public unsafe static CharacterDecompositionMapping Parse(string s)
{
if (string.IsNullOrEmpty(s))
{
return(default(CharacterDecompositionMapping));
}
CompatibilityFormattingTag tag = CompatibilityFormattingTag.Canonical;
int index;
if (s[0] == '<')
{
if (!EnumHelper <CompatibilityFormattingTag> .TryGetNamedValue(s.Substring(1, (index = s.IndexOf('>')) - 1), out tag))
{
throw new FormatException();
}
++index;
}
else
{
index = 0;
}
var buffer = stackalloc char[36]; // From the Unicode docs, a decomposition cannot have more than 18 code points.
int charIndex = 0;
while (index < s.Length && charIndex < 35)
{
char c = s[index];
if (c == ' ')
{
++index;
}
else
{
int codePoint = HexCodePoint.Parse(s, ref index);
if (codePoint < 0x10000)
{
buffer[charIndex++] = (char)codePoint;
}
else if (codePoint < 0x10FFFF)
{
codePoint -= 0x10000;
buffer[charIndex++] = (char)((codePoint >> 10) + 0xD800);
buffer[charIndex++] = (char)((codePoint & 0x3FF) + 0xDC00);
}
else
{
throw new FormatException("The code point was outside of the allowed range.");
}
}
}
return(new CharacterDecompositionMapping(tag, new string(buffer, 0, charIndex)));
}
public readonly int[] CrossRerefences; // NB: It seems that parsing NamesList is required in order to provide data for this field ?
internal UnicodeCharacterData
(
UnicodeCodePointRange codePointRange,
string name,
UnicodeNameAlias[] nameAliases,
UnicodeCategory category,
CanonicalCombiningClass canonicalCombiningClass,
BidirectionalClass bidirectionalClass,
CompatibilityFormattingTag decompositionType,
string decompositionMapping,
UnicodeNumericType numericType,
UnicodeRationalNumber numericValue,
bool bidirectionalMirrored,
string oldName,
string simpleUpperCaseMapping,
string simpleLowerCaseMapping,
string simpleTitleCaseMapping,
ContributoryProperties contributoryProperties,
int corePropertiesAndEmojiProperties,
int[] crossRerefences
)
{
this.CodePointRange = codePointRange;
this.Name = name;
this.NameAliases = nameAliases;
this.Category = category;
this.CanonicalCombiningClass = canonicalCombiningClass;
this.BidirectionalClass = bidirectionalClass;
this.DecompositionType = decompositionType;
this.DecompositionMapping = decompositionMapping;
this.NumericType = numericType;
this.numericValue = numericValue;
this.BidirectionalMirrored = bidirectionalMirrored;
this.OldName = oldName;
this.SimpleUpperCaseMapping = simpleUpperCaseMapping;
this.SimpleLowerCaseMapping = simpleLowerCaseMapping;
this.SimpleTitleCaseMapping = simpleTitleCaseMapping;
this.ContributoryProperties = contributoryProperties;
this.corePropertiesAndEmojiProperties = corePropertiesAndEmojiProperties;
this.CrossRerefences = crossRerefences;
}
private static void ReadUnicodeCharacterDataEntry(BinaryReader reader, byte[] nameBuffer, out UnicodeCharacterData value)
{
var fields = (UcdFields)reader.ReadUInt16();
var codePointRange = (fields & UcdFields.CodePointRange) != 0 ? new UnicodeCodePointRange(ReadCodePoint(reader), ReadCodePoint(reader)) : new UnicodeCodePointRange(ReadCodePoint(reader));
string name = null;
UnicodeNameAlias[] nameAliases = UnicodeNameAlias.EmptyArray;
// Read all the official names of the character.
if ((fields & UcdFields.Name) != 0)
{
int length = reader.ReadByte();
byte @case = (byte)(length & 0xC0);
if (@case < 0x80) // Handles the case where only the name is present.
