diff options
Diffstat (limited to 'vendor/golang.org/x/text/unicode/bidi')
-rw-r--r-- | vendor/golang.org/x/text/unicode/bidi/bidi.go | 221 | ||||
-rw-r--r-- | vendor/golang.org/x/text/unicode/bidi/core.go | 63 |
2 files changed, 229 insertions, 55 deletions
diff --git a/vendor/golang.org/x/text/unicode/bidi/bidi.go b/vendor/golang.org/x/text/unicode/bidi/bidi.go index e8edc54c..fd057601 100644 --- a/vendor/golang.org/x/text/unicode/bidi/bidi.go +++ b/vendor/golang.org/x/text/unicode/bidi/bidi.go @@ -12,15 +12,14 @@ // and without notice. package bidi // import "golang.org/x/text/unicode/bidi" -// TODO: -// The following functionality would not be hard to implement, but hinges on -// the definition of a Segmenter interface. For now this is up to the user. -// - Iterate over paragraphs -// - Segmenter to iterate over runs directly from a given text. -// Also: +// TODO // - Transformer for reordering? // - Transformer (validator, really) for Bidi Rule. +import ( + "bytes" +) + // This API tries to avoid dealing with embedding levels for now. Under the hood // these will be computed, but the question is to which extent the user should // know they exist. We should at some point allow the user to specify an @@ -49,7 +48,9 @@ const ( Neutral ) -type options struct{} +type options struct { + defaultDirection Direction +} // An Option is an option for Bidi processing. type Option func(*options) @@ -66,12 +67,62 @@ type Option func(*options) // DefaultDirection sets the default direction for a Paragraph. The direction is // overridden if the text contains directional characters. func DefaultDirection(d Direction) Option { - panic("unimplemented") + return func(opts *options) { + opts.defaultDirection = d + } } // A Paragraph holds a single Paragraph for Bidi processing. type Paragraph struct { - // buffers + p []byte + o Ordering + opts []Option + types []Class + pairTypes []bracketType + pairValues []rune + runes []rune + options options +} + +// Initialize the p.pairTypes, p.pairValues and p.types from the input previously +// set by p.SetBytes() or p.SetString(). Also limit the input up to (and including) a paragraph +// separator (bidi class B). +// +// The function p.Order() needs these values to be set, so this preparation could be postponed. +// But since the SetBytes and SetStrings functions return the length of the input up to the paragraph +// separator, the whole input needs to be processed anyway and should not be done twice. +// +// The function has the same return values as SetBytes() / SetString() +func (p *Paragraph) prepareInput() (n int, err error) { + p.runes = bytes.Runes(p.p) + bytecount := 0 + // clear slices from previous SetString or SetBytes + p.pairTypes = nil + p.pairValues = nil + p.types = nil + + for _, r := range p.runes { + props, i := LookupRune(r) + bytecount += i + cls := props.Class() + if cls == B { + return bytecount, nil + } + p.types = append(p.types, cls) + if props.IsOpeningBracket() { + p.pairTypes = append(p.pairTypes, bpOpen) + p.pairValues = append(p.pairValues, r) + } else if props.IsBracket() { + // this must be a closing bracket, + // since IsOpeningBracket is not true + p.pairTypes = append(p.pairTypes, bpClose) + p.pairValues = append(p.pairValues, r) + } else { + p.pairTypes = append(p.pairTypes, bpNone) + p.pairValues = append(p.pairValues, 0) + } + } + return bytecount, nil } // SetBytes configures p for the given paragraph text. It replaces text @@ -80,70 +131,150 @@ type Paragraph struct { // consumed from b including this separator. Error may be non-nil if options are // given. func (p *Paragraph) SetBytes(b []byte, opts ...Option) (n int, err error) { - panic("unimplemented") + p.p = b + p.opts = opts + return p.prepareInput() } -// SetString configures p for the given paragraph text. It replaces text -// previously set by SetBytes or SetString. If b contains a paragraph separator +// SetString configures s for the given paragraph text. It replaces text +// previously set by SetBytes or SetString. If s contains a paragraph separator // it will only process the first paragraph and report the number of bytes -// consumed from b including this separator. Error may be non-nil if options are +// consumed from s including this separator. Error may be non-nil if options are // given. func (p *Paragraph) SetString(s string, opts ...Option) (n int, err error) { - panic("unimplemented") + p.p = []byte(s) + p.opts = opts + return p.prepareInput() } // IsLeftToRight reports whether the principle direction of rendering for this // paragraphs is left-to-right. If this returns false, the principle direction // of rendering is right-to-left. func (p *Paragraph) IsLeftToRight() bool { - panic("unimplemented") + return p.Direction() == LeftToRight } // Direction returns the direction of the text of this paragraph. // // The direction may be LeftToRight, RightToLeft, Mixed, or Neutral. func (p *Paragraph) Direction() Direction { - panic("unimplemented") + return p.o.Direction() } +// TODO: what happens if the position is > len(input)? This should return an error. + // RunAt reports the Run at the given position of the input text. // // This method can be used for computing line breaks on paragraphs. func (p *Paragraph) RunAt(pos int) Run { - panic("unimplemented") + c := 0 + runNumber := 0 + for i, r := range p.o.runes { + c += len(r) + if pos < c { + runNumber = i + } + } + return p.o.Run(runNumber) +} + +func calculateOrdering(levels []level, runes []rune) Ordering { + var curDir Direction + + prevDir := Neutral + prevI := 0 + + o := Ordering{} + // lvl = 0,2,4,...: left to right + // lvl = 1,3,5,...: right to left + for i, lvl := range levels { + if lvl%2 == 0 { + curDir = LeftToRight + } else { + curDir = RightToLeft + } + if curDir != prevDir { + if i > 0 { + o.runes = append(o.runes, runes[prevI:i]) + o.directions = append(o.directions, prevDir) + o.startpos = append(o.startpos, prevI) + } + prevI = i + prevDir = curDir + } + } + o.runes = append(o.runes, runes[prevI:]) + o.directions = append(o.directions, prevDir) + o.startpos = append(o.startpos, prevI) + return o } // Order computes the visual ordering of all the runs in a Paragraph. func (p *Paragraph) Order() (Ordering, error) { - panic("unimplemented") + if len(p.types) == 0 { + return Ordering{}, nil + } + + for _, fn := range p.opts { + fn(&p.options) + } + lvl := level(-1) + if p.options.defaultDirection == RightToLeft { + lvl = 1 + } + para, err := newParagraph(p.types, p.pairTypes, p.pairValues, lvl) + if err != nil { + return Ordering{}, err + } + + levels := para.getLevels([]int{len(p.types)}) + + p.o = calculateOrdering(levels, p.runes) + return p.o, nil } // Line computes the visual ordering of runs for a single line starting and // ending at the given positions in the original text. func (p *Paragraph) Line(start, end int) (Ordering, error) { - panic("unimplemented") + lineTypes := p.types[start:end] + para, err := newParagraph(lineTypes, p.pairTypes[start:end], p.pairValues[start:end], -1) + if err != nil { + return Ordering{}, err + } + levels := para.getLevels([]int{len(lineTypes)}) + o := calculateOrdering(levels, p.runes[start:end]) + return o, nil } // An Ordering holds the computed visual order of runs of a Paragraph. Calling // SetBytes or SetString on the originating Paragraph invalidates an Ordering. // The methods of an Ordering should only be called by one goroutine at a time. -type Ordering struct{} +type Ordering struct { + runes [][]rune + directions []Direction + startpos []int +} // Direction reports the directionality of the runs. // // The direction may be LeftToRight, RightToLeft, Mixed, or Neutral. func (o *Ordering) Direction() Direction { - panic("unimplemented") + return o.directions[0] } // NumRuns returns the number of runs. func (o *Ordering) NumRuns() int { - panic("unimplemented") + return len(o.runes) } // Run returns the ith run within the ordering. func (o *Ordering) Run(i int) Run { - panic("unimplemented") + r := Run{ + runes: o.runes[i], + direction: o.directions[i], + startpos: o.startpos[i], + } + return r } // TODO: perhaps with options. @@ -155,16 +286,19 @@ func (o *Ordering) Run(i int) Run { // A Run is a continuous sequence of characters of a single direction. type Run struct { + runes []rune + direction Direction + startpos int } // String returns the text of the run in its original order. func (r *Run) String() string { - panic("unimplemented") + return string(r.runes) } // Bytes returns the text of the run in its original order. func (r *Run) Bytes() []byte { - panic("unimplemented") + return []byte(r.String()) } // TODO: methods for @@ -174,25 +308,52 @@ func (r *Run) Bytes() []byte { // Direction reports the direction of the run. func (r *Run) Direction() Direction { - panic("unimplemented") + return r.direction } -// Position of the Run within the text passed to SetBytes or SetString of the +// Pos returns the position of the Run within the text passed to SetBytes or SetString of the // originating Paragraph value. func (r *Run) Pos() (start, end int) { - panic("unimplemented") + return r.startpos, r.startpos + len(r.runes) - 1 } // AppendReverse reverses the order of characters of in, appends them to out, // and returns the result. Modifiers will still follow the runes they modify. // Brackets are replaced with their counterparts. func AppendReverse(out, in []byte) []byte { - panic("unimplemented") + ret := make([]byte, len(in)+len(out)) + copy(ret, out) + inRunes := bytes.Runes(in) + + for i, r := range