markitzero.go (view raw)
1//
2// Copyright (c) 2019 Ted Unangst <tedu@tedunangst.com>
3//
4// Permission to use, copy, modify, and distribute this software for any
5// purpose with or without fee is hereby granted, provided that the above
6// copyright notice and this permission notice appear in all copies.
7//
8// THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
9// WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
10// MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
11// ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
12// WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
13// ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
14// OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
15
16package main
17
18import (
19 "fmt"
20 "regexp"
21 "strings"
22
23 "golang.org/x/net/html"
24 "humungus.tedunangst.com/r/webs/synlight"
25)
26
27var re_bolder = regexp.MustCompile(`(^|\W)\*\*((?s:.*?))\*\*($|\W)`)
28var re_italicer = regexp.MustCompile(`(^|\W)\*((?s:.*?))\*($|\W)`)
29var re_bigcoder = regexp.MustCompile("```(.*)\n?((?s:.*?))\n?```\n?")
30var re_coder = regexp.MustCompile("`([^`]*)`")
31var re_quoter = regexp.MustCompile(`(?m:^> (.*)(\n- ?(.*))?\n?)`)
32var re_reciter = regexp.MustCompile(`(<cite><a href=".*?">)https://twitter.com/([^/]+)/.*?(</a></cite>)`)
33var re_link = regexp.MustCompile(`.?.?https?://[^\s"]+[\w/)!]`)
34var re_zerolink = regexp.MustCompile(`\[([^]]*)\]\(([^)]*\)?)\)`)
35var re_imgfix = regexp.MustCompile(`<img ([^>]*)>`)
36var re_lister = regexp.MustCompile(`((^|\n)(\+|-).*)+\n?`)
37var re_tabler = regexp.MustCompile(`((^|\n)\|.*)+\n?`)
38var re_header = regexp.MustCompile(`(^|\n)(#+) (.*)`)
39
40var lighter = synlight.New(synlight.Options{Format: synlight.HTML})
41
42func markitzero(s string) string {
43 // prepare the string
44 s = strings.TrimSpace(s)
45 s = strings.Replace(s, "\r", "", -1)
46
47 codeword := "`elided big code`"
48
49 // save away the code blocks so we don't mess them up further
50 var bigcodes, lilcodes, images []string
51 s = re_bigcoder.ReplaceAllStringFunc(s, func(code string) string {
52 bigcodes = append(bigcodes, code)
53 return codeword
54 })
55 s = re_coder.ReplaceAllStringFunc(s, func(code string) string {
56 lilcodes = append(lilcodes, code)
57 return "`x`"
58 })
59 s = re_imgfix.ReplaceAllStringFunc(s, func(img string) string {
60 images = append(images, img)
61 return "<img x>"
62 })
63
64 // fewer side effects than html.EscapeString
65 buf := make([]byte, 0, len(s))
66 for _, c := range []byte(s) {
67 switch c {
68 case '&':
69 buf = append(buf, []byte("&")...)
70 case '<':
71 buf = append(buf, []byte("<")...)
72 case '>':
73 buf = append(buf, []byte(">")...)
