private bool RegionalDialectsFilter(LanguageInfo li)
{
if (IncludeRegionalDialects)
return true;
// always include Chinese languages with region codes
if (li.LanguageTag.IsOneOf("zh-CN", "zh-TW"))
return true;
return string.IsNullOrEmpty(IetfLanguageTag.GetRegionPart(li.LanguageTag));
}
private LanguageInfo GetOrCreateLanguageFromCode(string code, string countryName)
{
LanguageInfo language;
if (!_codeToLanguageIndex.TryGetValue(code, out language))
{
language = new LanguageInfo {
LanguageTag = code
};
_codeToLanguageIndex.Add(code, language);
}
if (!string.IsNullOrEmpty(countryName))
{
language.Countries.Add(countryName);
}
return(language);
}
private bool AddLanguage(string code, string threelettercode, string full = null,
string name = null, string localName = null, string region = null, List <string> names = null, string regions = null, List <string> tags = null)
{
string primarycountry;
if (region == null)
{
primarycountry = "";
}
else if (StandardSubtags.IsValidIso3166RegionCode(region))
{
if (StandardSubtags.IsPrivateUseRegionCode(region))
{
if (region == "XK")
{
primarycountry = "Kosovo";
}
else
{
primarycountry = "Unknown private use";
}
}
else
{
primarycountry = StandardSubtags.RegisteredRegions[region].Name; // convert to full region name
}
}
else
{
primarycountry = "Invalid region";
}
LanguageInfo language = new LanguageInfo
{
LanguageTag = code,
ThreeLetterTag = threelettercode,
// DesiredName defaults to Names[0], which is set below.
PrimaryCountry = primarycountry
};
language.Countries.Add(primarycountry);
if (regions != null)
{
string[] countries = regions.Split();
foreach (string country in countries)
{
if (!country.Contains('?') && country != "")
{
language.Countries.Add(StandardSubtags.RegisteredRegions[country].Name);
}
}
}
// For sorting, it is better to store name first instead of localName, which may be in a local script.
// Names[0] is used in several ways in sorting languages in the list of possible matches: 1) bring
// to the top of the list languages where Names[0] matches what the user typed, 2) order by the
// "typing distance" of Names[0] from what the user typed, and 3) order by comparing the Names[0]
// value of the two languages if neither matches the search string and their typing distances from
// the search string are the same.
// Names[1] (if it exists) is used to move the language toward the top of the list if it exactly
// matches the search string. It is not used otherwise in the sorting heuristics. No other
// values in the Names list are involved in the sorting process.
if (name != null)
{
language.Names.Add(name.Trim());
}
if (localName != null && localName != name)
{
language.Names.Add(localName.Trim());
}
if (names != null)
{
foreach (string langname in names)
{
if (!language.Names.Contains(langname))
{
language.Names.Add(langname.Trim());
}
}
}
// If we end up needing to add the language code, that reflects a deficiency in the data. But
// having a bogus name value probably hurts less that not having any name at all. The sort
// process mentioned above using the language tag as well as the first two items in the Names list.
Debug.Assert(language.Names.Count > 0);
if (language.Names.Count == 0)
{
language.Names.Add(code);
}
// add language to _codeToLanguageIndex and _nameToLanguageIndex
// if 2 letter code then add both 2 and 3 letter codes to _codeToLanguageIndex
_codeToLanguageIndex[code] = language;
if (full != null && !string.Equals(full, code))
{
_codeToLanguageIndex[full] = language; // add the full expanded tag
}
if (threelettercode != null && !string.Equals(code, threelettercode))
{
_codeToLanguageIndex[threelettercode] = language;
}
if (tags != null)
{
foreach (string langtag in tags)
{
_codeToLanguageIndex[langtag] = language;
}
}
foreach (string langname in language.Names)
{
GetOrCreateListFromName(langname).Add(language);
}
// add to _countryToLanguageIndex
foreach (var country in language.Countries)
{
if (!string.IsNullOrEmpty(country))
{
List <LanguageInfo> list;
if (!_countryToLanguageIndex.TryGetValue(country, out list))
{
list = new List <LanguageInfo>();
_countryToLanguageIndex[country] = list;
}
list.Add(language);
}
}
return(true);
}
/// <summary>
/// Initializes a new instance of the <see cref="LanguageLookup"/> class.
