-
Notifications
You must be signed in to change notification settings - Fork 0
/
format.go
330 lines (275 loc) · 10.1 KB
/
format.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
318
319
320
321
322
323
324
325
326
327
328
329
330
package playground
import (
"bytes"
"encoding/json"
"errors"
"fmt"
xhtml "golang.org/x/net/html"
"html"
"io"
"log/slog"
"regexp"
"strings"
"unicode/utf8"
)
type textFormatterImpl struct{}
// format trims left and right spaces from input and returns the actual content as is.
func (textFormatterImpl) format(input []byte, output io.Writer, _ string) {
if len(input) == 0 {
return
}
_, err := output.Write(bytes.TrimSpace(input))
if nil != err {
slog.Error(err.Error())
}
}
type jsonFormatterImpl struct{}
// format implements a JSON formatter.
func (jsonFormatterImpl) format(input []byte, output io.Writer, indent string) {
if len(input) == 0 {
return
}
buffer := bytes.Buffer{}
err := json.Indent(&buffer, input, "", indent)
if nil != err {
slog.Error(err.Error())
return
}
_, err = output.Write(bytes.TrimSpace(buffer.Bytes()))
if nil != err {
slog.Error(err.Error())
}
}
var (
xmlReTag = regexp.MustCompile(`<([/!]?)([^>]+?)(/?)>`) // regexp xmlReTag matches an XML tag
xmlReComment = regexp.MustCompile(`(?s)(<!--)(.*?)(-->)`) // regexp xmlReComment matches an XML comment
xmlReXMLInsideComment = regexp.MustCompile(`<!--[^>]*?<[^>]+?[^-]*?-->`) // regexp xmlReXMLInsideComment matches XML tags inside comments
xmlReBlanksAround = regexp.MustCompile(`\s*(<|/?>)\s*`) // regexp xmlReBlanksAround matches blanks around XML tags
xmlReBlanksInsideTags = regexp.MustCompile(`>([^<]*[\n\r\t]| {3,})[^<]*<`) // regexp xmlReBlanksInsideTags matches the content of an XML tags that contains repeated spaces or \r\n\t
xmlReBlanks = regexp.MustCompile(`\s{2,}`) // regexp xmlReBlanks matches the content of an XML tag that contains more than two spaces together
)
type xmlFormatterImpl struct{}
// format implements basic formatting for XML input by applying indentation, normalizing whitespace, handling
// comments and XML tags inside comments.
func (xmlFormatterImpl) format(input []byte, output io.Writer, indent string) {
if nil == input || len(input) < 2 {
return
}
var needsUnescape bool
// Comments might contain further XML code. In that's the case, we want to escape
// that code to avoid formatting.
out := xmlReXMLInsideComment.ReplaceAllFunc(input, func(comment []byte) []byte {
needsUnescape = true
submatches := xmlReComment.FindSubmatch(comment)
b := bytes.Buffer{}
b.Grow(len(comment))
b.Write(submatches[1]) // <!--
b.WriteString(html.EscapeString(string(submatches[2]))) // ... (which includes XML code)
b.Write(submatches[3]) // -->
return b.Bytes()
})
out = xmlReBlanksAround.ReplaceAll(out, []byte("$1"))
out = xmlReBlanksInsideTags.ReplaceAllFunc(out, func(m []byte) []byte {
return xmlReBlanks.ReplaceAll(m, []byte(" "))
})
out = xmlReTag.ReplaceAllFunc(out, xmlTagReplacer(indent))
if needsUnescape {
// restore the original comment escaped content
out = xmlReComment.ReplaceAllFunc(out, func(comment []byte) []byte {
submatches := xmlReComment.FindSubmatch(comment)
b := bytes.Buffer{}
b.Grow(len(comment))
b.Write(submatches[1]) // <!--
b.WriteString(html.UnescapeString(string(submatches[2]))) // ... (which include XML code)
b.Write(submatches[3]) // -->
return b.Bytes()
})
}
_, err := output.Write(out[1:])
if nil != err {
slog.Error(err.Error())
}
}
// xmlTagReplacer returns a closure that processes XML tag slices, applying indentation and
// whitespace normalization based on the type of XML tag provided.
func xmlTagReplacer(indent string) func([]byte) []byte {
var (
depth = 0
wasEndOfTag = true
buffer = bytes.Buffer{}
)
buffer.Grow(64)
return func(tag []byte) []byte {
// XML declaration
if bytes.HasPrefix(tag, []byte("<?xml")) {
return append([]byte("\n"), xmlReBlanks.ReplaceAll(tag, []byte(" "))...)
}
defer buffer.Reset()
buffer.Write([]byte("\n"))
switch {
default: // start element
wasEndOfTag = false
buffer.WriteString(strings.Repeat(indent, depth))
buffer.Write(xmlReBlanks.ReplaceAll(tag, []byte(" ")))
depth++
return buffer.Bytes()
case bytes.HasSuffix(tag, []byte("/>")): // empty element
wasEndOfTag = true
buffer.WriteString(strings.Repeat(indent, depth))
buffer.Write(tag)
return buffer.Bytes()
case bytes.HasPrefix(tag, []byte("<!")): // comment or doctype
if bytes.HasPrefix(bytes.ToLower(tag), []byte("<!doctype")) {
tag = xmlReBlanks.ReplaceAll(tag, []byte(" "))
}
wasEndOfTag = true
buffer.WriteString(strings.Repeat(indent, depth))
buffer.Write(tag)
return buffer.Bytes()
case bytes.HasPrefix(tag, []byte("</")): // end element
depth--
if wasEndOfTag {
buffer.WriteString(strings.Repeat(indent, depth))
buffer.Write(tag)
return buffer.Bytes()
}
wasEndOfTag = true
return tag
}
}
}
type htmlFormatterImpl struct{}
var (
// htmlReStartingBlankBeforeText matches a string that starts with leading blank followed by a non-blank character
htmlReStartingBlankBeforeText = regexp.MustCompile(`^\s+\S`)
// htmlReTrailingBlankAfterText matches a string that ends with a non-blank character followed by trailing blank
htmlReTrailingBlankAfterText = regexp.MustCompile(`\S\s+$`)
// inlineTags is a dictionary that contains tags that typically do not start on a new line and only occupy the space
// required by their content.
