tsai
/
Avalonia
mirror of https://github.com/AvaloniaUI/Avalonia.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
Browse Source

Reduce Allocations in Text Rendering (#20175) 

* Reduce allocations for FontFamily.Parse

* Turn Trie data into static data

* Use static lambda in s_cachedLanguage.GetOrAdd

* Make LineBreakEnumerator+LineBreakState a ref struct to avoid allocations

* Use backing field for storing trie data in debug builds

.NET Standard compatibility
release/11.3.10
Washi 2 months ago
committed by Julien Lebosquain
parent
b4e391ee92
commit
eb59ffc0c1
No known key found for this signature in database GPG Key ID: 1833CAD10ACC46FD
9 changed files with 137 additions and 104 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 72
      src/Avalonia.Base/Media/FontFamily.cs
  2. 16
      src/Avalonia.Base/Media/TextFormatting/Unicode/BiDi.trie.cs
  3. 2
      src/Avalonia.Base/Media/TextFormatting/Unicode/EastAsianWidth.trie.cs
  4. 2
      src/Avalonia.Base/Media/TextFormatting/Unicode/GraphemeBreak.trie.cs
  5. 98
      src/Avalonia.Base/Media/TextFormatting/Unicode/LineBreakEnumerator.cs
  6. 18
      src/Avalonia.Base/Media/TextFormatting/Unicode/UnicodeData.trie.cs
  7. 4
      src/Markup/Avalonia.Markup/Markup/Parsers/ContainerQueryGrammar.cs
  8. 9
      src/Skia/Avalonia.Skia/TextShaperImpl.cs
  9. 20
      tests/Avalonia.Base.UnitTests/Media/TextFormatting/UnicodeDataGenerator.cs

72
src/Avalonia.Base/Media/FontFamily.cs
View File

@ -131,53 +131,55 @@ namespace Avalonia.Media
{
var result = new FrugalStructList<FontSourceIdentifier>(1);
var segments = name.Split(',');
for (int i = 0; i < segments.Length; i++)
int commaIndex = -1;
do
{
var segment = segments[i];
var innerSegments = segment.Split('#');
// Look for a comma separator to carve out a single segment.
int segmentStart = commaIndex + 1;
commaIndex = name.IndexOf(',', segmentStart);
int segmentEnd = commaIndex == -1
? name.Length
: commaIndex;
var segment = name.AsSpan(segmentStart, segmentEnd - segmentStart).Trim();
FontSourceIdentifier identifier = new FontSourceIdentifier(name, null);
FontSourceIdentifier? identifier = null;
switch (innerSegments.Length)
// Check if there is exactly one '#' (i.e., segment is in the format "path#innerName").
int separatorIndex = segment.IndexOf('#');
if (separatorIndex != -1 && segment.Slice(separatorIndex + 1).IndexOf('#') == -1)
{
case 1:
var pathSpan = segment.Slice(0, separatorIndex).Trim();
var innerName = segment.Slice(separatorIndex + 1).Trim();
if (pathSpan.IsEmpty)
{
identifier = new FontSourceIdentifier(innerName.ToString(), null);
}
else
{
string path = pathSpan.ToString();
if (pathSpan.IndexOf('/') >= 0 && Uri.TryCreate(path, UriKind.Relative, out var source))
{
identifier = new FontSourceIdentifier(innerSegments[0].Trim(), null);
break;
identifier = new FontSourceIdentifier(innerName.ToString(), source);
}
case 2:
else
{
var path = innerSegments[0].Trim();
var innerName = innerSegments[1].Trim();
if (string.IsNullOrEmpty(path))
{
identifier = new FontSourceIdentifier(innerName, null);
}
else
if (Uri.TryCreate(path, UriKind.Absolute, out source))
{
if (path.Contains('/') && Uri.TryCreate(path, UriKind.Relative, out var source))
{
identifier = new FontSourceIdentifier(innerName, source);
}
else
{
if (Uri.TryCreate(path, UriKind.Absolute, out source))
{
identifier = new FontSourceIdentifier(innerName, source);
}
}
identifier = new FontSourceIdentifier(innerName.ToString(), source);
}
break;
}
}
}
result.Add(identifier);
}
// If we didn't manage to match it to any known format, treat the entire segment as the font name.
identifier ??= new FontSourceIdentifier(
segment.Length == name.Length ? name : segment.ToString(),
null);
result.Add(identifier.Value);
} while (commaIndex != -1);
return result;
}