{
length = (length & 0x7F) + 1;
if (reader.Read(nameBuffer, 0, length) != length)
{
throw new EndOfStreamException();
}
name = Encoding.UTF8.GetString(nameBuffer, 0, length);
}
else
{
nameAliases = new UnicodeNameAlias[(length & 0x3F) + 1];
if ((@case & 0x40) != 0)
{
length = reader.ReadByte() + 1;
if (length > 128)
{
throw new InvalidDataException("Did not expect names longer than 128 bytes.");
}
if (reader.Read(nameBuffer, 0, length) != length)
{
throw new EndOfStreamException();
}
name = Encoding.UTF8.GetString(nameBuffer, 0, length);
}
for (int i = 0; i < nameAliases.Length; ++i)
{
nameAliases[i] = new UnicodeNameAlias(reader.ReadString(), (UnicodeNameAliasKind)(reader.ReadByte()));
}
}
}
var category = (fields & UcdFields.Category) != 0 ? (UnicodeCategory)reader.ReadByte() : UnicodeCategory.OtherNotAssigned;
var canonicalCombiningClass = (fields & UcdFields.CanonicalCombiningClass) != 0 ? (CanonicalCombiningClass)reader.ReadByte() : CanonicalCombiningClass.NotReordered;
var bidirectionalClass = (fields & UcdFields.BidirectionalClass) != 0 ? (BidirectionalClass)reader.ReadByte() : 0;
CompatibilityFormattingTag decompositionType = (fields & UcdFields.DecompositionMapping) != 0 ? (CompatibilityFormattingTag)reader.ReadByte() : CompatibilityFormattingTag.Canonical;
string decompositionMapping = (fields & UcdFields.DecompositionMapping) != 0 ? reader.ReadString() : null;
var numericType = (UnicodeNumericType)((int)(fields & UcdFields.NumericNumeric) >> 6);
UnicodeRationalNumber numericValue = numericType != UnicodeNumericType.None ?
new UnicodeRationalNumber(reader.ReadInt64(), reader.ReadByte()) :
default(UnicodeRationalNumber);
string oldName = (fields & UcdFields.OldName) != 0 ? reader.ReadString() : null;
string simpleUpperCaseMapping = (fields & UcdFields.SimpleUpperCaseMapping) != 0 ? reader.ReadString() : null;
string simpleLowerCaseMapping = (fields & UcdFields.SimpleLowerCaseMapping) != 0 ? reader.ReadString() : null;
string simpleTitleCaseMapping = (fields & UcdFields.SimpleTitleCaseMapping) != 0 ? reader.ReadString() : null;
ContributoryProperties contributoryProperties = (fields & UcdFields.ContributoryProperties) != 0 ? (ContributoryProperties)reader.ReadInt32() : 0;
int corePropertiesAndEmojiProperties = (fields & UcdFields.CorePropertiesAndEmojiProperties) != 0 ? ReadInt24(reader) : 0;
int[] crossReferences = (fields & UcdFields.CrossRerefences) != 0 ? new int[reader.ReadByte() + 1] : null;
if (crossReferences != null)
{
for (int i = 0; i < crossReferences.Length; ++i)
{
crossReferences[i] = ReadCodePoint(reader);
}
}
value = new UnicodeCharacterData
(
codePointRange,
name,
nameAliases,
category,
canonicalCombiningClass,
bidirectionalClass,
decompositionType,
decompositionMapping,
numericType,
numericValue,
(fields & UcdFields.BidirectionalMirrored) != 0,
oldName,
simpleUpperCaseMapping,
simpleLowerCaseMapping,
simpleTitleCaseMapping,
contributoryProperties,
corePropertiesAndEmojiProperties,
crossReferences
);
}
public CharacterDecompositionMapping(CompatibilityFormattingTag decompositionType, string decompositionMapping)
{
DecompositionType = decompositionType;
DecompositionMapping = decompositionMapping;
}
