aboutsummaryrefslogtreecommitdiff
path: root/vendor/golang.org/x/text/internal/ucd/ucd.go
blob: 60b27d51bec36a70629a59df6e12bcd9d9adecac (plain)
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
// Copyright 2014 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.

// Package ucd provides a parser for Unicode Character Database files, the
// format of which is defined in http://www.unicode.org/reports/tr44/. See
// http://www.unicode.org/Public/UCD/latest/ucd/ for example files.
//
// It currently does not support substitutions of missing fields.
package ucd // import "golang.org/x/text/internal/ucd"

import (
	"bufio"
	"bytes"
	"errors"
	"fmt"
	"io"
	"log"
	"regexp"
	"strconv"
	"strings"
)

// UnicodeData.txt fields.
const (
	CodePoint = iota
	Name
	GeneralCategory
	CanonicalCombiningClass
	BidiClass
	DecompMapping
	DecimalValue
	DigitValue
	NumericValue
	BidiMirrored
	Unicode1Name
	ISOComment
	SimpleUppercaseMapping
	SimpleLowercaseMapping
	SimpleTitlecaseMapping
)

// Parse calls f for each entry in the given reader of a UCD file. It will close
// the reader upon return. It will call log.Fatal if any error occurred.
//
// This implements the most common usage pattern of using Parser.
func Parse(r io.ReadCloser, f func(p *Parser)) {
	defer r.Close()

	p := New(r)
	for p.Next() {
		f(p)
	}
	if err := p.Err(); err != nil {
		r.Close() // os.Exit will cause defers not to be called.
		log.Fatal(err)
	}
}

// An Option is used to configure a Parser.
type Option func(p *Parser)

func keepRanges(p *Parser) {
	p.keepRanges = true
}

var (
	// KeepRanges prevents the expansion of ranges. The raw ranges can be
	// obtained by calling Range(0) on the parser.
	KeepRanges Option = keepRanges
)

// The Part option register a handler for lines starting with a '@'. The text
// after a '@' is available as the first field. Comments are handled as usual.
func Part(f func(p *Parser)) Option {
	return func(p *Parser) {
		p.partHandler = f
	}
}

// The CommentHandler option passes comments that are on a line by itself to
// a given handler.
func CommentHandler(f func(s string)) Option {
	return func(p *Parser) {
		p.commentHandler = f
	}
}

// A Parser parses Unicode Character Database (UCD) files.
type Parser struct {
	scanner *bufio.Scanner

	keepRanges bool // Don't expand rune ranges in field 0.

	err     error
	comment []byte
	field   [][]byte
	// parsedRange is needed in case Range(0) is called more than once for one
	// field. In some cases this requires scanning ahead.
	parsedRange          bool
	rangeStart, rangeEnd rune

	partHandler    func(p *Parser)
	commentHandler func(s string)
}

func (p *Parser) setError(err error) {
	if p.err == nil {
		p.err = err
	}
}

func (p *Parser) getField(i int) []byte {
	if i >= len(p.field) {
		p.setError(fmt.Errorf("ucd: index of field %d out of bounds", i))
		return nil
	}
	return p.field[i]
}

// Err returns a non-nil error if any error occurred during parsing.
func (p *Parser) Err() error {
	return p.err
}

// New returns a Parser for the given Reader.
func New(r io.Reader, o ...Option) *Parser {
	p := &Parser{
		scanner: bufio.NewScanner(r),
	}
	for _, f := range o {
		f(p)
	}
	return p
}

// Next parses the next line in the file. It returns true if a line was parsed
// and false if it reached the end of the file.
func (p *Parser) Next() bool {
	if !p.keepRanges && p.rangeStart < p.rangeEnd {
		p.rangeStart++
		return true
	}
	p.comment = nil
	p.field = p.field[:0]
	p.parsedRange = false

	for p.scanner.Scan() {
		b := p.scanner.Bytes()
		if len(b) == 0 {
			continue
		}
		if b[0] == '#' {
			if p.commentHandler != nil {
				p.commentHandler(strings.TrimSpace(string(b[1:])))
			}
			continue
		}

		// Parse line
		if i := bytes.IndexByte(b, '#'); i != -1 {
			p.comment = bytes.TrimSpace(b[i+1:])
			b = b[:i]
		}
		if b[0] == '@' {
			if p.partHandler != nil {
				p.field = append(p.field, bytes.TrimSpace(b[1:]))
				p.partHandler(p)
				p.field = p.field[:0]
			}
			p.comment = nil
			continue
		}
		for {
			i := bytes.IndexByte(b, ';')
			if i == -1 {
				p.field = append(p.field, bytes.TrimSpace(b))
				break
			}
			p.field = append(p.field, bytes.TrimSpace(b[:i]))
			b = b[i+1:]
		}
		if !p.keepRanges {
			p.rangeStart, p.rangeEnd = p.getRange(0)
		}
		return true
	}
	p.setError(p.scanner.Err())
	return false
}

func parseRune(b []byte) (rune, error) {
	if len(b) > 2 && b[0] == 'U' && b[1] == '+' {
		b = b[2:]
	}
	x, err := strconv.ParseUint(string(b), 16, 32)
	return rune(x), err
}

func (p *Parser) parseRune(b []byte) rune {
	x, err := parseRune(b)
	p.setError(err)
	return x
}

