You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
jellyfin/MediaBrowser.Controller/Extensions/StringExtensions.cs

53 lines
1.6 KiB

#pragma warning disable CS1591
using System;
using System.Globalization;
using System.Linq;
using System.Text;
using System.Text.RegularExpressions;
namespace MediaBrowser.Controller.Extensions
{
/// <summary>
/// Class BaseExtensions.
/// </summary>
public static class StringExtensions
{
public static string RemoveDiacritics(this string text)
{
var chars = Normalize(text, NormalizationForm.FormD)
.Where(ch => CharUnicodeInfo.GetUnicodeCategory(ch) != UnicodeCategory.NonSpacingMark);
return Normalize(string.Concat(chars), NormalizationForm.FormC);
}
private static string Normalize(string text, NormalizationForm form, bool stripStringOnFailure = true)
{
if (stripStringOnFailure)
{
try
{
return text.Normalize(form);
}
catch (ArgumentException)
{
// will throw if input contains invalid unicode chars
// https://mnaoumov.wordpress.com/2014/06/14/stripping-invalid-characters-from-utf-16-strings/
text = Regex.Replace(text, "([\ud800-\udbff](?![\udc00-\udfff]))|((?<![\ud800-\udbff])[\udc00-\udfff])", string.Empty);
return Normalize(text, form, false);
}
}
try
{
return text.Normalize(form);
}
catch (ArgumentException)
{
// if it still fails, return the original text
return text;
}
}
}
}