for glyph fallback enhancement
-// TODO: move into i18n module (maybe merge with svx/ucsubset.*
-//       or merge with i18nutil/source/utility/unicode_data.h)
-struct Unicode2LangType
-{
-    sal_UCS4 mnMinCode;
-    sal_UCS4 mnMaxCode;
-    LanguageType mnLangID;
-};
-
-// entries marked with default-CJK get replaced with the default-CJK language
-#define LANGUAGE_DEFAULT_CJK 0xFFF0
-
-// map unicode ranges to languages supported by OOo
-// NOTE: due to the binary search used this list must be sorted by mnMinCode
-static Unicode2LangType aLangFromCodeChart[]= {
-    {0x0000, 0x007F, LANGUAGE_ENGLISH},             // Basic Latin
-    {0x0080, 0x024F, LANGUAGE_ENGLISH},             // Latin Extended-A and Latin Extended-B
-    {0x0250, 0x02AF, LANGUAGE_SYSTEM},              // IPA Extensions
-    {0x0370, 0x03FF, LANGUAGE_GREEK},               // Greek
-    {0x0590, 0x05FF, LANGUAGE_HEBREW},              // Hebrew
-    {0x0600, 0x06FF, LANGUAGE_ARABIC_PRIMARY_ONLY}, // Arabic
-    {0x0900, 0x097F, LANGUAGE_HINDI},               // Devanagari
-    {0x0980, 0x09FF, LANGUAGE_BENGALI},             // Bengali
-    {0x0A80, 0x0AFF, LANGUAGE_GUJARATI},            // Gujarati
-    {0x0B00, 0x0B7F, LANGUAGE_ODIA},                // Odia
-    {0x0B80, 0x0BFF, LANGUAGE_TAMIL},               // Tamil
-    {0x0C00, 0x0C7F, LANGUAGE_TELUGU},              // Telugu
-    {0x0C80, 0x0CFF, LANGUAGE_KANNADA},             // Kannada
-    {0x0D00, 0x0D7F, LANGUAGE_MALAYALAM},           // Malayalam
-    {0x0D80, 0x0D7F, LANGUAGE_SINHALESE_SRI_LANKA}, // Sinhala
-    {0x0E00, 0x0E7F, LANGUAGE_THAI},                // Thai
-    {0x0E80, 0x0EFF, LANGUAGE_LAO},                 // Lao
-    {0x0F00, 0x0FFF, LANGUAGE_TIBETAN},             // Tibetan
-    {0x1000, 0x109F, LANGUAGE_BURMESE},             // Burmese
-    {0x10A0, 0x10FF, LANGUAGE_GEORGIAN},            // Georgian
-    {0x1100, 0x11FF, LANGUAGE_KOREAN},              // Hangul Jamo, Korean-specific
-//  {0x1200, 0x139F, LANGUAGE_AMHARIC_ETHIOPIA},    // Ethiopic
-//  {0x1200, 0x139F, LANGUAGE_TIGRIGNA_ETHIOPIA},   // Ethiopic
-    {0x13A0, 0x13FF, LANGUAGE_CHEROKEE_UNITED_STATES}, // Cherokee
-//  {0x1400, 0x167F, LANGUAGE_CANADIAN_ABORIGINAL}, // Canadian Aboriginal Syllabics
-//  {0x1680, 0x169F, LANGUAGE_OGHAM},               // Ogham
-//  {0x16A0, 0x16F0, LANGUAGE_RUNIC},               // Runic
-//  {0x1700, 0x171F, LANGUAGE_TAGALOG},             // Tagalog
-//  {0x1720, 0x173F, LANGUAGE_HANUNOO},             // Hanunoo
-//  {0x1740, 0x175F, LANGUAGE_BUHID},               // Buhid
-//  {0x1760, 0x177F, LANGUAGE_TAGBANWA},            // Tagbanwa
-    {0x1780, 0x17FF, LANGUAGE_KHMER},               // Khmer
-    {0x18A0, 0x18AF, LANGUAGE_MONGOLIAN_MONGOLIAN_MONGOLIA}, // Mongolian
-//  {0x1900, 0x194F, LANGUAGE_LIMBU},               // Limbu
-//  {0x1950, 0x197F, LANGUAGE_TAILE},               // Tai Le
-//  {0x1980, 0x19DF, LANGUAGE_TAILUE},              // Tai Lue
-    {0x19E0, 0x19FF, LANGUAGE_KHMER},               // Khmer Symbols
-//  {0x1A00, 0x1A1F, LANGUAGE_BUGINESE},            // Buginese/Lontara
