autoid.go (3222B)
1 // Copyright 2019 The Hugo Authors. All rights reserved.
2 //
3 // Licensed under the Apache License, Version 2.0 (the "License");
4 // you may not use this file except in compliance with the License.
5 // You may obtain a copy of the License at
6 // http://www.apache.org/licenses/LICENSE-2.0
7 //
8 // Unless required by applicable law or agreed to in writing, software
9 // distributed under the License is distributed on an "AS IS" BASIS,
10 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
11 // See the License for the specific language governing permissions and
12 // limitations under the License.
13
14 package goldmark
15
16 import (
17 "bytes"
18 "strconv"
19 "unicode"
20 "unicode/utf8"
21
22 "github.com/gohugoio/hugo/markup/blackfriday"
23
24 "github.com/gohugoio/hugo/markup/goldmark/goldmark_config"
25
26 "github.com/gohugoio/hugo/common/text"
27
28 "github.com/yuin/goldmark/ast"
29 "github.com/yuin/goldmark/parser"
30 "github.com/yuin/goldmark/util"
31
32 bp "github.com/gohugoio/hugo/bufferpool"
33 )
34
35 func sanitizeAnchorNameString(s string, idType string) string {
36 return string(sanitizeAnchorName([]byte(s), idType))
37 }
38
39 func sanitizeAnchorName(b []byte, idType string) []byte {
40 return sanitizeAnchorNameWithHook(b, idType, nil)
41 }
42
43 func sanitizeAnchorNameWithHook(b []byte, idType string, hook func(buf *bytes.Buffer)) []byte {
44 buf := bp.GetBuffer()
45
46 if idType == goldmark_config.AutoHeadingIDTypeBlackfriday {
47 // TODO(bep) make it more efficient.
48 buf.WriteString(blackfriday.SanitizedAnchorName(string(b)))
49 } else {
50 asciiOnly := idType == goldmark_config.AutoHeadingIDTypeGitHubAscii
51
52 if asciiOnly {
53 // Normalize it to preserve accents if possible.
54 b = text.RemoveAccents(b)
55 }
56
57 b = bytes.TrimSpace(b)
58
59 for len(b) > 0 {
60 r, size := utf8.DecodeRune(b)
61 switch {
62 case asciiOnly && size != 1:
63 case r == '-' || r == ' ':
64 buf.WriteRune('-')
65 case isAlphaNumeric(r):
66 buf.WriteRune(unicode.ToLower(r))
67 default:
68 }
69
70 b = b[size:]
71 }
72 }
73
74 if hook != nil {
75 hook(buf)
76 }
77
78 result := make([]byte, buf.Len())
79 copy(result, buf.Bytes())
80
81 bp.PutBuffer(buf)
82
83 return result
84 }
85
86 func isAlphaNumeric(r rune) bool {
87 return r == '_' || unicode.IsLetter(r) || unicode.IsDigit(r)
88 }
89
90 var _ parser.IDs = (*idFactory)(nil)
91
92 type idFactory struct {
93 idType string
94 vals map[string]struct{}
95 }
96
97 func newIDFactory(idType string) *idFactory {
98 return &idFactory{
99 vals: make(map[string]struct{}),
100 idType: idType,
101 }
102 }
103
104 func (ids *idFactory) Generate(value []byte, kind ast.NodeKind) []byte {
105 return sanitizeAnchorNameWithHook(value, ids.idType, func(buf *bytes.Buffer) {
106 if buf.Len() == 0 {
107 if kind == ast.KindHeading {
108 buf.WriteString("heading")
109 } else {
110 buf.WriteString("id")
111 }
112 }
113
114 if _, found := ids.vals[util.BytesToReadOnlyString(buf.Bytes())]; found {
115 // Append a hypen and a number, starting with 1.
116 buf.WriteRune('-')
117 pos := buf.Len()
118 for i := 1; ; i++ {
119 buf.WriteString(strconv.Itoa(i))
120 if _, found := ids.vals[util.BytesToReadOnlyString(buf.Bytes())]; !found {
121 break
122 }
123 buf.Truncate(pos)
124 }
125 }
126
127 ids.vals[buf.String()] = struct{}{}
128 })
129 }
130
131 func (ids *idFactory) Put(value []byte) {
132 ids.vals[util.BytesToReadOnlyString(value)] = struct{}{}
133 }