16
src/Avalonia.Base/Media/TextFormatting/Unicode/BiDi.trie.cs
View File

@ -16,10 +16,18 @@ internal static class BidiTrie
{
[MethodImpl(MethodImplOptions.AggressiveInlining)]
get => new(Data, 0x00100000, 0x00000000);
}
}
private static ReadOnlySpan<uint> Data => new uint[]
{
// For Debug builds, we store the data in a separate uint[] field to avoid RuntimeFieldInfoStub allocations.
// See also: https://github.com/AvaloniaUI/Avalonia/pull/20175
#if DEBUG
public static ReadOnlySpan<uint> Data => s_data;
private static uint[] s_data =
#else
public static ReadOnlySpan<uint> Data =>
#endif
[
0x0000038D, 0x00000395, 0x0000039D, 0x000003A5, 0x000003BD, 0x000003C5, 0x000003CD, 0x000003D5, 0x000003AD, 0x000003B5, 0x000003AD, 0x000003B5,
0x000003AD, 0x000003B5, 0x000003AD, 0x000003B5, 0x000003AD, 0x000003B5, 0x000003AD, 0x000003B5, 0x000003DB, 0x000003E3, 0x000003EB, 0x000003F3,
0x000003FB, 0x00000403, 0x000003FF, 0x00000407, 0x0000040F, 0x00000417, 0x00000412, 0x0000041A, 0x000003AD, 0x000003B5, 0x000003AD, 0x000003B5,
@ -1104,5 +1112,5 @@ internal static class BidiTrie
0x00000000, 0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00000000, 0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00040000,
0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00040000, 0x00040000,
0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000
};
];
}

2
src/Avalonia.Base/Media/TextFormatting/Unicode/EastAsianWidth.trie.cs
View File

@ -18,7 +18,7 @@ internal static class EastAsianWidthTrie
get => new(Data, 0x00110000, 0x00000000);
}
private static ReadOnlySpan<uint> Data => new uint[]
private static readonly uint[] Data = new uint[]
{
0x000003E7, 0x000003EF, 0x000003F7, 0x000003FF, 0x0000041F, 0x00000427, 0x0000042F, 0x00000437, 0x0000043F, 0x00000447, 0x0000044F, 0x00000457,
0x00000417, 0x0000041F, 0x0000045C, 0x00000464, 0x00000417, 0x0000041F, 0x00000468, 0x00000470, 0x00000417, 0x0000041F, 0x00000477, 0x0000047F,

2
src/Avalonia.Base/Media/TextFormatting/Unicode/GraphemeBreak.trie.cs
View File

@ -18,7 +18,7 @@ internal static class GraphemeBreakTrie
get => new(Data, 0x000E1000, 0x00000000);
}
private static ReadOnlySpan<uint> Data => new uint[]
private static readonly uint[] Data = new uint[]
{
0x0000035A, 0x00000362, 0x0000036A, 0x00000372, 0x0000038A, 0x00000392, 0x00000362, 0x0000036A, 0x00000362, 0x0000036A, 0x00000362, 0x0000036A,
0x00000362, 0x0000036A, 0x00000362, 0x0000036A, 0x00000362, 0x0000036A, 0x00000362, 0x0000036A, 0x00000362, 0x0000036A, 0x00000362, 0x0000036A,

98
src/Avalonia.Base/Media/TextFormatting/Unicode/LineBreakEnumerator.cs
View File