-//  {0x1B00, 0x1B7F, LANGUAGE_BALINESE},            // Balinese
-//  {0x1D00, 0x1DFF, LANGUAGE_NONE},                // Phonetic Symbols
-    {0x1E00, 0x1EFF, LANGUAGE_ENGLISH},             // Latin Extended Additional
-    {0x1F00, 0x1FFF, LANGUAGE_GREEK},               // Greek Extended
-    {0x2C60, 0x2C7F, LANGUAGE_ENGLISH},             // Latin Extended-C
-    {0x2E80, 0x2FFf, LANGUAGE_CHINESE_SIMPLIFIED},  // CJK Radicals Supplement + Kangxi Radical + Ideographic Description Characters
-    {0x3000, 0x303F, LANGUAGE_DEFAULT_CJK},         // CJK Symbols and punctuation
-    {0x3040, 0x30FF, LANGUAGE_JAPANESE},            // Japanese Hiragana + Katakana
-    {0x3100, 0x312F, LANGUAGE_CHINESE_TRADITIONAL}, // Bopomofo
-    {0x3130, 0x318F, LANGUAGE_KOREAN},              // Hangul Compatibility Jamo, Kocrean-specific
-    {0x3190, 0x319F, LANGUAGE_JAPANESE},            // Kanbun
-    {0x31A0, 0x31BF, LANGUAGE_CHINESE_TRADITIONAL}, // Bopomofo Extended
-    {0x31C0, 0x31EF, LANGUAGE_DEFAULT_CJK},         // CJK Ideographs
-    {0x31F0, 0x31FF, LANGUAGE_JAPANESE},            // Japanese Katakana Phonetic Extensions
-    {0x3200, 0x321F, LANGUAGE_KOREAN},              // Parenthesized Hangul
-    {0x3220, 0x325F, LANGUAGE_DEFAULT_CJK},         // Parenthesized Ideographs
-    {0x3260, 0x327F, LANGUAGE_KOREAN},              // Circled Hangul
-    {0x3280, 0x32CF, LANGUAGE_DEFAULT_CJK},         // Circled Ideographs
-    {0x32d0, 0x32FF, LANGUAGE_JAPANESE},            // Japanese Circled Katakana
-    {0x3400, 0x4DBF, LANGUAGE_DEFAULT_CJK},         // CJK Unified Ideographs Extension A
-    {0x4E00, 0x9FCF, LANGUAGE_DEFAULT_CJK},         // Unified CJK Ideographs
-    {0xA720, 0xA7FF, LANGUAGE_ENGLISH},             // Latin Extended-D
-    {0xAC00, 0xD7AF, LANGUAGE_KOREAN},              // Hangul Syllables, Korean-specific
-    {0xF900, 0xFAFF, LANGUAGE_DEFAULT_CJK},         // CJK Compatibility Ideographs
-    {0xFB00, 0xFB4F, LANGUAGE_HEBREW},              // Hebrew Presentation Forms
-    {0xFB50, 0xFDFF, LANGUAGE_ARABIC_PRIMARY_ONLY}, // Arabic Presentation Forms-A
-    {0xFE70, 0xFEFE, LANGUAGE_ARABIC_PRIMARY_ONLY}, // Arabic Presentation Forms-B
-    {0xFF65, 0xFF9F, LANGUAGE_JAPANESE},            // Japanese Halfwidth Katakana variant
-    {0xFFA0, 0xFFDC, LANGUAGE_KOREAN},              // Kocrean halfwidth hangual variant
-    {0x10140, 0x1018F, LANGUAGE_GREEK},             // Ancient Greak numbers
-    {0x1D200, 0x1D24F, LANGUAGE_GREEK},             // Ancient Greek Musical
-    {0x20000, 0x2A6DF, LANGUAGE_DEFAULT_CJK},       // CJK Unified Ideographs Extension B
-    {0x2F800, 0x2FA1F, LANGUAGE_DEFAULT_CJK}        // CJK Compatibility Ideographs Supplement
-};
-
-// get language matching to the missing char
-LanguageType MapCharToLanguage( sal_UCS4 uChar )
-{
-    // entries marked with default-CJK get replaced with the preferred CJK language
-    static bool bFirst = true;
-    if( bFirst )
-    {
-        bFirst = false;
-
-        // use method suggested in #i97086# to determine the systems default language
