summaryrefslogtreecommitdiffhomepage
path: root/vendor/golang.org/x/text/collate/collate.go
blob: d8c23cb696ac4c46f929a904a470d3df53571ac3 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
// Copyright 2012 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.

// TODO: remove hard-coded versions when we have implemented fractional weights.
// The current implementation is incompatible with later CLDR versions.
//go:generate go run maketables.go -cldr=23 -unicode=6.2.0

// Package collate contains types for comparing and sorting Unicode strings
// according to a given collation order.
package collate // import "golang.org/x/text/collate"

import (
	"bytes"
	"strings"

	"golang.org/x/text/internal/colltab"
	"golang.org/x/text/language"
)

// Collator provides functionality for comparing strings for a given
// collation order.
type Collator struct {
	options

	sorter sorter

	_iter [2]iter
}

func (c *Collator) iter(i int) *iter {
	// TODO: evaluate performance for making the second iterator optional.
	return &c._iter[i]
}

// Supported returns the list of languages for which collating differs from its parent.
func Supported() []language.Tag {
	// TODO: use language.Coverage instead.

	t := make([]language.Tag, len(tags))
	copy(t, tags)
	return t
}

func init() {
	ids := strings.Split(availableLocales, ",")
	tags = make([]language.Tag, len(ids))
	for i, s := range ids {
		tags[i] = language.Raw.MustParse(s)
	}
}

var tags []language.Tag

// New returns a new Collator initialized for the given locale.
func New(t language.Tag, o ...Option) *Collator {
	index := colltab.MatchLang(t, tags)
	c := newCollator(getTable(locales[index]))

	// Set options from the user-supplied tag.
	c.setFromTag(t)

	// Set the user-supplied options.
	c.setOptions(o)

	c.init()
	return c
}

// NewFromTable returns a new Collator for the given Weighter.
func NewFromTable(w colltab.Weighter, o ...Option) *Collator {
	c := newCollator(w)
	c.setOptions(o)
	c.init()
	return c
}

func (c *Collator) init() {
	if c.numeric {
		c.t = colltab.NewNumericWeighter(c.t)
	}
	c._iter[0].init(c)
	c._iter[1].init(c)
}

// Buffer holds keys generated by Key and KeyString.
type Buffer struct {
	buf [4096]byte
	key []byte
}

func (b *Buffer) init() {
	if b.key == nil {
		b.key = b.buf[:0]
	}
}

// Reset clears the buffer from previous results generated by Key and KeyString.
func (b *Buffer) Reset() {
	b.key = b.key[:0]
}

// Compare returns an integer comparing the two byte slices.
// The result will be 0 if a==b, -1 if a < b, and +1 if a > b.
func (c *Collator) Compare(a, b []byte) int {
	// TODO: skip identical prefixes once we have a fast way to detect if a rune is
	// part of a contraction. This would lead to roughly a 10% speedup for the colcmp regtest.
	c.iter(0).SetInput(a)
	c.iter(1).SetInput(b)
	if res := c.compare(); res != 0 {
		return res
	}
	if !c.ignore[colltab.Identity] {
		return bytes.Compare(a, b)
	}
	return 0
}

// CompareString returns an integer comparing the two strings.
// The result will be 0 if a==b, -1 if a < b, and +1 if a > b.
func (c *Collator) CompareString(a, b string) int {
	// TODO: skip identical prefixes once we have a fast way to detect if a rune is
	// part of a contraction. This would lead to roughly a 10% speedup for the colcmp regtest.
	c.iter(0).SetInputString(a)
	c.iter(1).SetInputString(b)
	if res := c.compare(); res != 0 {
		return res
	}
	if !c.ignore[colltab.Identity] {
		if a < b {
			return -1
		} else if a > b {
			return 1
		}
	}
	return 0
}

func compareLevel(f func(i *iter) int, a, b *iter) int {
	a.pce = 0
	b.pce = 0
	for {
		va := f(a)
		vb := f(b)
		if va != vb {
			if va < vb {
				return -1
			}
			return 1
		} else if va == 0 {
			break
		}
	}
	return 0
}

func (c *Collator) compare() int {
	ia, ib := c.iter(0), c.iter(1)
	// Process primary level
	if c.alternate != altShifted {
		// TODO: implement script reordering
		if res := compareLevel((*iter).nextPrimary, ia, ib); res != 0 {
			return res
		}
	} else {
		// TODO: handle shifted
	}
	if !c.ignore[colltab.Secondary] {
		f := (*iter).nextSecondary
		if c.backwards {
			f = (*iter).prevSecondary
		}
		if res := compareLevel(f, ia, ib); res != 0 {
			return res
		}
	}
	// TODO: special case handling (Danish?)
	if !c.ignore[colltab.Tertiary] || c.caseLevel {
		if res := compareLevel((*iter).nextTertiary, ia, ib); res != 0 {
			return res
		}
		if !c.ignore[colltab.Quaternary] {
			if res := compareLevel((*iter).nextQuaternary, ia, ib); res != 0 {
				return res
			}
		}
	}
	return 0
}

// Key returns the collation key for str.
// Passing the buffer buf may avoid memory allocations.
// The returned slice will point to an allocation in Buffer and will remain
// valid until the next call to buf.Reset().
func (c *Collator) Key(buf *Buffer, str []byte) []byte {
	// See https://www.unicode.org/reports/tr10/#Main_Algorithm for more details.
	buf.init()
	return c.key(buf, c.getColElems(str))
}