@ -12,7 +12,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
private static readonly BreakUnit s_eot = new() { EndOfText = true };
public readonly ReadOnlySpan<char> _text;
private readonly LineBreakState _state;
private LineBreakState _state;
public LineBreakEnumerator(ReadOnlySpan<char> text)
{
@ -20,7 +20,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
_state = new LineBreakState();
}
public readonly bool MoveNext([NotNullWhen(true)] out LineBreak lineBreak)
public bool MoveNext([NotNullWhen(true)] out LineBreak lineBreak)
{
lineBreak = default;
@ -35,7 +35,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
{
_state.Read(_text);
result = ExecuteRules(_text, _state);
result = ExecuteRules(_text, ref _state);
}
if (result == null)
@ -120,11 +120,11 @@ namespace Avalonia.Media.TextFormatting.Unicode
return from;
}
private static LineBreak? ExecuteRules(ReadOnlySpan<char> text, LineBreakState state)
private static LineBreak? ExecuteRules(ReadOnlySpan<char> text, ref LineBreakState state)
{
foreach (var rule in s_rules)
{
var res = rule.Invoke(text, state);
var res = rule.Invoke(text, ref state);
switch (res)
{
@ -143,7 +143,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
return null;
}
private static RuleResult QuotationAndRegionalIndicator(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult QuotationAndRegionalIndicator(ReadOnlySpan<char> text, ref LineBreakState state)
{
if (state.Current.Inherited)
{
@ -176,7 +176,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB3: Always break at the end of text.
/// </summary>
private static RuleResult LB03(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB03(ReadOnlySpan<char> text, ref LineBreakState state)
{
if (state.Current.EndOfText)
{
@ -189,7 +189,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB4: Always break after hard line breaks.
/// </summary>
private static RuleResult LB04(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB04(ReadOnlySpan<char> text, ref LineBreakState state)
{
// BK !
if (state.Current.LineBreakClass == LineBreakClass.MandatoryBreak)
@ -203,7 +203,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB5: Treat CR followed by LF, as well as CR, LF, and NL as hard line
/// </summary>
private static RuleResult LB05(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB05(ReadOnlySpan<char> text, ref LineBreakState state)
{
switch (state.Current.LineBreakClass)
{
@ -227,7 +227,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// LB6: Do not break before hard line breaks.
/// </summary>
/// <returns></returns>
private static RuleResult LB06(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB06(ReadOnlySpan<char> text, ref LineBreakState state)
{
// × ( BK | CR | LF | NL )
if (IsBreakClass(state.Next(text).LineBreakClass))
@ -241,7 +241,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB7: Do not break before spaces or zero width space.
/// </summary>
private static RuleResult LB07(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB07(ReadOnlySpan<char> text, ref LineBreakState state)
{
// × SP
// × ZW
@ -261,7 +261,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB8: Break before any character following a zero-width space, even if one or more spaces intervene.
/// </summary>
private static RuleResult LB08(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB08(ReadOnlySpan<char> text, ref LineBreakState state)
{
if (state.LastBeforeSpace.LineBreakClass == LineBreakClass.ZWSpace && state.Next(text).LineBreakClass != LineBreakClass.Space)
{
@ -274,7 +274,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB8a: Do not break after a zero width joiner.
/// </summary>
private static RuleResult LB08a(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB08a(ReadOnlySpan<char> text, ref LineBreakState state)
{
// ZWJ ×
if (state.Current.LineBreakClass == LineBreakClass.ZWJ)
@ -290,7 +290,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// treat it as if it has the line breaking class of the base character in all of the following rules.
/// Treat ZWJ as if it were CM.
/// </summary>
private static RuleResult LB09(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB09(ReadOnlySpan<char> text, ref LineBreakState state)
{
// Treat X (CM | ZWJ)* as if it were X.
// where X is any line break class except BK, CR, LF, NL, SP, or ZW.
@ -319,7 +319,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB10: Treat any remaining combining mark or ZWJ as AL.
/// </summary>
private static RuleResult LB10(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB10(ReadOnlySpan<char> text, ref LineBreakState state)
{
if (state.Current.LineBreakClass == LineBreakClass.CombiningMark)
{
@ -338,7 +338,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB11: Do not break before or after Word joiner and related characters.
/// </summary>
private static RuleResult LB11(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB11(ReadOnlySpan<char> text, ref LineBreakState state)
{
if (state.Next(text).LineBreakClass == LineBreakClass.WordJoiner /* × WJ */
|| state.Current.LineBreakClass == LineBreakClass.WordJoiner /* WJ × */)
@ -352,7 +352,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB12: Do not break after NBSP and related characters.
/// </summary>
private static RuleResult LB12(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB12(ReadOnlySpan<char> text, ref LineBreakState state)
{
// GL ×
if (state.Current.LineBreakClass == LineBreakClass.Glue)
@ -366,7 +366,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB12a: Do not break before NBSP and related characters, except after spaces and hyphens.
/// </summary>
private static RuleResult LB12a(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB12a(ReadOnlySpan<char> text, ref LineBreakState state)
{
// [^SP BA HY] × GL
if (state.Next(text).LineBreakClass == LineBreakClass.Glue)
