Repository navigation
Expand file tree
/
Copy pathparser.go
More file actions
565 lines (496 loc) · 14.1 KB
/
Copy pathparser.go
File metadata and controls
565 lines (496 loc) · 14.1 KB
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
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
10000
554
555
556
557
558
559
560
561
562
563
564
565
// Copyright 2015 Unknwon
//
// Licensed under the Apache License, Version 2.0 (the "License"): you may
// not use this file except in compliance with the License. You may obtain
// a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
// WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
// License for the specific language governing permissions and limitations
// under the License.
package ini
import (
"bufio"
"bytes"
"fmt"
"io"
"regexp"
"strconv"
"strings"
"unicode"
)
const minReaderBufferSize = 4096
var pythonMultiline = regexp.MustCompile(`^([\t\f ]+)(.*)`)
type parserOptions struct {
IgnoreContinuation bool
IgnoreInlineComment bool
AllowPythonMultilineValues bool
SpaceBeforeInlineComment bool
UnescapeValueDoubleQuotes bool
UnescapeValueCommentSymbols bool
PreserveSurroundedQuote bool
DebugFunc DebugFunc
ReaderBufferSize int
}
type parser struct {
buf *bufio.Reader
options parserOptions
isEOF bool
count int
comment *bytes.Buffer
}
func (p *parser) debug(format string, args ...interface{}) {
if p.options.DebugFunc != nil {
p.options.DebugFunc(fmt.Sprintf(format, args...))
}
}
func newParser(r io.Reader, opts parserOptions) *parser {
size := opts.ReaderBufferSize
if size < minReaderBufferSize {
size = minReaderBufferSize
}
return &parser{
buf: bufio.NewReaderSize(r, size),
options: opts,
count: 1,
comment: &bytes.Buffer{},
}
}
// BOM handles header of UTF-8, UTF-16 LE and UTF-16 BE's BOM format.
// http://en.wikipedia.org/wiki/Byte_order_mark#Representations_of_byte_order_marks_by_encoding
func (p *parser) BOM() error {
// Strip every leading BOM: a file may accidentally carry more than one
// (e.g. a tool prepending a BOM to already-BOM'd content). Stopping after
// the first one leaves the remaining BOM on the next token and makes the
// parser fail with a misleading "key-value delimiter not found" (#390).
for {
mask, err := p.buf.Peek(2)
if err != nil && err != io.EOF {
return err
} else if len(mask) < 2 {
return nil
}
switch {
case mask[0] == 254 && mask[1] == 255:
fallthrough
case mask[0] == 255 && mask[1] == 254:
if _, err = p.buf.Read(mask); err != nil {
return err
}
case mask[0] == 239 && mask[1] == 187:
mask, err := p.buf.Peek(3)
if err != nil && err != io.EOF {
return err
} else if len(mask) < 3 {
return nil
}
if mask[2] != 191 {
return nil
}
if _, err = p.buf.Read(mask); err != nil {
return err
}
default:
return nil
}
}
}
func (p *parser) readUntil(delim byte) ([]byte, error) {
data, err := p.buf.ReadBytes(delim)
if err != nil {
if err == io.EOF {
p.isEOF = true
} else {
return nil, err
}
}
return data, nil
}
func cleanComment(in []byte) ([]byte, bool) {
i := bytes.IndexAny(in, "#;")
if i == -1 {
return nil, false
}
return in[i:], true
}
func readKeyName(delimiters string, in []byte) (string, int, error) {
line := string(in)
// Check if key name surrounded by quotes.
