This repository has been archived by the owner on Nov 10, 2020. It is now read-only.
forked from russross/blackfriday
-
Notifications
You must be signed in to change notification settings - Fork 24
/
code.go
317 lines (293 loc) · 6.92 KB
/
code.go
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
// 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.
// Adapted for mmark, by Miek Gieben, 2015.
package mmark
import (
"bufio"
"bytes"
"errors"
"io/ioutil"
"regexp"
"strconv"
"unicode/utf8"
)
// SourceCodeTypes are the different languages that are supported as
// a type attribute in sourcecode, see Section 2.48.4 of XML2RFC v3 (-21).
var SourceCodeTypes = map[string]bool{
"abnf": true,
"asn.1": true,
"bash": true,
"c++": true,
"c": true,
"cbor": true,
"dtd": true,
"java": true,
"javascript": true,
"json": true,
"mib": true,
"perl": true,
"pseudocode": true,
"python": true,
"rnc": true,
"xml": true,
"go": true,
}
// parseAddress parses a code address directive and returns the bytes.
func parseAddress(addr []byte, file []byte) []byte {
bytes.TrimSpace(addr)
textBytes, err := ioutil.ReadFile(string(file))
if err != nil {
printf(nil, "failed: `%s': %s", string(file), err)
return nil
}
lo, hi, err := addrToByteRange(string(addr), 0, textBytes)
if err != nil {
printf(nil, "code include address: %s", err.Error())
return textBytes
}
// Acme pattern matches can stop mid-line,
// so run to end of line in both directions if not at line start/end.
for lo > 0 && textBytes[lo-1] != '\n' {
lo--
}
if hi > 0 {
for hi < len(textBytes) && textBytes[hi-1] != '\n' {
hi++
}
}
lines := codeLines(textBytes, lo, hi)
return lines
}
// codeLines takes a source file and returns the lines that
// span the byte range specified by start and end.
// It discards lines that end in "OMIT" and in "OMIT -->"
func codeLines(src []byte, start, end int) (lines []byte) {
startLine := 1
for i, b := range src {
if i == start {
break
}
if b == '\n' {
startLine++
}
}
s := bufio.NewScanner(bytes.NewReader(src[start:end]))
for n := startLine; s.Scan(); n++ {
l := s.Bytes()
if bytes.HasSuffix(l, []byte("OMIT")) {
continue
}
if bytes.HasSuffix(l, []byte("OMIT -->")) {
continue
}
lines = append(lines, l...)
lines = append(lines, '\n')
}
// TODO(miek): trim leading and trailing blanklines
return
}
// This file is stolen from go/src/cmd/godoc/codewalk.go.
// It's an evaluator for the file address syntax implemented by acme and sam,
// but using Go-native regular expressions.
// To keep things reasonably close, this version uses (?m:re) for all user-provided
// regular expressions. That is the only change to the code from codewalk.go.
// See http://plan9.bell-labs.com/sys/doc/sam/sam.html Table II
// for details on the syntax.
// addrToByte evaluates the given address starting at offset start in data.
// It returns the lo and hi byte offset of the matched region within data.