@ -388,7 +388,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB13: Do not break before ‘]’ or ‘!’ or ‘;’ or ‘/’, even after spaces.
/// </summary>
private static RuleResult LB13(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB13(ReadOnlySpan<char> text, ref LineBreakState state)
{
// × CL
// × CP
@ -409,7 +409,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB14: Do not break after ‘[’, even after spaces.
/// </summary>
private static RuleResult LB14(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB14(ReadOnlySpan<char> text, ref LineBreakState state)
{
// OP SP* ×
if (state.LastBeforeWhitespace.LineBreakClass == LineBreakClass.OpenPunctuation)
@ -424,7 +424,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// LB15a: Do not break after an unresolved initial punctuation that lies at the start of the line,
/// after a space, after opening punctuation, or after an unresolved quotation mark, even after spaces.
/// </summary>
private static RuleResult LB15a(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB15a(ReadOnlySpan<char> text, ref LineBreakState state)
{
// (sot | BK | CR | LF | NL | OP | QU | GL | SP | ZW) [\p{Pi}&QU] SP* ×
if (state.Quotation > 0 && state.LastBeforeWhitespace.Codepoint.GeneralCategory == GeneralCategory.InitialPunctuation &&
@ -487,7 +487,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// LB15b: Do not break before an unresolved final punctuation that lies at the end of the line,
/// before a space, before a prohibited break, or before an unresolved quotation mark, even after spaces.
/// </summary>
private static RuleResult LB15b(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB15b(ReadOnlySpan<char> text, ref LineBreakState state)
{
// × [\p{Pf}&QU] ( SP | GL | WJ | CL | QU | CP | EX | IS | SY | BK | CR | LF | NL | ZW | eot)
if (state.Next(text).Codepoint.GeneralCategory == GeneralCategory.FinalPunctuation && (state.Next(text).LineBreakClass == LineBreakClass.Quotation))
@ -526,7 +526,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB15c: Break before a decimal mark that follows a space, for instance, in ‘subtract .5’.
/// </summary>
private static RuleResult LB15c(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB15c(ReadOnlySpan<char> text, ref LineBreakState state)
{
// SP ÷ IS NU
if (state.Current.LineBreakClass == LineBreakClass.Space)
@ -546,7 +546,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB15d: Otherwise, do not break before ‘;’, ‘,’, or ‘.’, even after spaces.
/// </summary>
private static RuleResult LB15d(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB15d(ReadOnlySpan<char> text, ref LineBreakState state)
{
// × IS
if (state.Next(text).LineBreakClass == LineBreakClass.InfixNumeric)
@ -562,7 +562,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// LB16: Do not break between closing punctuation and a nonstarter (lb=NS),
/// even with intervening spaces.
/// </summary>
private static RuleResult LB16(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB16(ReadOnlySpan<char> text, ref LineBreakState state)
{
switch (state.LastBeforeWhitespace.LineBreakClass)
{
@ -590,7 +590,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB17: Do not break within ‘——’, even with intervening spaces.
/// </summary>
private static RuleResult LB17(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB17(ReadOnlySpan<char> text, ref LineBreakState state)
{
// B2 SP* × B2
if (state.LastBeforeWhitespace.LineBreakClass == LineBreakClass.BreakBoth)
@ -607,7 +607,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB18: Break after spaces.
/// </summary>
private static RuleResult LB18(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB18(ReadOnlySpan<char> text, ref LineBreakState state)
{
// SP ÷
if (state.Current.LineBreakClass == LineBreakClass.Space)
@ -621,7 +621,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB19: Do not break before or after quotation marks, such as ‘ ” ’.
/// </summary>
private static RuleResult LB19(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB19(ReadOnlySpan<char> text, ref LineBreakState state)
{
var next = state.Next(text);
@ -677,7 +677,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB20: Break before and after unresolved CB.
/// </summary>
private static RuleResult LB20(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB20(ReadOnlySpan<char> text, ref LineBreakState state)
{
// ÷ CB
// CB ÷
@ -692,7 +692,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB20a: Do not break after a word-initial hyphen.
/// </summary>
private static RuleResult LB20a(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB20a(ReadOnlySpan<char> text, ref LineBreakState state)
{
// (sot | BK | CR | LF | NL | SP | ZW | CB | GL)(HY | [\u2010]) × AL
if (IsMatch(state.Previous) && state.Next(text).LineBreakClass == LineBreakClass.Alphabetic)
@ -733,7 +733,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB21: Do not break before hyphen-minus, other hyphens, fixed-width spaces, small kana, and other non-starters, or after acute accents.
/// </summary>
private static RuleResult LB21(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB21(ReadOnlySpan<char> text, ref LineBreakState state)
{
// × (BA | HY | NS)
switch (state.Next(text).LineBreakClass)
@ -760,7 +760,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB21a: Don't break after Hebrew + Hyphen.
/// </summary>
private static RuleResult LB21a(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB21a(ReadOnlySpan<char> text, ref LineBreakState state)
{
if(state.Next(text).LineBreakClass != LineBreakClass.HebrewLetter)
{
@ -778,7 +778,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB21b: Don’t break between Solidus and Hebrew letters.
/// </summary>
private static RuleResult LB21b(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB21b(ReadOnlySpan<char> text, ref LineBreakState state)
{
// [21.2] SY × HL