var keyQuote string
switch line[0] {
case '"':
if len(line) > 6 && line[0:3] == `"""` {
keyQuote = `"""`
} else {
keyQuote = `"`
}
case '`':
keyQuote = "`"
}
// Get out key name
var endIdx int
if len(keyQuote) > 0 {
startIdx := len(keyQuote)
// FIXME: fail case -> """"""name"""=value
pos := strings.Index(line[startIdx:], keyQuote)
if pos == -1 {
return "", -1, fmt.Errorf("missing closing key quote: %s", line)
}
pos += startIdx
// Find key-value delimiter
i := strings.IndexAny(line[pos+startIdx:], delimiters)
if i < 0 {
return "", -1, ErrDelimiterNotFound{line}
}
endIdx = pos + i
return strings.TrimSpace(line[startIdx:pos]), endIdx + startIdx + 1, nil
}
endIdx = strings.IndexAny(line, delimiters)
if endIdx < 0 {
return "", -1, ErrDelimiterNotFound{line}
}
if endIdx == 0 {
return "", -1, ErrEmptyKeyName{line}
}
return strings.TrimSpace(line[0:endIdx]), endIdx + 1, nil
}
func (p *parser) readMultilines(line, val, valQuote string) (string, error) {
for {
data, err := p.readUntil('\n')
if err != nil {
return "", err
}
next := string(data)
pos := strings.LastIndex(next, valQuote)
if pos > -1 {
// Check if the line ends with backslash continuation after the quote
restOfLine := strings.TrimRight(next[pos+len(valQuote):], "\r\n")
if !p.options.IgnoreContinuation && strings.HasSuffix(strings.TrimSpace(restOfLine), `\`) {
val += next
continue
}
val += next[:pos]
comment, has := cleanComment([]byte(next[pos:]))
if has {
p.comment.Write(bytes.TrimSpace(comment))
}
break
}
val += next
if p.isEOF {
return "", fmt.Errorf("missing closing key quote from %q to %q", line, next)
}
}
return val, nil
}
func (p *parser) readContinuationLines(val string) (string, error) {
for {
data, err := p.readUntil('\n')
if err != nil {
return "", err
}
next := strings.TrimSpace(string(data))
if len(next) == 0 {
break
}
val += next
if val[len(val)-1] != '\\' {
break
}
val = val[:len(val)-1]
}
return val, nil
}
func inlineCommentIndex(line string, spaceBefore bool) int {
var quote byte
escaped := false
for i := 0; i < len(line); i++ {
ch := line[i]
if quote != 0 {
if escaped {
escaped = false
continue
}
if quote == '"' && ch == '\\' {
escaped = true
continue
}
if ch == quote {
quote = 0
}
continue
}
if i == 0 && (ch == '\'' || ch == '"') {
quote = ch
continue
}
if ch != '#' && ch != ';' {
continue
}
if !spaceBefore {
return i
}
if i > 0 && line[i-1] == ' ' {
return i - 1
}
}
return -1
}
// hasSurroundedQuote check if and only if the first and last characters
// are quotes \" or \'.
// It returns false if any other parts also contain same kind of quotes.
func hasSurroundedQuote(in string, quote byte) bool {
return len(in) >= 2 && in[0] == quote && in[len(in)-1] == quote &&
strings.IndexByte(in[1:], quote) == len(in)-2
}
func (p *parser) readValue(in []byte, bufferSize int) (string, error) {
line := strings.TrimLeftFunc(string(in), unicode.IsSpace)
if len(line) == 0 {
if p.options.AllowPythonMultilineValues && len(in) > 0 && in[len(in)-1] == '\n' {
return p.readPythonMultilines(line, bufferSize)
}
return "", nil
}
var valQuote string
if len(line) > 3 && line[0:3] == `"""` {
valQuote = `"""`
} else if line[0] == '`' {
valQuote = "`"
} else if p.options.UnescapeValueDoubleQuotes && line[0] == '"' {
valQuote = `"`
}
if len(valQuote) > 0 {
startIdx := len(valQuote)
pos := strings.LastIndex(line[startIdx:], valQuote)
// Check for multi-line value
if pos == -1 {
return p.readMultilines(line, line[startIdx:], valQuote)
}
if p.options.UnescapeValueDoubleQuotes && valQuote == `"` {
return strings.ReplaceAll(line[startIdx:pos+startIdx], `\"`, `"`), nil
}
return line[startIdx : pos+startIdx], nil
}
lastChar := line[len(line)-1]
// Won't be able to reach here if value only contains whitespace
line = strings.TrimSpace(line)
trimmedLastChar := line[len(line)-1]
// Check continuation lines when desired
if !p.options.IgnoreContinuation && trimmedLastChar == '\\' {
return p.readContinuationLines(line[:len(line)-1])
}
// Check for inline comments only outside a value's surrounding quotes.