// KeyFromString returns the collation key for str.
// Passing the buffer buf may avoid memory allocations.
// The returned slice will point to an allocation in Buffer and will retain
// valid until the next call to buf.ResetKeys().
func (c *Collator) KeyFromString(buf *Buffer, str string) []byte {
	// See https://www.unicode.org/reports/tr10/#Main_Algorithm for more details.
	buf.init()
	return c.key(buf, c.getColElemsString(str))
}

func (c *Collator) key(buf *Buffer, w []colltab.Elem) []byte {
	processWeights(c.alternate, c.t.Top(), w)
	kn := len(buf.key)
	c.keyFromElems(buf, w)
	return buf.key[kn:]
}

func (c *Collator) getColElems(str []byte) []colltab.Elem {
	i := c.iter(0)
	i.SetInput(str)
	for i.Next() {
	}
	return i.Elems
}

func (c *Collator) getColElemsString(str string) []colltab.Elem {
	i := c.iter(0)
	i.SetInputString(str)
	for i.Next() {
	}
	return i.Elems
}

type iter struct {
	wa [512]colltab.Elem

	colltab.Iter
	pce int
}

func (i *iter) init(c *Collator) {
	i.Weighter = c.t
	i.Elems = i.wa[:0]
}

func (i *iter) nextPrimary() int {
	for {
		for ; i.pce < i.N; i.pce++ {
			if v := i.Elems[i.pce].Primary(); v != 0 {
				i.pce++
				return v
			}
		}
		if !i.Next() {
			return 0
		}
	}
	panic("should not reach here")
}

func (i *iter) nextSecondary() int {
	for ; i.pce < len(i.Elems); i.pce++ {
		if v := i.Elems[i.pce].Secondary(); v != 0 {
			i.pce++
			return v
		}
	}
	return 0
}

func (i *iter) prevSecondary() int {
	for ; i.pce < len(i.Elems); i.pce++ {
		if v := i.Elems[len(i.Elems)-i.pce-1].Secondary(); v != 0 {
			i.pce++
			return v
		}
	}
	return 0
}

func (i *iter) nextTertiary() int {
	for ; i.pce < len(i.Elems); i.pce++ {
		if v := i.Elems[i.pce].Tertiary(); v != 0 {
			i.pce++
			return int(v)
		}
	}
	return 0
}

func (i *iter) nextQuaternary() int {
	for ; i.pce < len(i.Elems); i.pce++ {
		if v := i.Elems[i.pce].Quaternary(); v != 0 {
			i.pce++
			return v
		}
	}
	return 0
}

func appendPrimary(key []byte, p int) []byte {
	// Convert to variable length encoding; supports up to 23 bits.
	if p <= 0x7FFF {
		key = append(key, uint8(p>>8), uint8(p))
	} else {
		key = append(key, uint8(p>>16)|0x80, uint8(p>>8), uint8(p))
	}
	return key
}

// keyFromElems converts the weights ws to a compact sequence of bytes.
// The result will be appended to the byte buffer in buf.
func (c *Collator) keyFromElems(buf *Buffer, ws []colltab.Elem) {
	for _, v := range ws {
		if w := v.Primary(); w > 0 {
			buf.key = appendPrimary(buf.key, w)
		}
	}
	if !c.ignore[colltab.Secondary] {
		buf.key = append(buf.key, 0, 0)
		// TODO: we can use one 0 if we can guarantee that all non-zero weights are > 0xFF.
		if !c.backwards {
			for _, v := range ws {
				if w := v.Secondary(); w > 0 {
					buf.key = append(buf.key, uint8(w>>8), uint8(w))
				}
			}
		} else {
			for i := len(ws) - 1; i >= 0; i-- {
				if w := ws[i].Secondary(); w > 0 {
					buf.key = append(buf.key, uint8(w>>8), uint8(w))
				}
			}
		}
	} else if c.caseLevel {
		buf.key = append(buf.key, 0, 0)
	}
	if !c.ignore[colltab.Tertiary] || c.caseLevel {
		buf.key = append(buf.key, 0, 0)
		for _, v := range ws {
			if w := v.Tertiary(); w > 0 {
				buf.key = append(buf.key, uint8(w))
			}
		}
		// Derive the quaternary weights from the options and other levels.
		// Note that we represent MaxQuaternary as 0xFF. The first byte of the
		// representation of a primary weight is always smaller than 0xFF,
		// so using this single byte value will compare correctly.
		if !c.ignore[colltab.Quaternary] && c.alternate >= altShifted {
			if c.alternate == altShiftTrimmed {
				lastNonFFFF := len(buf.key)
				buf.key = append(buf.key, 0)
				for _, v := range ws {
					if w := v.Quaternary(); w == colltab.MaxQuaternary {
						buf.key = append(buf.key, 0xFF)
					} else if w > 0 {
						buf.key = appendPrimary(buf.key, w)
						lastNonFFFF = len(buf.key)
					}
				}
				buf.key = buf.key[:lastNonFFFF]
			} else {
				buf.key = append(buf.key, 0)
				for _, v := range ws {
					if w := v.Quaternary(); w == colltab.MaxQuaternary {
						buf.key = append(buf.key, 0xFF)
					} else if w > 0 {
						buf.key = appendPrimary(buf.key, w)
					}
				}
			}
		}
	}
}

func processWeights(vw alternateHandling, top uint32, wa []colltab.Elem) {
	ignore := false
	vtop := int(top)
	switch vw {
	case altShifted, altShiftTrimmed:
		for i := range wa {
			if p := wa[i].Primary(); p <= vtop && p != 0 {
				wa[i] = colltab.MakeQuaternary(p)
				ignore = true
			} else if p == 0 {
				if ignore {
					wa[i] = colltab.Ignore
				}
			} else {
				ignore = false
			}
		}
	case altBlanked:
		for i := range wa {
			if p := wa[i].Primary(); p <= vtop && (ignore || p != 0) {
				wa[i] = colltab.Ignore
				ignore = true
			} else {
				ignore = false
			}
		}
	}
}