diff --git a/src/SixLabors.Fonts/Unicode/LineBreakEnumerator.cs b/src/SixLabors.Fonts/Unicode/LineBreakEnumerator.cs index 1d81a2be..864576a2 100644 --- a/src/SixLabors.Fonts/Unicode/LineBreakEnumerator.cs +++ b/src/SixLabors.Fonts/Unicode/LineBreakEnumerator.cs @@ -6,22 +6,45 @@ namespace SixLabors.Fonts.Unicode; /// -/// Supports a simple iteration over a linebreak collection. -/// Implementation of the Unicode Line Break Algorithm. UAX:14 -/// +/// Enumerates potential line break opportunities for a span of text. +/// This is the engine behind the Unicode Line Breaking Algorithm as defined by +/// Unicode Standard Annex #14 (UAX #14): +/// . +/// +/// The Unicode rules are the source of truth for the logic below. Comments intentionally +/// call out the UAX rule numbers so readers can cross-check the implementation against the +/// specification instead of treating the local state flags as standalone behavior. +/// +/// The implementation walks one code point at a time, keeps a two-code-point window +/// ( on the left and on the right), and +/// carries a small amount of extra state for rules that depend on more than the current pair. /// Methods are pattern-matched by compiler to allow using foreach pattern. /// internal ref struct LineBreakEnumerator { + // Iteration state: + // - charPosition is the UTF-16 offset into the source span. + // - position is the code point index after the most recently consumed code point. + // - lastPosition is the candidate break boundary between currentClass and nextClass. private readonly ReadOnlySpan source; private int charPosition; private readonly int pointsLength; private int position; private int lastPosition; + + // The active pair under consideration. currentClass is the left side of the boundary and + // nextClass is the right side. Most of UAX #14 reduces to deciding whether that pair breaks. private LineBreakClass currentClass; private LineBreakClass nextClass; private bool first; + + // Tracks whether we are inside an AL/HL/NU run, including trailing combining marks. + // This is used by LB30, which needs more context than the pair table alone can express. private int alphaNumericCount; + + // Stateful rule flags for the UAX #14 rules that cannot be represented by a simple + // currentClass/nextClass lookup. The field names intentionally mirror the rule numbers + // so the implementation can be read side by side with the spec. private bool lb8a; private bool lb21a; private bool lb22ex; @@ -31,6 +54,10 @@ internal ref struct LineBreakEnumerator private int lb30a; private bool lb31; + /// + /// Initializes a new instance of the struct. + /// + /// The source text to inspect for UAX #14 break opportunities. public LineBreakEnumerator(ReadOnlySpan source) : this() { @@ -53,6 +80,9 @@ public LineBreakEnumerator(ReadOnlySpan source) this.lb30a = 0; } + /// + /// Gets the most recently discovered line break opportunity. + /// public LineBreak Current { get; private set; } /// @@ -70,7 +100,8 @@ public LineBreakEnumerator(ReadOnlySpan source) /// public bool MoveNext() { - // Get the first char if we're at the beginning of the string. + // Prime the left side of the pair window. After this block the loop below always + // decides the boundary between currentClass (left) and nextClass (right). if (this.first) { LineBreakClass firstClass = this.NextCharClass(); @@ -87,7 +118,8 @@ public bool MoveNext() LineBreakClass lastClass = this.nextClass; this.nextClass = this.NextCharClass(); - // Explicit newline + // Required breaks from BK/CR/LF are emitted before any pair-table logic. + // This matches the UAX handling for explicit line terminators. switch (this.currentClass) { case LineBreakClass.BK: @@ -97,9 +129,12 @@ public bool MoveNext() return true; } + // Handle the classes that have bespoke UAX behavior first, then fall back to the + // pair table plus the stateful exceptions tracked on this enumerator. bool? shouldBreak = this.GetSimpleBreak() ?? (bool?)this.GetPairTableBreak(lastClass); - // Rule LB8a + // LB8a suppresses breaks after a ZWJ. We record it after processing the current + // boundary so it applies to the next boundary instead of the one we just decided. this.lb8a = this.nextClass == LineBreakClass.ZWJ; if (shouldBreak.Value) @@ -121,6 +156,8 @@ public bool MoveNext() break; } + // UAX also exposes an end-of-text boundary. Callers use that final boundary to + // finalize the trailing line even when no earlier break opportunity was taken. this.Current = new LineBreak(this.FindPriorNonWhitespace(this.pointsLength), this.lastPosition, required); return true; } @@ -129,6 +166,9 @@ public bool MoveNext() return false; } + /// + /// Applies the LB1 class remapping required before any pair-table decisions are made. + /// private static LineBreakClass MapClass(CodePoint cp, LineBreakClass