inlineTags = map[string]struct{}{
"a": {}, "b": {}, "i": {}, "em": {}, "strong": {}, "code": {}, "span": {},
"ins": {}, "big": {}, "small": {}, "tt": {}, "abbr": {}, "acronym": {},
"cite": {}, "dfn": {}, "kbd": {}, "samp": {}, "var": {}, "bdo": {},
"map": {}, "q": {}, "sub": {}, "sup": {},
}
// selfClosingTags is a dictionary that contains tags that do not have a closing counterpart.
selfClosingTags = map[string]struct{}{
"input": {}, "link": {}, "meta": {}, "hr": {}, "img": {}, "br": {}, "area": {},
"base": {}, "col": {}, "param": {}, "command": {}, "embed": {}, "keygen": {},
"source": {}, "track": {}, "wbr": {},
}
)
// isInlineTag checks if the given HTML tag is an inline tag.
func isInlineTag(tag []byte) bool {
_, found := inlineTags[string(tag)]
return found
}
// isSelfClosingTag checks if the given HTML tag is a self-closing tag.
func isSelfClosingTag(tag []byte) bool {
_, found := selfClosingTags[string(tag)]
return found
}
// adjustTextIndentation adjusts the indentation of a block of text. It first determines the minimum indentation level
// (leading spaces) present in the non-empty lines of the text; then it removes this common indentation and re-indents
// the text with a specified depth of spaces.
func adjustTextIndentation(text []byte, depth int) []byte {
minIndent := 1000
for _, line := range bytes.Split(text, []byte{'\n'}) {
if trimmed := bytes.TrimLeft(line, " "); len(trimmed) > 0 {
indent := len(line) - len(trimmed)
if indent < minIndent {
minIndent = indent
}
}
}
if minIndent == 1000 {
minIndent = 0
}
re := regexp.MustCompile(fmt.Sprintf(`\n\s{%d}`, minIndent))
return re.ReplaceAllLiteral(text, append([]byte{'\n'}, bytes.Repeat([]byte(" "), depth)...))
}
// format formats the HTML input with proper indentation.
func (htmlFormatterImpl) format(input []byte, output io.Writer, indent string) {
var (
reader = bytes.NewReader(input)
tokenizer = xhtml.NewTokenizer(reader)
depth = 0
prevType = xhtml.ErrorToken
tagName []byte
prvName []byte
longText = false
skipFirstNewline = true
)
writeIndent := func(depth int) {
if skipFirstNewline {
skipFirstNewline = false
} else {
output.Write([]byte("\n"))
}
output.Write(bytes.Repeat([]byte(indent), depth))
}
for {
tokenType := tokenizer.Next()
if tokenType != xhtml.TextToken {
prvName = tagName
tagName, _ = tokenizer.TagName()
}
switch tokenType {
case xhtml.ErrorToken:
if errors.Is(tokenizer.Err(), io.EOF) {
return
}
slog.Error(tokenizer.Err().Error())
return
case xhtml.StartTagToken:
if !(isInlineTag(tagName) && prevType == xhtml.TextToken) {
writeIndent(depth)
}
output.Write(tokenizer.Raw())
if !isSelfClosingTag(tagName) {
depth++
}
case xhtml.SelfClosingTagToken, xhtml.CommentToken, xhtml.DoctypeToken:
writeIndent(depth)
output.Write(tokenizer.Raw())
case xhtml.EndTagToken:
if depth > 0 {
depth--
}
if !bytes.Equal(prvName, tagName) || prevType == xhtml.SelfClosingTagToken || prevType == xhtml.CommentToken ||
prevType == xhtml.DoctypeToken || (prevType == xhtml.TextToken && longText) {
writeIndent(depth)
}
output.Write(tokenizer.Raw())
case xhtml.TextToken:
t := bytes.Replace(tokenizer.Raw(), []byte{'\t'}, []byte(indent), -1)
text := bytes.Trim(t, "\n\r ")
if htmlReTrailingBlankAfterText.Match(t) {
text = append(text, ' ')
}
longText = false
if len(text) > 0 {
if bytes.Contains(text, []byte{'\n'}) {
if !(prevType == xhtml.EndTagToken && isInlineTag(tagName)) {
writeIndent(depth)
} else if htmlReStartingBlankBeforeText.Match(t) {
text = append([]byte{' '}, text...)
}
output.Write(adjustTextIndentation(text, depth))
longText = true
} else {
if utf8.RuneCount(text) > 80 || prevType != xhtml.StartTagToken {
if !(prevType == xhtml.EndTagToken && isInlineTag(tagName)) {
writeIndent(depth)
longText = true
} else if htmlReStartingBlankBeforeText.Match(t) {
text = append([]byte{' '}, text...)
}
}
output.Write(text)
}
}
}
prevType = tokenType
}
}