func addrToByteRange(addr string, start int, data []byte) (lo, hi int, err error) {
if addr == "" {
lo, hi = start, len(data)
return
}
var (
dir byte
prevc byte
charOffset bool
)
lo = start
hi = start
for addr != "" && err == nil {
c := addr[0]
switch c {
default:
err = errors.New("invalid address syntax near " + string(c))
case ',':
if len(addr) == 1 {
hi = len(data)
} else {
_, hi, err = addrToByteRange(addr[1:], hi, data)
}
return
case '+', '-':
if prevc == '+' || prevc == '-' {
lo, hi, err = addrNumber(data, lo, hi, prevc, 1, charOffset)
}
dir = c
case '$':
lo = len(data)
hi = len(data)
if len(addr) > 1 {
dir = '+'
}
case '#':
charOffset = true
case '0', '1', '2', '3', '4', '5', '6', '7', '8', '9':
var i int
for i = 1; i < len(addr); i++ {
if addr[i] < '0' || addr[i] > '9' {
break
}
}
var n int
n, err = strconv.Atoi(addr[0:i])
if err != nil {
break
}
lo, hi, err = addrNumber(data, lo, hi, dir, n, charOffset)
dir = 0
charOffset = false
prevc = c
addr = addr[i:]
continue
case '/':
var i, j int
Regexp:
for i = 1; i < len(addr); i++ {
switch addr[i] {
case '\\':
i++
case '/':
j = i + 1
break Regexp
}
}
if j == 0 {
j = i
}
pattern := addr[1:i]
lo, hi, err = addrRegexp(data, lo, hi, dir, pattern)
prevc = c
addr = addr[j:]
continue
}
prevc = c
addr = addr[1:]
}
if err == nil && dir != 0 {
lo, hi, err = addrNumber(data, lo, hi, dir, 1, charOffset)
}
if err != nil {
return 0, 0, err
}
return lo, hi, nil
}
// addrNumber applies the given dir, n, and charOffset to the address lo, hi.
// dir is '+' or '-', n is the count, and charOffset is true if the syntax
// used was #n. Applying +n (or +#n) means to advance n lines
// (or characters) after hi. Applying -n (or -#n) means to back up n lines
// (or characters) before lo.
// The return value is the new lo, hi.
func addrNumber(data []byte, lo, hi int, dir byte, n int, charOffset bool) (int, int, error) {
switch dir {
case 0:
lo = 0
hi = 0
fallthrough
case '+':
if charOffset {
pos := hi
for ; n > 0 && pos < len(data); n-- {
_, size := utf8.DecodeRune(data[pos:])
pos += size
}
if n == 0 {
return pos, pos, nil
}
break
}
// find next beginning of line
if hi > 0 {
for hi < len(data) && data[hi-1] != '\n' {
hi++
}
}
lo = hi
if n == 0 {
return lo, hi, nil
}
for ; hi < len(data); hi++ {
if data[hi] != '\n' {
continue
}
switch n--; n {
case 1:
lo = hi + 1
case 0:
return lo, hi + 1, nil
}
}
case '-':
if charOffset {
// Scan backward for bytes that are not UTF-8 continuation bytes.
pos := lo
for ; pos > 0 && n > 0; pos-- {
if data[pos]&0xc0 != 0x80 {
n--
}
}
if n == 0 {
return pos, pos, nil
}
break
}
// find earlier beginning of line
for lo > 0 && data[lo-1] != '\n' {
lo--
}
hi = lo
if n == 0 {
return lo, hi, nil
}
for ; lo >= 0; lo-- {
if lo > 0 && data[lo-1] != '\n' {
continue
}
switch n--; n {
case 1:
hi = lo
case 0:
return lo, hi, nil
}
}
}
return 0, 0, errors.New("address out of range")
}
// addrRegexp searches for pattern in the given direction starting at lo, hi.
// The direction dir is '+' (search forward from hi) or '-' (search backward from lo).
// Backward searches are unimplemented.
func addrRegexp(data []byte, lo, hi int, dir byte, pattern string) (int, int, error) {
// We want ^ and $ to work as in sam/acme, so use ?m.
re, err := regexp.Compile("(?m:" + pattern + ")")
if err != nil {
return 0, 0, err
}
if dir == '-' {
// Could implement reverse search using binary search
// through file, but that seems like overkill.
return 0, 0, errors.New("reverse search not implemented")
}
m := re.FindIndex(data[hi:])
if len(m) > 0 {
m[0] += hi
m[1] += hi
} else if hi > 0 {
// No match. Wrap to beginning of data.
m = re.FindIndex(data)
}
if len(m) == 0 {
return 0, 0, errors.New("no match for " + pattern)
}
return m[0], m[1], nil
}