if !p.options.IgnoreInlineComment {
i := inlineCommentIndex(line, p.options.SpaceBeforeInlineComment)
if i > -1 {
p.comment.WriteString(line[i:])
line = strings.TrimSpace(line[:i])
}
}
// Trim single and double quotes
if (hasSurroundedQuote(line, '\'') ||
hasSurroundedQuote(line, '"')) && !p.options.PreserveSurroundedQuote {
line = line[1 : len(line)-1]
} else if len(valQuote) == 0 && p.options.UnescapeValueCommentSymbols {
line = strings.ReplaceAll(line, `\;`, ";")
line = strings.ReplaceAll(line, `\#`, "#")
} else if p.options.AllowPythonMultilineValues && lastChar == '\n' {
return p.readPythonMultilines(line, bufferSize)
}
return line, nil
}
func (p *parser) readPythonMultilines(line string, bufferSize int) (string, error) {
parserBufferPeekResult, _ := p.buf.Peek(bufferSize)
peekBuffer := bytes.NewBuffer(parserBufferPeekResult)
for {
peekData, peekErr := peekBuffer.ReadBytes('\n')
if peekErr != nil && peekErr != io.EOF {
p.debug("readPythonMultilines: failed to peek with error: %v", peekErr)
return "", peekErr
}
p.debug("readPythonMultilines: parsing %q", string(peekData))
peekMatches := pythonMultiline.FindStringSubmatch(string(peekData))
p.debug("readPythonMultilines: matched %d parts", len(peekMatches))
for n, v := range peekMatches {
p.debug(" %d: %q", n, v)
}
// Return if not a Python multiline value.
if len(peekMatches) != 3 {
p.debug("readPythonMultilines: end of value, got: %q", line)
return line, nil
}
// Advance the parser reader (buffer) in-sync with the peek buffer.
_, err := p.buf.Discard(len(peekData))
if err != nil {
p.debug("readPythonMultilines: failed to skip to the end, returning error")
return "", err
}
line += "\n" + peekMatches[0]
}
}
// parse parses data through an io.Reader.
func (f *File) parse(reader io.Reader) (err error) {
p := newParser(reader, parserOptions{
IgnoreContinuation: f.options.IgnoreContinuation,
IgnoreInlineComment: f.options.IgnoreInlineComment,
AllowPythonMultilineValues: f.options.AllowPythonMultilineValues,
SpaceBeforeInlineComment: f.options.SpaceBeforeInlineComment,
UnescapeValueDoubleQuotes: f.options.UnescapeValueDoubleQuotes,
UnescapeValueCommentSymbols: f.options.UnescapeValueCommentSymbols,
PreserveSurroundedQuote: f.options.PreserveSurroundedQuote,
DebugFunc: f.options.DebugFunc,
ReaderBufferSize: f.options.ReaderBufferSize,
})
if err = p.BOM(); err != nil {
return fmt.Errorf("BOM: %v", err)
}
// Ignore error because default section name is never empty string.
name := DefaultSection
if f.options.Insensitive || f.options.InsensitiveSections {
name = strings.ToLower(DefaultSection)
}
section, _ := f.NewSection(name)
// This "last" is not strictly equivalent to "previous one" if current key is not the first nested key
var isLastValueEmpty bool
var lastRegularKey *Key
var line []byte
var inUnparseableSection bool
// NOTE: Iterate and increase `currentPeekSize` until
// the size of the parser buffer is found.