if ((state.Current.LineBreakClass == LineBreakClass.BreakSymbols) && (state.Next(text).LineBreakClass == LineBreakClass.HebrewLetter))
@ -792,7 +792,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB22: Do not break before ellipses.
/// </summary>
private static RuleResult LB22(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB22(ReadOnlySpan<char> text, ref LineBreakState state)
{
// × IN
if (state.Next(text).LineBreakClass == LineBreakClass.Inseparable)
@ -806,7 +806,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB23: Do not break between digits and letters.
/// </summary>
private static RuleResult LB23(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB23(ReadOnlySpan<char> text, ref LineBreakState state)
{
switch (state.Current.LineBreakClass)
{
@ -842,7 +842,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// LB23a: Do not break between numeric prefixes and ideographs, or between
/// ideographs and numeric postfixes.
/// </summary>
private static RuleResult LB23a(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB23a(ReadOnlySpan<char> text, ref LineBreakState state)
{
// PR × (ID | EB | EM)
if ((state.Current.LineBreakClass == LineBreakClass.PrefixNumeric)
@ -878,7 +878,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// LB24: Do not break between numeric prefix/postfix and letters, or between
/// letters and prefix/postfix.
/// </summary>
private static RuleResult LB24(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB24(ReadOnlySpan<char> text, ref LineBreakState state)
{
// (PR | PO) × (AL | HL)
if (state.Current.LineBreakClass is LineBreakClass.PrefixNumeric or LineBreakClass.PostfixNumeric
@ -900,7 +900,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB25: Do not break between the following pairs of classes relevant to numbers
/// </summary>
private static RuleResult LB25(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB25(ReadOnlySpan<char> text, ref LineBreakState state)
{
switch (state.Next(text).LineBreakClass)
{
@ -1128,7 +1128,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB26: Do not break a Korean syllable.
/// </summary>
private static RuleResult LB26(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB26(ReadOnlySpan<char> text, ref LineBreakState state)
{
switch (state.Current.LineBreakClass)
{
@ -1176,7 +1176,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB27: Treat a Korean Syllable Block the same as ID.
/// </summary>
private static RuleResult LB27(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB27(ReadOnlySpan<char> text, ref LineBreakState state)
{
switch (state.Current.LineBreakClass)
{
@ -1216,7 +1216,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB28: Do not break between alphabetics (“at”).
/// </summary>
private static RuleResult LB28(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB28(ReadOnlySpan<char> text, ref LineBreakState state)
{
// [28.0] (AL | HL) × (AL | HL)
switch (state.Current.LineBreakClass)
@ -1243,7 +1243,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB28a: Do not break inside the orthographic syllables of Brahmic scripts.
/// </summary>
private static RuleResult LB28a(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB28a(ReadOnlySpan<char> text, ref LineBreakState state)
{
// [28.11] AP × (AK | DottedCircle | AS)
if ((state.Current.LineBreakClass == LineBreakClass.AksaraPrebase) && isMatch(state.Next(text)))
@ -1286,7 +1286,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB29: Do not break between numeric punctuation and alphabetics (“e.g.”).
/// </summary>
private static RuleResult LB29(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB29(ReadOnlySpan<char> text, ref LineBreakState state)
{
// IS × (AL | HL)
if ((state.Current.LineBreakClass == LineBreakClass.InfixNumeric)
@ -1301,7 +1301,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB30: Do not break between letters, numbers, or ordinary symbols and opening or closing parentheses.
/// </summary>
private static RuleResult LB30(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB30(ReadOnlySpan<char> text, ref LineBreakState state)
{
switch (state.Current.LineBreakClass)
{
@ -1346,7 +1346,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// break.
/// </summary>
/// <returns></returns>
private static RuleResult LB30a(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB30a(ReadOnlySpan<char> text, ref LineBreakState state)
{
if (state.RegionalIndicator > 0 && state.Next(text).LineBreakClass == LineBreakClass.RegionalIndicator)
{
@ -1363,7 +1363,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// LB30b: Do not break between an emoji base (or potential emoji) and an emoji modifier.
/// </summary>
/// <returns></returns>
private static RuleResult LB30b(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB30b(ReadOnlySpan<char> text, ref LineBreakState state)
{
// EB × EM
if ((state.Current.LineBreakClass == LineBreakClass.EBase) && (state.Next(text).LineBreakClass == LineBreakClass.EModifier))
@ -1392,7 +1392,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
/// <summary>
/// LB31: Break everywhere else.
/// </summary>
private static RuleResult LB31(ReadOnlySpan<char> text, LineBreakState state)
private static RuleResult LB31(ReadOnlySpan<char> text, ref LineBreakState state)
{
return RuleResult.MayBreak;
}
@ -1491,7 +1491,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
}
}
private class LineBreakState
private ref struct LineBreakState
{
private BreakUnit? _next;
private BreakUnit _previous;
@ -1651,7 +1651,7 @@ namespace Avalonia.Media.TextFormatting.Unicode
}
}
private delegate RuleResult BreakUnitDelegate(ReadOnlySpan<char> text, LineBreakState state);
private delegate RuleResult BreakUnitDelegate(ReadOnlySpan<char> text, ref LineBreakState state);
private static readonly BreakUnitDelegate[] s_rules = [
QuotationAndRegionalIndicator,

