private static bool IsPrintableCharacter(char c)
{
// See https://stackoverflow.com/questions/3253247/how-do-i-detect-non-printable-characters-in-net
if (c == 0x2028 || c == 0x2029)
{
return(false);
}
// To ensure that we don't get new lines and whatnot.
if (char.IsWhiteSpace(c))
{
return(false);
}
// See https://docs.microsoft.com/en-us/dotnet/api/system.globalization.unicodecategory?view=netframework-4.8
var nonRenderingCategories = new UnicodeCategory[] {
UnicodeCategory.Control,
UnicodeCategory.OtherNotAssigned,
UnicodeCategory.Surrogate,
UnicodeCategory.LineSeparator,
UnicodeCategory.ParagraphSeparator
};
return(!nonRenderingCategories.Contains(Char.GetUnicodeCategory(c)));
}
public static string RemoveNonAlphanumericCharacters(this string s)
{
var alphanumericCategories = new UnicodeCategory[]
{
UnicodeCategory.LowercaseLetter,
UnicodeCategory.UppercaseLetter,
UnicodeCategory.DecimalDigitNumber,
};
return(new string(s.Where(c => alphanumericCategories.Contains(CharUnicodeInfo.GetUnicodeCategory(c))).ToArray()));
}
public static string clean(string name, HashSet <char> invalid)
{
if (string.IsNullOrEmpty(name))
{
return(name);
}
if (invalid.Count <= 0)
{
invalid = nremove;
}
// https://stackoverflow.com/questions/146134/how-to-remove-illegal-characters-from-path-and-filenames
// https://stackoverflow.com/questions/3253247/how-do-i-detect-non-printable-characters-in-net
// The set of Unicode character categories containing non-rendering,
// unknown, or incomplete characters.
// !! Unicode.Format and Unicode.PrivateUse can NOT be included in
// !! this set, because they may (private-use) or do (format)
// !! contain at least *some* rendering characters.
var nonRenderingCategories = new UnicodeCategory[]
{
UnicodeCategory.Control,
UnicodeCategory.OtherNotAssigned,
UnicodeCategory.Surrogate
};
var sb = new StringBuilder();
foreach (var c in name)
{
// Char.IsWhiteSpace() includes the ASCII whitespace characters that
// are categorized as control characters. Any other character is
// printable, unless it falls into the non-rendering categories.
var isPrintable = Char.IsWhiteSpace(c) || !nonRenderingCategories.Contains(Char.GetUnicodeCategory(c));
if (isPrintable && invalid.Count > 0)
{
isPrintable = (invalid.Contains(c) == false);
}
// ASCII/Unicode characters 1 through 31
if (c < 32)
{
isPrintable = false;
}
if (isPrintable)
{
sb.Append(c);
}
}
return(sb.ToString());
}
static public bool CharIsPrintable(char character)
{
// The set of Unicode character categories containing non-rendering,
// unknown, or incomplete characters.
// !! Unicode.Format and Unicode.PrivateUse can NOT be included in this set,
var nonRenderingCategories = new UnicodeCategory[] {
UnicodeCategory.Control,
UnicodeCategory.OtherNotAssigned,
UnicodeCategory.Surrogate
};
// Char.IsWhiteSpace() includes the ASCII whitespace characters that
// are categorized as control characters. Any other character is
// printable, unless it falls into the non-rendering categories.
bool isPrintable = Char.IsWhiteSpace(character) || !nonRenderingCategories.Contains(Char.GetUnicodeCategory(character));
return(isPrintable);
}
/// <summary>
/// 剔除乱码
/// </summary>
/// <param name="content"></param>
/// <returns></returns>
public static string TrimMessyChars(this string content)
{
string result = null;
if (!string.IsNullOrWhiteSpace(content))
{
content = content.Trim();
var arr = new UnicodeCategory[] { UnicodeCategory.OtherSymbol, UnicodeCategory.Surrogate, UnicodeCategory.Control };
foreach (var c in content)
{
if (!arr.Contains(CharUnicodeInfo.GetUnicodeCategory(c)))
{
result += c;
}
}
}
return(result);
}
public static char[] InvisibleChars()
{
List <char> e = new List <char>();
UnicodeCategory[] uc = new UnicodeCategory[]
{
UnicodeCategory.Format,
};
char b = '\0';
for (int i = 1; i < 0xFFFF; i++)
{
b = (char)i;
if (uc.Contains(CharUnicodeInfo.GetUnicodeCategory(b)))
{
e.Add(b);
}
}
return(e.ToArray());
}
/// <summary>
/// Will determine if a string contains printable char or not
/// (extension)
/// </summary>
/// <param name="text"></param>
/// <returns>true if string contains printable char and false otherwise</returns>
public static bool IsPrintable(this string text)
{
var nonRenderingCategories = new UnicodeCategory[] {
UnicodeCategory.Control,
UnicodeCategory.OtherNotAssigned,
UnicodeCategory.Surrogate
};
bool pritable = true;
foreach (char c in text)
{
var isPrintable = char.IsWhiteSpace(c) || !nonRenderingCategories.Contains(char.GetUnicodeCategory(c));
if (!isPrintable)
{
return(false);
}
}
return(pritable);
}