c) { // LB 1 @@ -160,6 +200,9 @@ private static LineBreakClass MapClass(CodePoint cp, LineBreakClass c) } } + /// + /// Applies the start-of-text normalization required for the first class in the stream. + /// private static LineBreakClass MapFirst(LineBreakClass c) => c switch { @@ -168,18 +211,29 @@ private static LineBreakClass MapFirst(LineBreakClass c) _ => c, }; + /// + /// Returns for the classes treated as alphanumeric by the + /// stateful LB30 handling. + /// private static bool IsAlphaNumeric(LineBreakClass cls) => cls is LineBreakClass.AL or LineBreakClass.HL or LineBreakClass.NU; + /// + /// Reads the next class without advancing the enumerator. This is only used by rules such as + /// LB25 that need one-code-point lookahead to confirm a numeric punctuation sequence. + /// private readonly LineBreakClass PeekNextCharClass() { CodePoint cp = CodePoint.DecodeFromUtf16At(this.source, this.charPosition); return MapClass(cp, CodePoint.GetLineBreakClass(cp)); } - // Get the next character class + /// + /// Consumes the next code point, applies LB1 class resolution, and updates any stateful + /// rule flags that depend on more than the current pair. + /// private LineBreakClass NextCharClass() { CodePoint cp = CodePoint.DecodeFromUtf16At(this.source, this.charPosition, out int count); @@ -187,14 +241,17 @@ private LineBreakClass NextCharClass() this.charPosition += count; this.position++; - // Keep track of alphanumeric + any combining marks. - // This is used for LB22 and LB30. + // Track an alphanumeric run together with any trailing combining marks. LB30 needs to + // know whether an opening punctuation follows such a run, but once currentClass advances + // we would otherwise lose that earlier context. if (IsAlphaNumeric(this.currentClass) || (this.alphaNumericCount > 0 && cls == LineBreakClass.CM)) { this.alphaNumericCount++; } - // Track combining mark exceptions. LB22 + // LB22 distinguishes between "CM after one of the explicitly allowed classes" and + // "CM after anything else" when the next class is IN. Record that now before currentClass + // collapses to CM on the next iteration. if (cls == LineBreakClass.CM) { switch (this.currentClass) @@ -212,7 +269,8 @@ private LineBreakClass NextCharClass() } } - // Track combining mark exceptions. LB31 + // LB31 is another context-sensitive rule. We record the contexts that permit a break + // before an opening punctuation and defer the final decision until OP is nextClass. if (this.first && cls == LineBreakClass.CM) { this.lb31 = true; @@ -257,19 +315,21 @@ private LineBreakClass NextCharClass() this.lb31 = false; } - // Rule LB24 + // Seed the multi-code-point context used later by LB24. if (this.first && (cls == LineBreakClass.CL || cls == LineBreakClass.CP)) { this.lb24ex = true; } - // Rule LB25 + // Seed the multi-code-point context used later by LB25. if (this.first && (cls == LineBreakClass.CL || cls == LineBreakClass.IS || cls == LineBreakClass.SY)) { this.lb25ex = true; } + // LB25 spans punctuation, spaces, and surrounding numeric runs. Look ahead one code point + // so punctuation after a space or letter can still participate in the same numeric context. if (cls is LineBreakClass.SP or LineBreakClass.WJ or LineBreakClass.AL) { LineBreakClass next = this.PeekNextCharClass(); @@ -294,9 +354,15 @@ private LineBreakClass NextCharClass() return cls; } + /// + /// Handles the UAX rules for spaces and explicit line terminators before falling back to the + /// pair table for the ordinary pair-based decisions. + /// private bool? GetSimpleBreak() { - // handle classes not handled by the pair table + // These classes are easier to express directly than through the pair table: + // - spaces never break immediately before themselves, + // - hard line terminators update currentClass so the next iteration emits a required break. switch (this.nextClass) { case LineBreakClass.SP: @@ -316,9 +382,14 @@ private LineBreakClass NextCharClass() return null; } + /// + /// Applies the pair-table result and then layers on the stateful UAX exceptions that require + /// additional context beyond the current pair. + /// private bool GetPairTableBreak(LineBreakClass lastClass) { - // If not handled already, use the pair table + // The pair table is the baseline answer from UAX #14. The rule-specific flags below + // then tighten or loosen that answer where the spec requires extra context. bool shouldBreak = false; switch (LineBreakPairTable.Table[(int)this.currentClass][(int)this.nextClass]) { @@ -387,6 +458,20 @@ private bool GetPairTableBreak(LineBreakClass lastClass) break; } + if (lastClass == LineBreakClass.SP + && this.nextClass is LineBreakClass.AL or LineBreakClass.HL) + { + // Once we have already broken at the punctuation-adjacent space, keep the + // LB25 state alive only