18
src/Avalonia.Base/Media/TextFormatting/Unicode/UnicodeData.trie.cs
View File

@ -16,10 +16,18 @@ internal static class UnicodeDataTrie
{
[MethodImpl(MethodImplOptions.AggressiveInlining)]
get => new(Data, 0x00100000, 0x00000000);
}
private static ReadOnlySpan<uint> Data => new uint[]
{
}
// For Debug builds, we store the data in a separate uint[] field to avoid RuntimeFieldInfoStub allocations.
// See also: https://github.com/AvaloniaUI/Avalonia/pull/20175
#if DEBUG
public static ReadOnlySpan<uint> Data => s_data;
private static uint[] s_data =
#else
public static ReadOnlySpan<uint> Data =>
#endif
[
0x00000482, 0x0000048A, 0x00000492, 0x0000049A, 0x000004B2, 0x000004BA, 0x000004C2, 0x000004CA, 0x000004D2, 0x000004DA, 0x000004E0, 0x000004E8,
0x000004F0, 0x000004F8, 0x00000500, 0x00000508, 0x0000050E, 0x00000516, 0x0000051E, 0x00000526, 0x00000529, 0x00000531, 0x00000539, 0x00000541,
0x00000549, 0x00000551, 0x00000556, 0x0000055E, 0x00000566, 0x0000056E, 0x00000573, 0x0000057B, 0x00000583, 0x0000058B, 0x0000058F, 0x00000597,
@ -2351,5 +2359,5 @@ internal static class UnicodeDataTrie
0x00038061, 0x00038061, 0x00038061, 0x00038061, 0x00038061, 0x00074061, 0x00074061, 0x00074061, 0x00038061, 0x00074061, 0x00074061, 0x00074061,
0x00074061, 0x00074061, 0x00074061, 0x00074061, 0x00074061, 0x00074061, 0x00074061, 0x00074061, 0x00074061, 0x00074061, 0x00038061, 0x00038061,
0x00000000, 0x00000000, 0x00000000, 0x00000000
};
];
}