// Rune parses and returns field i as a rune.
func (p *Parser) Rune(i int) rune {
	if i > 0 || p.keepRanges {
		return p.parseRune(p.getField(i))
	}
	return p.rangeStart
}

// Runes interprets and returns field i as a sequence of runes.
func (p *Parser) Runes(i int) (runes []rune) {
	add := func(b []byte) {
		if b = bytes.TrimSpace(b); len(b) > 0 {
			runes = append(runes, p.parseRune(b))
		}
	}
	for b := p.getField(i); ; {
		i := bytes.IndexByte(b, ' ')
		if i == -1 {
			add(b)
			break
		}
		add(b[:i])
		b = b[i+1:]
	}
	return
}

var (
	errIncorrectLegacyRange = errors.New("ucd: unmatched <* First>")

	// reRange matches one line of a legacy rune range.
	reRange = regexp.MustCompile("^([0-9A-F]*);<([^,]*), ([^>]*)>(.*)$")
)

// Range parses and returns field i as a rune range. A range is inclusive at
// both ends. If the field only has one rune, first and last will be identical.
// It supports the legacy format for ranges used in UnicodeData.txt.
func (p *Parser) Range(i int) (first, last rune) {
	if !p.keepRanges {
		return p.rangeStart, p.rangeStart
	}
	return p.getRange(i)
}

func (p *Parser) getRange(i int) (first, last rune) {
	b := p.getField(i)
	if k := bytes.Index(b, []byte("..")); k != -1 {
		return p.parseRune(b[:k]), p.parseRune(b[k+2:])
	}
	// The first field may not be a rune, in which case we may ignore any error
	// and set the range as 0..0.
	x, err := parseRune(b)
	if err != nil {
		// Disable range parsing henceforth. This ensures that an error will be
		// returned if the user subsequently will try to parse this field as
		// a Rune.
		p.keepRanges = true
	}
	// Special case for UnicodeData that was retained for backwards compatibility.
	if i == 0 && len(p.field) > 1 && bytes.HasSuffix(p.field[1], []byte("First>")) {
		if p.parsedRange {
			return p.rangeStart, p.rangeEnd
		}
		mf := reRange.FindStringSubmatch(p.scanner.Text())
		if mf == nil || !p.scanner.Scan() {
			p.setError(errIncorrectLegacyRange)
			return x, x
		}
		// Using Bytes would be more efficient here, but Text is a lot easier
		// and this is not a frequent case.
		ml := reRange.FindStringSubmatch(p.scanner.Text())
		if ml == nil || mf[2] != ml[2] || ml[3] != "Last" || mf[4] != ml[4] {
			p.setError(errIncorrectLegacyRange)
			return x, x
		}
		p.rangeStart, p.rangeEnd = x, p.parseRune(p.scanner.Bytes()[:len(ml[1])])
		p.parsedRange = true
		return p.rangeStart, p.rangeEnd
	}
	return x, x
}

// bools recognizes all valid UCD boolean values.
var bools = map[string]bool{
	"":      false,
	"N":     false,
	"No":    false,
	"F":     false,
	"False": false,
	"Y":     true,
	"Yes":   true,
	"T":     true,
	"True":  true,
}

// Bool parses and returns field i as a boolean value.
func (p *Parser) Bool(i int) bool {
	b := p.getField(i)
	for s, v := range bools {
		if bstrEq(b, s) {
			return v
		}
	}
	p.setError(strconv.ErrSyntax)
	return false
}

// Int parses and returns field i as an integer value.
func (p *Parser) Int(i int) int {
	x, err := strconv.ParseInt(string(p.getField(i)), 10, 64)
	p.setError(err)
	return int(x)
}

// Uint parses and returns field i as an unsigned integer value.
func (p *Parser) Uint(i int) uint {
	x, err := strconv.ParseUint(string(p.getField(i)), 10, 64)
	p.setError(err)
	return uint(x)
}

// Float parses and returns field i as a decimal value.
func (p *Parser) Float(i int) float64 {
	x, err := strconv.ParseFloat(string(p.getField(i)), 64)
	p.setError(err)
	return x
}

// String parses and returns field i as a string value.
func (p *Parser) String(i int) string {
	return string(p.getField(i))
}

// Strings parses and returns field i as a space-separated list of strings.
func (p *Parser) Strings(i int) []string {
	ss := strings.Split(string(p.getField(i)), " ")
	for i, s := range ss {
		ss[i] = strings.TrimSpace(s)
	}
	return ss
}

// Comment returns the comments for the current line.
func (p *Parser) Comment() string {
	return string(p.comment)
}

var errUndefinedEnum = errors.New("ucd: undefined enum value")

// Enum interprets and returns field i as a value that must be one of the values
// in enum.
func (p *Parser) Enum(i int, enum ...string) string {
	b := p.getField(i)
	for _, s := range enum {
		if bstrEq(b, s) {
			return s
		}
	}
	p.setError(errUndefinedEnum)
	return ""
}

func bstrEq(b []byte, s string) bool {
	if len(b) != len(s) {
		return false
	}
	for i, c := range b {
		if c != s[i] {
			return false
		}
	}
	return true
}