Use LanguageTagIcu::getDisplayName() for on-the-fly language list entries
So the for {tag} entries "too geeky" argument of spell-checker
dictionary developers and language enthusiasts will be moot in
future and we can stop adding 100 more tags if it's not for locale
data.
Change-Id: Ic44fe5b2f794ddb258e56ec073f310ccaf6e470d
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/120799
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>
diff --git a/svtools/source/misc/langtab.cxx b/svtools/source/misc/langtab.cxx
index 7bd30f3..9981cf6 100644
--- a/svtools/source/misc/langtab.cxx
+++ b/svtools/source/misc/langtab.cxx
@@ -29,8 +29,11 @@
#include <i18nlangtag/lang.h>
#include <i18nlangtag/mslangid.hxx>
#include <i18nlangtag/languagetag.hxx>
#include <i18nlangtag/languagetagicu.hxx>
#include <sal/log.hxx>
#include <vcl/svapp.hxx>
#include <vcl/settings.hxx>
#include <svtools/svtresid.hxx>
#include <svtools/langtab.hxx>
#include <unotools/syslocale.hxx>
@@ -156,16 +159,25 @@ OUString ApplyLreOrRleEmbedding( const OUString &rText )
return aRes;
}
static OUString lcl_getDescription( std::u16string_view rBcp47 )
static OUString lcl_getDescription( const LanguageTag& rTag )
{
// Place in curly brackets, so all on-the-fly tags are grouped together at
// the top of a listbox (but behind the "[None]" entry), and not sprinkled
// all over, which alphabetically might make sense in an English UI only
// anyway. Also a visual indicator that it is a programmatical name, IMHO.
/* TODO: pulling descriptive names (language, script, country, subtags)
* from liblangtag or ISO databases might be nice, but those are English
* only. Maybe ICU, that has translations for language and country. */
return OUString::Concat("{") + rBcp47 + "}";
OUString aStr( LanguageTagIcu::getDisplayName( rTag, Application::GetSettings().GetUILanguageTag()));
if (aStr.isEmpty() || aStr == rTag.getBcp47())
{
// Place in curly brackets, so all on-the-fly tags without display name
// are grouped together at the top of a listbox (but behind the
// "[None]" entry), and not sprinkled all over, which alphabetically
// might make sense in an English UI only anyway. Also a visual
// indicator that it is a programmatical name, IMHO.
return OUString::Concat("{") + aStr + "}";
}
else
{
// The ICU display name might be identical to a predefined name or even
// to another tag's ICU name; clarify that this is a generated name and
// append the language tag in curly brackets to distinguish.
return aStr + " {" + rTag.getBcp47() + "}";
}
}
SvtLanguageTableImpl::SvtLanguageTableImpl()
@@ -193,7 +205,7 @@ SvtLanguageTableImpl::SvtLanguageTableImpl()
aLang.setScriptType(LanguageTag::ScriptType(nType));
sal_uInt32 nPos = FindIndex(nLangType);
if (nPos == RESARRAY_INDEX_NOTFOUND)
AddItem((aName.isEmpty() ? lcl_getDescription(rBcp47) : aName), nLangType);
AddItem((aName.isEmpty() ? lcl_getDescription(aLang) : aName), nLangType);
}
}
}
@@ -219,8 +231,8 @@ OUString SvtLanguageTableImpl::GetString( const LanguageType eType ) const
if ( RESARRAY_INDEX_NOTFOUND != nPos && nPos < GetEntryCount() )
return m_aStrings[nPos].first;
//Rather than return a fairly useless "Unknown" name, return a geeky but usable-in-a-pinch lang-tag
OUString sLangTag( lcl_getDescription( LanguageTag::convertToBcp47(eType)));
// Obtain from ICU, or a geeky but usable-in-a-pinch lang-tag.
OUString sLangTag( lcl_getDescription( LanguageTag(eType)));
SAL_WARN("svtools.misc", "Language: 0x"
<< std::hex << eType
<< " with unknown name, so returning lang-tag of: "
@@ -287,7 +299,7 @@ LanguageType SvtLanguageTable::GetLanguageTypeAtIndex( sal_uInt32 nIndex )
sal_uInt32 SvtLanguageTable::AddLanguageTag( const LanguageTag& rLanguageTag )
{
return theLanguageTable::get().AddItem( lcl_getDescription(rLanguageTag.getBcp47()),
return theLanguageTable::get().AddItem( lcl_getDescription(rLanguageTag),
rLanguageTag.getLanguageType());
}