if the following letters immediately continue into + // another CL/IS/SY sequence. Otherwise the punctuation context has ended + // and must not leak forward to a later AL|NU pair. + LineBreakClass ahead = this.PeekNextCharClass(); + if (ahead is not LineBreakClass.CL and not LineBreakClass.IS and not LineBreakClass.SY) + { + this.lb25ex = false; + } + } + if (this.nextClass is LineBreakClass.PR or LineBreakClass.NU) { shouldBreak = true; @@ -428,7 +513,9 @@ private bool GetPairTableBreak(LineBreakClass lastClass) break; } - // Rule LB22 + // Apply the remaining non-pair-table rules in the same place every time so the + // interaction between pair-table output and rule-specific overrides stays obvious. + // LB22 if (this.nextClass == LineBreakClass.IN) { switch (lastClass) @@ -459,6 +546,7 @@ private bool GetPairTableBreak(LineBreakClass lastClass) } } + // LB8a suppresses a break after ZWJ regardless of the pair-table answer. if (this.lb8a) { shouldBreak = false; @@ -490,7 +578,9 @@ private bool GetPairTableBreak(LineBreakClass lastClass) this.lb30a = 0; } - // Rule LB30b + // LB30b depends on the Extended_Pictographic property, but Mahjong tiles are a special case: + // their Line_Break class is ID even though they live in an Extended_Pictographic-adjacent + // block, so we need an explicit guard here to mirror the Unicode test data. if (this.nextClass == LineBreakClass.EM && this.lastPosition > 0) { // Mahjong Tiles (Unicode block) are extended pictographics but have a class of ID @@ -508,6 +598,11 @@ private bool GetPairTableBreak(LineBreakClass lastClass) return shouldBreak; } + /// + /// Walks backward from a wrap position to the nearest non-breaking trailing content so that + /// measurement excludes trailing spaces and hard line terminators while wrapping still occurs + /// at the original boundary. + /// private readonly int FindPriorNonWhitespace(int from) { if (from > 0) diff --git a/src/SixLabors.Fonts/Unicode/LineBreakPairTable.cs b/src/SixLabors.Fonts/Unicode/LineBreakPairTable.cs index 00515d63..10ae6369 100644 --- a/src/SixLabors.Fonts/Unicode/LineBreakPairTable.cs +++ b/src/SixLabors.Fonts/Unicode/LineBreakPairTable.cs @@ -3,6 +3,16 @@ namespace SixLabors.Fonts.Unicode; +/// +/// Provides Unicode line break pair classification values and the pair table used to determine line break opportunities +/// between character classes. +/// +/// +/// This class supplies constants representing different types of line break opportunities, as defined by +/// the Unicode Line Breaking Algorithm (UAX #14), and a lookup table for resolving break opportunities between pairs of +/// line break classes. The table is based on the Unicode specification and is intended for use in text layout and line +/// breaking implementations. +/// internal static class LineBreakPairTable { /// @@ -33,41 +43,42 @@ internal static class LineBreakPairTable // Based on example pair table from https://www.unicode.org/reports/tr14/tr14-37.html#Table2 // - ZWJ special processing for LB8a // - CB manually added as per Rule LB20 - public static byte[][] Table { get; } = new byte[][] - { + public static byte[][] Table { get; } = + [ + // . OP CL CP QU GL NS EX SY IS PR PO NU AL HL ID IN HY BA BB B2 ZW CM WJ H2 H3 JL JV JT RI EB EM ZWJ CB - new byte[] { PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, CPBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK }, // OP - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // CL - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // CP - new byte[] { PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK }, // QU - new byte[] { INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK }, // GL - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // NS - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // EX - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, INBRK, DIBRK, INBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // SY - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // IS - new byte[] { INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK }, // PR - new byte[] { INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // PO - new byte[] { INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // NU - new byte[] { INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // AL - new byte[] { INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // HL - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // ID - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // IN - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, DIBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // HY - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, DIBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // BA - new byte[] { INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK }, // BB - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, PRBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // B2 - new byte[] { DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK }, // ZW - new byte[] { INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // CM - new byte[] { INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK }, // WJ - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // H2 - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // H3 - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // JL - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // JV - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // JT - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK, DIBRK, INBRK, DIBRK }, // RI - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK }, // EB - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // EM - new byte[] { INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK }, // ZWJ - new byte[] { DIBRK, PRBRK, PRBRK, INBRK, INBRK, DIBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK } // CB - }; + [PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, CPBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK, PRBRK], // OP + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // CL + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // CP + [PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK], // QU + [INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK], // GL + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // NS + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // EX + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, INBRK, DIBRK, INBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // SY + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // IS + [INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK], // PR + [INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // PO + [INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // NU + [INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // AL + [INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // HL + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // ID + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // IN + [DIBRK, PRBRK, PRBRK, INBRK, DIBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // HY + [DIBRK, PRBRK, PRBRK, INBRK, DIBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // BA + [INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK], // BB + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, PRBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // B2 + [DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK], // ZW + [INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // CM + [INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK, INBRK], // WJ + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // H2 + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // H3 + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // JL + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // JV + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // JT + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK, DIBRK, INBRK, DIBRK], // RI + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, DIBRK], // EB + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, DIBRK, INBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // EM + [INBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, PRBRK, PRBRK, PRBRK, INBRK, INBRK, INBRK, INBRK, INBRK, DIBRK, INBRK, INBRK, INBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK], // ZWJ + [DIBRK, PRBRK, PRBRK, INBRK, INBRK, DIBRK, PRBRK, PRBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, PRBRK, CIBRK, PRBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, DIBRK, INBRK, DIBRK] // CB + ]; } diff --git a/tests/SixLabors.Fonts.Tests/Issues/Issues_522.cs b/tests/SixLabors.Fonts.Tests/Issues/Issues_522.cs new file mode 100644 index 00000000..3a87e277 --- /dev/null +++ b/tests/SixLabors.Fonts.Tests/Issues/Issues_522.cs @@ -0,0 +1,37 @@ +// Copyright (c) Six Labors. +// Licensed under the Six Labors Split License. + +using SixLabors.Fonts.Unicode; + +namespace SixLabors.Fonts.Tests.Issues; + +public class Issues_522 +{ + [Fact] + public void LineBreakEnumerator_DoesNotBreakWithinAlphaNumericRunAfterColon() + { + string text = "số khung: RRKWCH1UM7XJ00693"; + List breaks = [.. new LineBreakEnumerator(text.AsSpan())]; + + Assert.Collection( + breaks, + x => + { + Assert.Equal(2, x.PositionMeasure); + Assert.Equal(3, x.PositionWrap); + Assert.False(x.Required); + }, + x => + { + Assert.Equal(9, x.PositionMeasure); + Assert.Equal(10, x.PositionWrap); + Assert.False(x.Required); + }, + x => + { + Assert.Equal(27, x.PositionMeasure); + Assert.Equal(27, x.PositionWrap); + Assert.False(x.Required); + }); + } +}