/// </summary>
public LanguageLookup()
{
var threeToTwoLetter = new Dictionary <string, string>();
foreach (string line in LanguageRegistryResources.TwoToThreeCodes.Replace("\r\n", "\n").Split(new[] { "\n" }, StringSplitOptions.RemoveEmptyEntries))
{
string[] items = line.Split('\t');
threeToTwoLetter.Add(items[1].Trim(), items[0].Trim());
}
//LanguageIndex.txt Format: LangID CountryID NameType Name
//a language appears on one row for each of its alternative langauges
var entries = new List <string>(LanguageRegistryResources.LanguageIndex.Split(new[] { "\n" }, StringSplitOptions.RemoveEmptyEntries));
entries.Add("qaa\t?\tL\tUnlisted Language");
foreach (string entry in entries.Skip(1)) //skip the header
{
string[] items = entry.Split('\t');
if (items.Length != 4)
{
continue;
}
if (items[2].Contains('!')) //temporary suppression of entries while waiting for Ethnologue changes
{
continue;
}
string code = items[0].Trim();
string twoLetterCode;
if (threeToTwoLetter.TryGetValue(code, out twoLetterCode))
{
code = twoLetterCode;
}
string regionCode = items[1].Trim();
LanguageInfo language = GetOrCreateLanguageFromCode(code, regionCode == "?" ? "?" : StandardSubtags.RegisteredRegions[regionCode].Name);
string name = items[3].Trim();
if (items[2] == "L")
{
while (language.Names.Contains(name))
{
language.Names.Remove(name);
}
language.Names.Insert(0, name);
}
else
{
if (items[2].Contains("P"))
{
//Skip pejorative
}
else if (items[1] == ("ET"))
{
//Skip alternatives for Ethiopia, as per request
}
else if (items[0] == "gax" || items[0] == "om")
{
//For these two "Oromo" languages, skip all related languages as per request
}
else if (!language.Names.Contains(name))
{
language.Names.Add(name); //intentionally not lower-casing
}
}
}
IEnumerable <IGrouping <string, string> > languageGroups = Sldr.LanguageTags.Where(info => info.IsAvailable && IetfLanguageTag.IsValid(info.LanguageTag))
.Select(info => IetfLanguageTag.Canonicalize(info.LanguageTag))
.GroupBy(IetfLanguageTag.GetLanguagePart);
foreach (IGrouping <string, string> languageGroup in languageGroups)
{
string[] langTags = languageGroup.ToArray();
if (langTags.Length == 1)
{
string langTag = langTags[0];
LanguageInfo language;
if (langTag != languageGroup.Key && _codeToLanguageIndex.TryGetValue(languageGroup.Key, out language))
{
_codeToLanguageIndex.Remove(languageGroup.Key);
language.LanguageTag = langTag;
_codeToLanguageIndex[langTag] = language;
}
}
else
{
foreach (string langTag in langTags)
{
LanguageSubtag languageSubtag;
ScriptSubtag scriptSubtag;
RegionSubtag regionSubtag;
IEnumerable <VariantSubtag> variantSubtags;
if (IetfLanguageTag.TryGetSubtags(langTag, out languageSubtag, out scriptSubtag, out regionSubtag, out variantSubtags))
{
if (langTag == languageSubtag)
{
continue;
}
LanguageInfo language = GetOrCreateLanguageFromCode(langTag, regionSubtag == null ? "?" : regionSubtag.Name);
bool displayScript = scriptSubtag != null && !IetfLanguageTag.IsScriptImplied(langTag);
LanguageInfo otherLanguage;
if (langTag != languageSubtag && !displayScript && _codeToLanguageIndex.TryGetValue(languageSubtag, out otherLanguage) && language.Countries.SetEquals(otherLanguage.Countries))
{
language.Names.AddRange(otherLanguage.Names);
}
else
{
string name = displayScript ? string.Format("{0} ({1})", languageSubtag.Name, scriptSubtag.Name) : languageSubtag.Name;
if (!language.Names.Contains(name))
{
language.Names.Add(name); //intentionally not lower-casing
}
}
}
}
}
}
foreach (LanguageInfo languageInfo in _codeToLanguageIndex.Values)
{
foreach (string name in languageInfo.Names)
{
GetOrCreateListFromName(name).Add(languageInfo);
}
if (languageInfo.Names.Count == 0)
{
continue; // this language is suppressed
}
//Why just this small set? Only out of convenience. Ideally we'd have a db of all languages as they write it in their literature.