-        // TODO: move into i18npool or sal/osl/w32/nlsupport.c
-        LanguageType nDefaultLang = 0;
-        HKEY hKey = nullptr;
-        LONG lResult = ::RegOpenKeyExA( HKEY_LOCAL_MACHINE,
-            "SYSTEM\\CurrentControlSet\\Control\\Nls\\Language",
-            0, KEY_QUERY_VALUE, &hKey );
-        char aKeyValBuf[16];
-        DWORD nKeyValSize = sizeof(aKeyValBuf);
-        if( ERROR_SUCCESS == lResult )
-            lResult = RegQueryValueExA( hKey, "Default", nullptr, nullptr, reinterpret_cast<LPBYTE>(aKeyValBuf), &nKeyValSize );
-        aKeyValBuf[ sizeof(aKeyValBuf)-1 ] = '\0';
-        if( ERROR_SUCCESS == lResult )
-            nDefaultLang = (LanguageType)rtl_str_toInt32( aKeyValBuf, 16 );
-
-        // TODO: use the default-CJK language selected in
-        //  Tools->Options->LangSettings->Languages when it becomes available here
-        if( !nDefaultLang )
-            nDefaultLang = Application::GetSettings().GetUILanguageTag().getLanguageType();
-
-        LanguageType nDefaultCJK = MsLangId::isCJK(nDefaultLang) ? nDefaultLang : LANGUAGE_CHINESE;
-
-        // change the marked entries to preferred language
-        static const int nCount = SAL_N_ELEMENTS(aLangFromCodeChart);
-        for( int i = 0; i < nCount; ++i )
-        {
-            if( aLangFromCodeChart[ i].mnLangID == LANGUAGE_DEFAULT_CJK )
-                aLangFromCodeChart[ i].mnLangID = nDefaultCJK;
-        }
-    }
-
-    // binary search
-    int nLow = 0;
-    int nHigh = SAL_N_ELEMENTS(aLangFromCodeChart) - 1;
-    while( nLow <= nHigh )
-    {
-        int nMiddle = (nHigh + nLow) / 2;
-        if( uChar < aLangFromCodeChart[ nMiddle].mnMinCode )
-            nHigh = nMiddle - 1;
-        else if( uChar > aLangFromCodeChart[ nMiddle].mnMaxCode )
-            nLow = nMiddle + 1;
-        else
-            return aLangFromCodeChart[ nMiddle].mnLangID;
-    }
-
-    return LANGUAGE_DONTKNOW;
-}
 
 class WinPreMatchFontSubstititution
 :    public ImplPreMatchFontSubstitution
@@ -404,22 +260,10 @@ bool WinPreMatchFontSubstititution::FindFontSubstitute(FontSelectPattern& rFontS
 bool WinGlyphFallbackSubstititution::FindFontSubstitute( FontSelectPattern& rFontSelData, OUString& rMissingChars ) const
 {
     // guess a locale matching to the missing chars
-    LanguageType eLang = LANGUAGE_DONTKNOW;
+    LanguageType eLang = rFontSelData.meLanguage;
     LanguageTag aLanguageTag( eLang);
 
-    sal_Int32 nStrIdx = 0;
-    const sal_Int32 nStrLen = rMissingChars.getLength();
-    while( nStrIdx < nStrLen )
-    {
-        const sal_UCS4 uChar = rMissingChars.iterateCodePoints( &nStrIdx );
-        eLang = MapCharToLanguage( uChar );
-        if( eLang == LANGUAGE_DONTKNOW )
-            continue;
-        aLanguageTag.reset( eLang);
-        break;
-    }
-
-    // fall back to default UI locale if the missing characters are inconclusive
+    // fall back to default UI locale if the font language is inconclusive
     if( eLang == LANGUAGE_DONTKNOW )
         aLanguageTag = Application::GetSettings().GetUILanguageTag();