inRunes { + prop, _ := LookupRune(r) + if prop.IsBracket() { + inRunes[i] = prop.reverseBracket(r) + } + } + + for i, j := 0, len(inRunes)-1; i < j; i, j = i+1, j-1 { + inRunes[i], inRunes[j] = inRunes[j], inRunes[i] + } + copy(ret[len(out):], string(inRunes)) + + return ret } // ReverseString reverses the order of characters in s and returns a new string. // Modifiers will still follow the runes they modify. Brackets are replaced with // their counterparts. func ReverseString(s string) string { - panic("unimplemented") + input := []rune(s) + li := len(input) + ret := make([]rune, li) + for i, r := range input { + prop, _ := LookupRune(r) + if prop.IsBracket() { + ret[li-i-1] = prop.reverseBracket(r) + } else { + ret[li-i-1] = r + } + } + return string(ret) } diff --git a/vendor/golang.org/x/text/unicode/bidi/core.go b/vendor/golang.org/x/text/unicode/bidi/core.go index 50deb660..e4c08110 100644 --- a/vendor/golang.org/x/text/unicode/bidi/core.go +++ b/vendor/golang.org/x/text/unicode/bidi/core.go @@ -4,7 +4,10 @@ package bidi -import "log" +import ( + "fmt" + "log" +) // This implementation is a port based on the reference implementation found at: // https://www.unicode.org/Public/PROGRAMS/BidiReferenceJava/ @@ -97,13 +100,20 @@ type paragraph struct { // rune (suggested is the rune of the open bracket for opening and matching // close brackets, after normalization). The embedding levels are optional, but // may be supplied to encode embedding levels of styled text. -// -// TODO: return an error. -func newParagraph(types []Class, pairTypes []bracketType, pairValues []rune, levels level) *paragraph { - validateTypes(types) - validatePbTypes(pairTypes) - validatePbValues(pairValues, pairTypes) - validateParagraphEmbeddingLevel(levels) +func newParagraph(types []Class, pairTypes []bracketType, pairValues []rune, levels level) (*paragraph, error) { + var err error + if err = validateTypes(types); err != nil { + return nil, err + } + if err = validatePbTypes(pairTypes); err != nil { + return nil, err + } + if err = validatePbValues(pairValues, pairTypes); err != nil { + return nil, err + } + if err = validateParagraphEmbeddingLevel(levels); err != nil { + return nil, err + } p := ¶graph{ initialTypes: append([]Class(nil), types...), @@ -115,7 +125,7 @@ func newParagraph(types []Class, pairTypes []bracketType, pairValues []rune, lev resultTypes: append([]Class(nil), types...), } p.run() - return p + return p, nil } func (p *paragraph) Len() int { return len(p.initialTypes) } @@ -1001,58 +1011,61 @@ func typeForLevel(level level) Class { return R } -// TODO: change validation to not panic - -func validateTypes(types []Class) { +func validateTypes(types []Class) error { if len(types) == 0 { - log.Panic("types is null") + return fmt.Errorf("types is null") } for i, t := range types[:len(types)-1] { if t == B { - log.Panicf("B type before end of paragraph at index: %d", i) + return fmt.Errorf("B type before end of paragraph at index: %d", i) } } + return nil } -func validateParagraphEmbeddingLevel(embeddingLevel level) { +func validateParagraphEmbeddingLevel(embeddingLevel level) error { if embeddingLevel != implicitLevel && embeddingLevel != 0 && embeddingLevel != 1 { - log.Panicf("illegal paragraph embedding level: %d", embeddingLevel) + return fmt.Errorf("illegal paragraph embedding level: %d", embeddingLevel) } + return nil } -func validateLineBreaks(linebreaks []int, textLength int) { +func validateLineBreaks(linebreaks []int, textLength int) error { prev := 0 for i, next := range linebreaks { if next <= prev { - log.Panicf("bad linebreak: %d at index: %d", next, i) + return fmt.Errorf("bad linebreak: %d at index: %d", next, i) } prev = next } if prev != textLength { - log.Panicf("last linebreak was %d, want %d", prev, textLength) + return fmt.Errorf("last linebreak was %d, want %d", prev, textLength) } + return nil } -func validatePbTypes(pairTypes []bracketType) { +func validatePbTypes(pairTypes []bracketType) error { if len(pairTypes) == 0 { - log.Panic("pairTypes is null") + return fmt.Errorf("pairTypes is null") } for i, pt := range pairTypes { switch pt { case bpNone, bpOpen, bpClose: default: - log.Panicf("illegal pairType value at %d: %v", i, pairTypes[i]) + return fmt.Errorf("illegal pairType value at %d: %v", i, pairTypes[i]) } } + return nil } -func validatePbValues(pairValues []rune, pairTypes []bracketType) { +func validatePbValues(pairValues []rune, pairTypes []bracketType) error { if pairValues == nil { - log.Panic("pairValues is null") + return fmt.Errorf("pairValues is null") } if len(pairTypes) != len(pairValues) { - log.Panic("pairTypes is different length from pairValues") + return fmt.Errorf("pairTypes is different length from pairValues") } + return nil } |