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?`)
38
39var lighter = synlight.New(synlight.Options{Format: synlight.HTML})
40
41func markitzero(s string) string {
42 // prepare the string
43 s = strings.TrimSpace(s)
44 s = strings.Replace(s, "\r", "", -1)
45
46 codeword := "`elided big code`"
47
48 // save away the code blocks so we don't mess them up further
49 var bigcodes, lilcodes, images []string
50 s = re_bigcoder.ReplaceAllStringFunc(s, func(code string) string {
51 bigcodes = append(bigcodes, code)
52 return codeword
53 })
54 s = re_coder.ReplaceAllStringFunc(s, func(code string) string {
55 lilcodes = append(lilcodes, code)
56 return "`x`"
57 })
58 s = re_imgfix.ReplaceAllStringFunc(s, func(img string) string {
59 images = append(images, img)
60 return "<img x>"
61 })
62
63 // fewer side effects than html.EscapeString
64 buf := make([]byte, 0, len(s))
65 for _, c := range []byte(s) {
66 switch c {
67 case '&':
68 buf = append(buf, []byte("&")...)
69 case '<':
70 buf = append(buf, []byte("<")...)
71 case '>':
72 buf = append(buf, []byte(">")...)
73 default:
74 buf = append(buf, c)
75 }
76 }
77 s = string(buf)
78
79 // mark it zero
80 if strings.Contains(s, "http") {
81 s = re_link.ReplaceAllStringFunc(s, linkreplacer)
82 }
83 s = re_zerolink.ReplaceAllString(s, `<a href="$2">$1</a>`)
84 if strings.Contains(s, "*") {
85 s = re_bolder.ReplaceAllString(s, "$1<b>$2</b>$3")
86 s = re_italicer.ReplaceAllString(s, "$1<i>$2</i>$3")
87 }
88 s = re_quoter.ReplaceAllString(s, "<blockquote>$1<br><cite>$3</cite></blockquote><p>")
89 s = re_reciter.ReplaceAllString(s, "$1$2$3")
90 s = strings.Replace(s, "\n---\n", "<hr><p>", -1)
91
92 s = re_lister.ReplaceAllStringFunc(s, func(m string) string {
93 m = strings.Trim(m, "\n")
94 items := strings.Split(m, "\n")
95 r := "<ul>"
96 for _, item := range items {
97 r += "<li>" + strings.Trim(item[1:], " ")
98 }
99 r += "</ul><p>"
100 return r
101 })
102 s = re_tabler.ReplaceAllStringFunc(s, func(m string) string {
103 m = strings.Trim(m, "\n")
104 rows := strings.Split(m, "\n")
105 var r strings.Builder
106 r.WriteString("<table>")
107 alignments := make(map[int]string)
108 for _, row := range rows {
109 hastr := false
110 cells := strings.Split(row, "|")
111 for i, cell := range cells {
112 cell = strings.TrimSpace(cell)
113 if cell == "" && (i == 0 || i == len(cells)-1) {
114 continue
115 }
116 switch cell {
117 case ":---":
118 alignments[i] = ` style="text-align: left"`
119 continue
120 case ":---:":
121 alignments[i] = ` style="text-align: center"`
122 continue
123 case "---:":
124 alignments[i] = ` style="text-align: right"`
125 continue
126 }
127 if !hastr {
128 r.WriteString("<tr>")
129 hastr = true
130 }
131 fmt.Fprintf(&r, "<td%s>", alignments[i])
132 r.WriteString(cell)
133 }
134 }
135 r.WriteString("</table><p>")
136 return r.String()
137 })
138
139 // restore images
140 s = strings.Replace(s, "<img x>", "<img x>", -1)
141 s = re_imgfix.ReplaceAllStringFunc(s, func(string) string {
142 img := images[0]
143 images = images[1:]
144 return img
145 })
146
147 // now restore the code blocks
148 s = re_coder.ReplaceAllStringFunc(s, func(string) string {
149 code := lilcodes[0]
150 lilcodes = lilcodes[1:]
151 if code == codeword && len(bigcodes) > 0 {
152 code := bigcodes[0]
153 bigcodes = bigcodes[1:]
154 m := re_bigcoder.FindStringSubmatch(code)
155 return "<pre><code>" + lighter.HighlightString(m[2], m[1]) + "</code></pre><p>"
156 }
157 code = html.EscapeString(code)
158 return code
159 })
160
161 s = re_coder.ReplaceAllString(s, "<code>$1</code>")
162
163 // some final fixups
164 s = strings.Replace(s, "\n", "<br>", -1)
165 s = strings.Replace(s, "<br><blockquote>", "<blockquote>", -1)
166 s = strings.Replace(s, "<br><cite></cite>", "", -1)
167 s = strings.Replace(s, "<br><pre>", "<pre>", -1)
168 s = strings.Replace(s, "<br><ul>", "<ul>", -1)
169 s = strings.Replace(s, "<br><table>", "<table>", -1)
170 s = strings.Replace(s, "<p><br>", "<p>", -1)
171 return s
172}
173
174func linkreplacer(url string) string {
175 if url[0:2] == "](" {
176 return url
177 }
178 prefix := ""
179 for !strings.HasPrefix(url, "http") {
180 prefix += url[0:1]
181 url = url[1:]
182 }
183 addparen := false
184 adddot := false
185 if strings.HasSuffix(url, ")") && strings.IndexByte(url, '(') == -1 {
186 url = url[:len(url)-1]
187 addparen = true
188 }
189 if strings.HasSuffix(url, ".") {
190 url = url[:len(url)-1]
191 adddot = true
192 }
193 url = fmt.Sprintf(`<a href="%s">%s</a>`, url, url)
194 if adddot {
195 url += "."
196 }
197 if addparen {
198 url += ")"
199 }
200 return prefix + url
201}