string localName = null;
switch (languageInfo.Names[0])
{
case "French":
localName = "français";
break;
case "Spanish":
localName = "español";
break;
case "Chinese":
localName = "中文";
break;
case "Hindi":
localName = "हिन्दी";
break;
case "Bengali":
localName = "বাংলা";
break;
case "Telugu":
localName = "తెలుగు";
break;
case "Tamil":
localName = "தமிழ்";
break;
case "Urdu":
localName = "اُردُو";
break;
case "Arabic":
localName = "العربية/عربي";
break;
case "Thai":
localName = "ภาษาไทย";
break;
case "Indonesian":
localName = "Bahasa Indonesia";
break;
}
if (!string.IsNullOrEmpty(localName))
{
if (!languageInfo.Names.Remove(localName))
{
GetOrCreateListFromName(localName).Add(languageInfo);
}
languageInfo.Names.Insert(0, localName);
}
}
}
private LanguageInfo ChangeLanguage(string iso639Code, string potentiallyCustomName=null)
{
using (var dlg = new LanguageLookupDialog())
{
//at this point, we don't let them customize the national languages
dlg.IsDesiredLanguageNameFieldVisible = potentiallyCustomName != null;
var language = new LanguageInfo() { LanguageTag = iso639Code};
if (!string.IsNullOrEmpty(potentiallyCustomName))
{
language.DesiredName = potentiallyCustomName; // to be noticed, must set before dlg.SelectedLanguage
}
dlg.SelectedLanguage = language;
dlg.SearchText = iso639Code;
if (DialogResult.OK != dlg.ShowDialog())
{
return null;
}
return dlg.SelectedLanguage;
}
}
/// <summary>
/// Initializes a new instance of the <see cref="LanguageLookup"/> class.
/// It gets its data from the LanguageDataIndex resource
/// </summary>
public LanguageLookup()
{
// Load from file into the data structures instead of creating it from scratch
var entries = LanguageRegistryResources.LanguageDataIndex.Replace("\r\n", "\n").Split(new[] { "\n" }, StringSplitOptions.RemoveEmptyEntries);
foreach (string entry in entries)
{
// Fields in LanguageDataIndex
// Code ThreeLetterCode DesiredName Names Countries PrimaryCountry
string[] items = entry.Split('\t');
if (items.Length != 7) // This needs to be changed if the number of fields changes
{
continue;
}
string code = items[0];
string threelettercode = items[1];
string desiredname = items[2];
bool macrolanguage = String.Equals("M", items[3]);
string[] names = items[4].Split(';');
string[] countries = items[5].Split(';');
string primarycountry = items[6];
LanguageInfo language = new LanguageInfo {
LanguageTag = code, ThreeLetterTag = threelettercode, DesiredName = desiredname,
IsMacroLanguage = macrolanguage, PrimaryCountry = primarycountry
};
foreach (string country in countries)
{
language.Countries.Add(country);
}
foreach (string langname in names)
{
language.Names.Add(langname.Trim());
}
// Do not add anything to LanguageInfo manually here if it would be useful in LanguageDataIndex.txt/json
// add language to _codeToLanguageIndex and _nameToLanguageIndex
// if 2 letter code then add both 2 and 3 letter codes to _codeToLanguageIndex
_codeToLanguageIndex[code] = language;
if (!String.Equals(code, threelettercode))
{
_codeToLanguageIndex[threelettercode] = language;
}
foreach (string langname in language.Names)
{
GetOrCreateListFromName(langname).Add(language);
}
// add to _countryToLanguageIndex
foreach (var country in language.Countries)
{
if (!string.IsNullOrEmpty(country))
{
List <LanguageInfo> list;
if (!_countryToLanguageIndex.TryGetValue(country, out list))
{
list = new List <LanguageInfo>();
_countryToLanguageIndex[country] = list;
}
list.Add(language);
}
}
}
}