74 default:
75 buf = append(buf, c)
76 }
77 }
78 s = string(buf)
79
80 // mark it zero
81 if strings.Contains(s, "http") {
82 s = re_link.ReplaceAllStringFunc(s, linkreplacer)
83 }
84 s = re_zerolink.ReplaceAllString(s, `<a href="$2">$1</a>`)
85 if strings.Contains(s, "*") {
86 s = re_bolder.ReplaceAllString(s, "$1<b>$2</b>$3")
87 s = re_italicer.ReplaceAllString(s, "$1<i>$2</i>$3")
88 }
89 s = re_quoter.ReplaceAllString(s, "<blockquote>$1<br><cite>$3</cite></blockquote><p>")
90 s = re_reciter.ReplaceAllString(s, "$1$2$3")
91 s = strings.Replace(s, "\n---\n", "<hr><p>", -1)
92
93 s = re_lister.ReplaceAllStringFunc(s, func(m string) string {
94 m = strings.Trim(m, "\n")
95 items := strings.Split(m, "\n")
96 r := "<ul>"
97 for _, item := range items {
98 r += "<li>" + strings.Trim(item[1:], " ")
99 }
100 r += "</ul><p>"
101 return r
102 })
103 s = re_tabler.ReplaceAllStringFunc(s, func(m string) string {
104 m = strings.Trim(m, "\n")
105 rows := strings.Split(m, "\n")
106 var r strings.Builder
107 r.WriteString("<table>")
108 alignments := make(map[int]string)
109 for _, row := range rows {
110 hastr := false
111 cells := strings.Split(row, "|")
112 for i, cell := range cells {
113 cell = strings.TrimSpace(cell)
114 if cell == "" && (i == 0 || i == len(cells)-1) {
115 continue
116 }
117 switch cell {
118 case ":---":
119 alignments[i] = ` style="text-align: left"`
120 continue
121 case ":---:":
122 alignments[i] = ` style="text-align: center"`
123 continue
124 case "---:":
125 alignments[i] = ` style="text-align: right"`
126 continue
127 }
128 if !hastr {
129 r.WriteString("<tr>")
130 hastr = true
131 }
132 fmt.Fprintf(&r, "<td%s>", alignments[i])
133 r.WriteString(cell)
134 }
135 }
136 r.WriteString("</table><p>")
137 return r.String()
138 })
139 s = re_header.ReplaceAllStringFunc(s, func(s string) string {
140 m := re_header.FindStringSubmatch(s)
141 num := len(m[2])
142 return fmt.Sprintf("<h%d>%s</h%d><p>", num, m[3], num)
143 })
144
145 // restore images
146 s = strings.Replace(s, "<img x>", "<img x>", -1)
147 s = re_imgfix.ReplaceAllStringFunc(s, func(string) string {
148 img := images[0]
149 images = images[1:]
150 return img
151 })
152
153 // now restore the code blocks
154 s = re_coder.ReplaceAllStringFunc(s, func(string) string {
155 code := lilcodes[0]
156 lilcodes = lilcodes[1:]
157 if code == codeword && len(bigcodes) > 0 {
158 code := bigcodes[0]
159 bigcodes = bigcodes[1:]
160 m := re_bigcoder.FindStringSubmatch(code)
161 return "<pre><code>" + lighter.HighlightString(m[2], m[1]) + "</code></pre><p>"
162 }
163 code = html.EscapeString(code)
164 return code
165 })
166
167 s = re_coder.ReplaceAllString(s, "<code>$1</code>")
168
169 // some final fixups
170 s = strings.Replace(s, "\n", "<br>", -1)
171 s = strings.Replace(s, "<br><blockquote>", "<blockquote>", -1)
172 s = strings.Replace(s, "<br><cite></cite>", "", -1)
173 s = strings.Replace(s, "<br><pre>", "<pre>", -1)
174 s = strings.Replace(s, "<br><ul>", "<ul>", -1)
175 s = strings.Replace(s, "<br><table>", "<table>", -1)
176 s = strings.Replace(s, "<p><br>", "<p>", -1)
177 return s
178}
179
180func linkreplacer(url string) string {
181 if url[0:2] == "](" {
182 return url
183 }
184 prefix := ""
185 for !strings.HasPrefix(url, "http") {
186 prefix += url[0:1]
187 url = url[1:]
188 }
189 addparen := false
190 adddot := false
191 if strings.HasSuffix(url, ")") && strings.IndexByte(url, '(') == -1 {
192 url = url[:len(url)-1]
193 addparen = true
194 }
195 if strings.HasSuffix(url, ".") {
196 url = url[:len(url)-1]
197 adddot = true
198 }
199 url = fmt.Sprintf(`<a href="%s">%s</a>`, url, url)
200 if adddot {
201 url += "."
202 }
203 if addparen {
204 url += ")"
205 }
206 return prefix + url
207}