// TODO(unknwon): When Golang 1.10 is the lowest version supported, replace with `parserBufferSize := p.buf.Size()`.
parserBufferSize := 0
// NOTE: Peek 4kb at a time.
currentPeekSize := minReaderBufferSize
if f.options.AllowPythonMultilineValues {
for {
peekBytes, _ := p.buf.Peek(currentPeekSize)
peekBytesLength := len(peekBytes)
if parserBufferSize >= peekBytesLength {
break
}
currentPeekSize *= 2
parserBufferSize = peekBytesLength
}
}
for !p.isEOF {
line, err = p.readUntil('\n')
if err != nil {
return err
}
if f.options.AllowNestedValues &&
isLastValueEmpty && len(line) > 0 {
if line[0] == ' ' || line[0] == '\t' {
err = lastRegularKey.addNestedValue(string(bytes.TrimSpace(line)))
if err != nil {
return err
}
continue
}
}
line = bytes.TrimLeftFunc(line, unicode.IsSpace)
if len(line) == 0 {
continue
}
// Comments
if line[0] == '#' || line[0] == ';' {
// Note: we do not care ending line break,
// it is needed for adding second line,
// so just clean it once at the end when set to value.
p.comment.Write(line)
continue
}
// Section
if line[0] == '[' {
// Read to the next ']' (TODO: support quoted strings)
closeIdx := bytes.LastIndexByte(line, ']')
if closeIdx == -1 {
return fmt.Errorf("unclosed section: %s", line)
}
name := string(line[1:closeIdx])
section, err = f.NewSection(name)
if err != nil {
return err
}
comment, has := cleanComment(line[closeIdx+1:])
if has {
p.comment.Write(comment)
}
section.Comment = strings.TrimSpace(p.comment.String())
// Reset auto-counter and comments
p.comment.Reset()
p.count = 1
// Nested values can't span sections
isLastValueEmpty = false
inUnparseableSection = false
for i := range f.options.UnparseableSections {
if f.options.UnparseableSections[i] == name ||
((f.options.Insensitive || f.options.InsensitiveSections) && strings.EqualFold(f.options.UnparseableSections[i], name)) {
inUnparseableSection = true
section.isRawSection = true
continue
}
}
continue
}
if inUnparseableSection {
section.rawBody += string(line)
continue
}
kname, offset, err := readKeyName(f.options.KeyValueDelimiters, line)
if err != nil {
switch {
// Treat as boolean key when desired, and whole line is key name.
case IsErrDelimiterNotFound(err):
switch {
case f.options.AllowBooleanKeys:
kname, err := p.readValue(line, parserBufferSize)
if err != nil {
return err
}
key, err := section.NewBooleanKey(kname)
if err != nil {
return err
}
key.Comment = strings.TrimSpace(p.comment.String())
p.comment.Reset()
continue
case f.options.SkipUnrecognizableLines:
continue
}
case IsErrEmptyKeyName(err) && f.options.SkipUnrecognizableLines:
continue
}
return err
}
// Auto increment.
isAutoIncr := false
if kname == "-" {
isAutoIncr = true
kname = "#" + strconv.Itoa(p.count)
p.count++
}
value, err := p.readValue(line[offset:], parserBufferSize)
if err != nil {
return err
}
isLastValueEmpty = len(value) == 0
key, err := section.NewKey(kname, value)
if err != nil {
return err
}
key.isAutoIncrement = isAutoIncr
key.Comment = strings.TrimSpace(p.comment.String())
p.comment.Reset()
lastRegularKey = key
}
// Comments after the last key/section are otherwise dropped on EOF
// (e.g. Ansible blockinfile # END markers). Keep them on the last section.
if p.comment.Len() > 0 {
section.TrailingComment = strings.TrimSpace(p.comment.String())
p.comment.Reset()
}
return nil
}