line.go

  1package uniseg
  2
  3import "unicode/utf8"
  4
  5// FirstLineSegment returns the prefix of the given byte slice after which a
  6// decision to break the string over to the next line can or must be made,
  7// according to the rules of [Unicode Standard Annex #14]. This is used to
  8// implement line breaking.
  9//
 10// Line breaking, also known as word wrapping, is the process of breaking a
 11// section of text into lines such that it will fit in the available width of a
 12// page, window or other display area.
 13//
 14// The returned "segment" may not be broken into smaller parts, unless no other
 15// breaking opportunities present themselves, in which case you may break by
 16// grapheme clusters (using the [FirstGraphemeCluster] function to determine the
 17// grapheme clusters).
 18//
 19// The "mustBreak" flag indicates whether you MUST break the line after the
 20// given segment (true), for example after newline characters, or you MAY break
 21// the line after the given segment (false).
 22//
 23// This function can be called continuously to extract all non-breaking sub-sets
 24// from a byte slice, as illustrated in the example below.
 25//
 26// If you don't know the current state, for example when calling the function
 27// for the first time, you must pass -1. For consecutive calls, pass the state
 28// and rest slice returned by the previous call.
 29//
 30// The "rest" slice is the sub-slice of the original byte slice "b" starting
 31// after the last byte of the identified line segment. If the length of the
 32// "rest" slice is 0, the entire byte slice "b" has been processed. The
 33// "segment" byte slice is the sub-slice of the input slice containing the
 34// identified line segment.
 35//
 36// Given an empty byte slice "b", the function returns nil values.
 37//
 38// Note that in accordance with [UAX #14 LB3], the final segment will end with
 39// "mustBreak" set to true. You can choose to ignore this by checking if the
 40// length of the "rest" slice is 0 and calling [HasTrailingLineBreak] or
 41// [HasTrailingLineBreakInString] on the last rune.
 42//
 43// Note also that this algorithm may break within grapheme clusters. This is
 44// addressed in Section 8.2 Example 6 of UAX #14. To avoid this, you can use
 45// the [Step] function instead.
 46//
 47// [Unicode Standard Annex #14]: https://www.unicode.org/reports/tr14/
 48// [UAX #14 LB3]: https://www.unicode.org/reports/tr14/#Algorithm
 49func FirstLineSegment(b []byte, state int) (segment, rest []byte, mustBreak bool, newState int) {
 50	// An empty byte slice returns nothing.
 51	if len(b) == 0 {
 52		return
 53	}
 54
 55	// Extract the first rune.
 56	r, length := utf8.DecodeRune(b)
 57	if len(b) <= length { // If we're already past the end, there is nothing else to parse.
 58		return b, nil, true, lbAny // LB3.
 59	}
 60
 61	// If we don't know the state, determine it now.
 62	if state < 0 {
 63		state, _ = transitionLineBreakState(state, r, b[length:], "")
 64	}
 65
 66	// Transition until we find a boundary.
 67	var boundary int
 68	for {
 69		r, l := utf8.DecodeRune(b[length:])
 70		state, boundary = transitionLineBreakState(state, r, b[length+l:], "")
 71
 72		if boundary != LineDontBreak {
 73			return b[:length], b[length:], boundary == LineMustBreak, state
 74		}
 75
 76		length += l
 77		if len(b) <= length {
 78			return b, nil, true, lbAny // LB3
 79		}
 80	}
 81}
 82
 83// FirstLineSegmentInString is like [FirstLineSegment] but its input and outputs
 84// are strings.
 85func FirstLineSegmentInString(str string, state int) (segment, rest string, mustBreak bool, newState int) {
 86	// An empty byte slice returns nothing.
 87	if len(str) == 0 {
 88		return
 89	}
 90
 91	// Extract the first rune.
 92	r, length := utf8.DecodeRuneInString(str)
 93	if len(str) <= length { // If we're already past the end, there is nothing else to parse.
 94		return str, "", true, lbAny // LB3.
 95	}
 96
 97	// If we don't know the state, determine it now.
 98	if state < 0 {
 99		state, _ = transitionLineBreakState(state, r, nil, str[length:])
100	}
101
102	// Transition until we find a boundary.
103	var boundary int
104	for {
105		r, l := utf8.DecodeRuneInString(str[length:])
106		state, boundary = transitionLineBreakState(state, r, nil, str[length+l:])
107
108		if boundary != LineDontBreak {
109			return str[:length], str[length:], boundary == LineMustBreak, state
110		}
111
112		length += l
113		if len(str) <= length {
114			return str, "", true, lbAny // LB3.
115		}
116	}
117}
118
119// HasTrailingLineBreak returns true if the last rune in the given byte slice is
120// one of the hard line break code points defined in LB4 and LB5 of [UAX #14].
121//
122// [UAX #14]: https://www.unicode.org/reports/tr14/#Algorithm
123func HasTrailingLineBreak(b []byte) bool {
124	r, _ := utf8.DecodeLastRune(b)
125	property, _ := propertyLineBreak(r)
126	return property == prBK || property == prCR || property == prLF || property == prNL
127}
128
129// HasTrailingLineBreakInString is like [HasTrailingLineBreak] but for a string.
130func HasTrailingLineBreakInString(str string) bool {
131	r, _ := utf8.DecodeLastRuneInString(str)
132	property, _ := propertyLineBreak(r)
133	return property == prBK || property == prCR || property == prLF || property == prNL
134}