/// <summary>
/// A new locale identifier with the all the empty subtags filled in
/// with a likely value.
/// </summary>
/// <returns>
/// A new locale identifier with all subtags filled in.
/// </returns>
public LocaleIdentifier MostLikelySubtags()
{
var result = (LocaleIdentifier)this.MemberwiseClone();
// Remove the script code 'Zzzz' and the region code 'ZZ' if they occur.
if (result.Script == "Zzzz")
{
result.Script = String.Empty;
}
if (result.Region == "ZZ")
{
result.Region = String.Empty;
}
// Short cut if all subtags have a value.
if (result.Language != "" && result.Script != "" && result.Region != "")
{
return(result);
}
// Find the language in likely subtags.
var likely = new[]
{
$"{result.Language}_{result.Script}_{result.Region}",
$"{result.Language}_{result.Region}",
$"{result.Language}_{result.Script}",
$"{result.Language}",
$"und_{result.Script}",
}
.Select(k => k.Replace("__", "_").Trim('_'))
.Where(k => k != String.Empty)
.Distinct()
.Select(k => Cldr.Instance
.GetDocuments("common/supplemental/likelySubtags.xml")
.FirstElementOrDefault($"supplementalData/likelySubtags/likelySubtag[@from='{k}']")
)
.FirstOrDefault(e => e != null);
if (likely != null)
{
var defaults = LocaleIdentifier.ParseBcp47(likely.GetAttribute("to", ""));
if (result.Language == "")
{
result.Language = defaults.Language;
}
if (result.Script == "")
{
result.Script = defaults.Script;
}
if (result.Region == "")
{
result.Region = defaults.Region;
}
}
return(result);
}
/// <summary>
/// A new locale idenyifer with empty subtags that <see cref="MostLikelySubtags"/> would fill.
/// </summary>
/// <returns>
/// A new locale identifier.
/// </returns>
public LocaleIdentifier RemoveMostLikelySubtags()
{
var max = this.MostLikelySubtags();
max.Variants = new string[0];
var trials = new[] { max.Language, $"{max.Language}-{max.Region}", $"{max.Language}-{max.Script}" };
foreach (var trial in trials)
{
var id = LocaleIdentifier.ParseBcp47(trial);
if (max.ToUnicodeLanguage() == id.MostLikelySubtags().ToUnicodeLanguage())
{
id.Variants = this.Variants;
return(id);
}
}
max.Variants = this.Variants;
return(max);
}
/// <summary>
/// Convert BCP 47 tag to a valid Unicode locale identifier
/// </summary>
string TransformFromBcp47()
{
// 1. Canonicalize the language tag (afterwards, there will be no extlang subtag)
// 2. Replace the BCP 47 primary language subtag "und" with "root" if no script, region,
// or variant subtags are present
if (Language == "und" && Script == "" && Region == "" && Variants.Count() == 0)
{
Language = "root";
}
// 3. If the BCP 47 primary language subtag matches the type attribute of a languageAlias
// element in Supplemental Data, replace the language subtag with the replacement value.
// 3.1 If there are additional subtags in the replacement value, add them to the result, but only
// if there is no corresponding subtag already in the tag.
var languageAlias = Cldr.Instance
.GetDocuments("common/supplemental/supplementalMetadata.xml")
.FirstElementOrDefault($"supplementalData/metadata/alias/languageAlias[@type='{Language}']");
if (languageAlias != null)
{
var replacement = LocaleIdentifier.ParseBcp47(languageAlias.GetAttribute("replacement", ""));
Language = replacement.Language;
if (Script == "")
{
Script = replacement.Script;
}
if (Region == "")
{
Region = replacement.Region;
}
}
// 4. If the BCP 47 region subtag matches the type attribute of a territoryAlias
// element in Supplemental Data, replace the language subtag with the replacement value, as follows:
var territoryAlias = Cldr.Instance
.GetDocuments("common/supplemental/supplementalMetadata.xml")
.FirstElementOrDefault($"supplementalData/metadata/alias/territoryAlias[@type='{Region}']");
if (territoryAlias != null)
{
var replacements = territoryAlias.GetAttribute("replacement", "").Split(' ');
// 4.1 If there is a single territory in the replacement, use it.
var replacementValue = replacements[0];
// 4.2 If there are multiple territories:
// 4.2.1 Look up the most likely territory for the base language code(and script, if there is one).
// 4.2.2 If that likely territory is in the list, use it.
// 4.2.3 Otherwise, use the first territory in the list.
if (replacements.Length > 1)
{
var best = Cldr.Instance
.GetDocuments("common/supplemental/likelySubtags.xml")
.FirstElementOrDefault($"supplementalData/likelySubtags/likelySubtag[@from='{Language}']");
if (best != null)
{
var to = LocaleIdentifier.ParseBcp47(best.GetAttribute("to", ""));
if (replacements.Contains(to.Region))
{
replacementValue = to.Region;
}
}
}
Region = replacementValue;
}
// Verify that the subtags are defined.
// TODO: Need StringRanges, see https://github.com/richardschneider/net-cldr/issues/2
if (Language != "" && !Cldr.Instance.IsLanguageDefined(Language))
{
return($"Language '{Language}' is not defined.");
}
if (Script != "" && !Cldr.Instance.IsScriptDefined(Script))
{
return($"Script '{Script}' is not defined.");
}
if (Region != "" && !Cldr.Instance.IsRegionDefined(Region))
{
return($"Region '{Region}' is not defined.");
}
foreach (var variant in Variants)
{
if (!Cldr.Instance.IsVariantDefined(variant))
{
return($"Language variant '{variant}' is not defined.");
}
}
// TODO: U extensions
// TODO: T extensions
return(null);
}