4
src/Markup/Avalonia.Markup/Markup/Parsers/ContainerQueryGrammar.cs
View File

@ -84,7 +84,7 @@ namespace Avalonia.Markup.Parsers
if (identifier.SequenceEqual(MinWidthKeyword.AsSpan()) || identifier.SequenceEqual(MaxWidthKeyword.AsSpan()) || identifier.SequenceEqual(WidthKeyword.AsSpan()))
{
if (!r.TakeIf(':'))
throw new ExpressionParseException(r.Position, $"Expected ':' after '{identifier}'.");
throw new ExpressionParseException(r.Position, $"Expected ':' after '{identifier.ToString()}'.");
double val = ParseDecimal(ref r);
var syntax = new WidthSyntax()
@ -101,7 +101,7 @@ namespace Avalonia.Markup.Parsers
if (identifier.SequenceEqual(MinHeightKeyword.AsSpan()) || identifier.SequenceEqual(MaxHeightKeyword.AsSpan()) || identifier.SequenceEqual(HeightKeyword.AsSpan()))
{
if (!r.TakeIf(':'))
throw new ExpressionParseException(r.Position, $"Expected ':' after '{identifier}'.");
throw new ExpressionParseException(r.Position, $"Expected ':' after '{identifier.ToString()}'.");
double val = ParseDecimal(ref r);
var syntax = new HeightSyntax()

9
src/Skia/Avalonia.Skia/TextShaperImpl.cs
View File

@ -43,7 +43,14 @@ namespace Avalonia.Skia
var usedCulture = culture ?? CultureInfo.CurrentCulture;
buffer.Language = s_cachedLanguage.GetOrAdd(usedCulture.LCID, _ => new Language(usedCulture));
#if NETSTANDARD2_0
buffer.Language = s_cachedLanguage.GetOrAdd(usedCulture.LCID, _ => new Language(culture));
#else
buffer.Language = s_cachedLanguage.GetOrAdd(
usedCulture.LCID,
static (_, culture) => new Language(culture),
usedCulture);
#endif
var font = ((GlyphTypefaceImpl)typeface).Font;

20
tests/Avalonia.Base.UnitTests/Media/TextFormatting/UnicodeDataGenerator.cs
View File

@ -67,7 +67,7 @@ namespace Avalonia.Base.UnitTests.Media.TextFormatting
public static void GenerateTrieClass(string name, UnicodeTrie trie)
{
using var fileStream = File.Create($"Generated\\{name}.trie.cs");
using var fileStream = File.Create(Path.Combine("Generated", $"{name}.trie.cs"));
using var writer = new StreamWriter(fileStream);
writer.Write(
@ -90,10 +90,18 @@ namespace Avalonia.Base.UnitTests.Media.TextFormatting
{
[MethodImpl(MethodImplOptions.AggressiveInlining)]
get => new(Data, 0x{{trie.HighStart:X8}}, 0x{{trie.ErrorValue:X8}});
}
private static ReadOnlySpan<uint> Data => new uint[]
{
}
// For Debug builds, we store the data in a separate uint[] field to avoid RuntimeFieldInfoStub allocations.
// See also: https://github.com/AvaloniaUI/Avalonia/pull/20175
#if DEBUG
public static ReadOnlySpan<uint> Data => s_data;
private static uint[] s_data =
#else
public static ReadOnlySpan<uint> Data =>
#endif
[
""");
for (int i = 0; i < trie.Data.Length; ++i)
@ -114,7 +122,7 @@ namespace Avalonia.Base.UnitTests.Media.TextFormatting
writer.Write(
"""
};
];
}
""");

Write Preview
Loading…
Cancel
Save