summaryrefslogtreecommitdiff
path: root/packages/backend/src/mfm/from-html.ts
blob: 7751bac5639ed0f4a7a1b3ee8b0ab1285d37dd5a (plain)
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
import { URL } from 'node:url';
import * as parse5 from 'parse5';
import * as TreeAdapter from '../../node_modules/parse5/dist/tree-adapters/default.js';

const treeAdapter = TreeAdapter.defaultTreeAdapter;

const urlRegex = /^https?:\/\/[\w\/:%#@$&?!()\[\]~.,=+\-]+/;
const urlRegexFull = /^https?:\/\/[\w\/:%#@$&?!()\[\]~.,=+\-]+$/;

export function fromHtml(html: string, hashtagNames?: string[]): string {
	// some AP servers like Pixelfed use br tags as well as newlines
	html = html.replace(/<br\s?\/?>\r?\n/gi, '\n');

	const dom = parse5.parseFragment(html);

	let text = '';

	for (const n of dom.childNodes) {
		analyze(n);
	}

	return text.trim();

	function getText(node: TreeAdapter.Node): string {
		if (treeAdapter.isTextNode(node)) return node.value;
		if (!treeAdapter.isElementNode(node)) return '';
		if (node.nodeName === 'br') return '\n';

		if (node.childNodes) {
			return node.childNodes.map(n => getText(n)).join('');
		}

		return '';
	}

	function appendChildren(childNodes: TreeAdapter.ChildNode[]): void {
		if (childNodes) {
			for (const n of childNodes) {
				analyze(n);
			}
		}
	}

	function analyze(node: TreeAdapter.Node) {
		if (treeAdapter.isTextNode(node)) {
			text += node.value;
			return;
		}

		// Skip comment or document type node
		if (!treeAdapter.isElementNode(node)) return;

		switch (node.nodeName) {
			case 'br': {
				text += '\n';
				break;
			}

			case 'a':
			{
				const txt = getText(node);
				const rel = node.attrs.find(x => x.name === 'rel');
				const href = node.attrs.find(x => x.name === 'href');

				// ハッシュタグ
				if (hashtagNames && href && hashtagNames.map(x => x.toLowerCase()).includes(txt.toLowerCase())) {
					text += txt;
				// メンション
				} else if (txt.startsWith('@') && !(rel && rel.value.match(/^me /))) {
					const part = txt.split('@');

					if (part.length === 2 && href) {
						//#region ホスト名部分が省略されているので復元する
						const acct = `${txt}@${(new URL(href.value)).hostname}`;
						text += acct;
						//#endregion
					} else if (part.length === 3) {
						text += txt;
					}
				// その他
				} else {
					const generateLink = () => {
						if (!href && !txt) {
							return '';
						}
						if (!href) {
							return txt;
						}
						if (!txt || txt === href.value) {	// #6383: Missing text node
							if (href.value.match(urlRegexFull)) {
								return href.value;
							} else {
								return `<${href.value}>`;
							}
						}
						if (href.value.match(urlRegex) && !href.value.match(urlRegexFull)) {
							return `[${txt}](<${href.value}>)`;	// #6846
						} else {
							return `[${txt}](${href.value})`;
						}
					};

					text += generateLink();
				}
				break;
			}

			case 'h1':
			{
				text += '【';
				appendChildren(node.childNodes);
				text += '】\n';
				break;
			}

			case 'b':
			case 'strong':
			{
				text += '**';
				appendChildren(node.childNodes);
				text += '**';
				break;
			}

			case 'small':
			{
				text += '<small>';
				appendChildren(node.childNodes);
				text += '</small>';
				break;
			}

			case 's':
			case 'del':
			{
				text += '~~';
				appendChildren(node.childNodes);
				text += '~~';
				break;
			}

			case 'i':
			case 'em':
			{
				text += '<i>';
				appendChildren(node.childNodes);
				text += '</i>';
				break;
			}

			// block code (<pre><code>)
			case 'pre': {
				if (node.childNodes.length === 1 && node.childNodes[0].nodeName === 'code') {
					text += '\n```\n';
					text += getText(node.childNodes[0]);
					text += '\n```\n';
				} else {
					appendChildren(node.childNodes);
				}
				break;
			}

			// inline code (<code>)
			case 'code': {
				text += '`';
				appendChildren(node.childNodes);
				text += '`';
				break;
			}

			case 'blockquote': {
				const t = getText(node);
				if (t) {
					text += '\n> ';
					text += t.split('\n').join('\n> ');
				}
				break;
			}

			case 'p':
			case 'h2':
			case 'h3':
			case 'h4':
			case 'h5':
			case 'h6':
			{
				text += '\n\n';
				appendChildren(node.childNodes);
				break;
			}

			// other block elements
			case 'div':
			case 'header':
			case 'footer':
			case 'article':
			case 'li':
			case 'dt':
			case 'dd':
			{
				text += '\n';
				appendChildren(node.childNodes);
				break;
			}

			default:	// includes inline elements
			{
				appendChildren(node.childNodes);
				break;
			}
		}
	}
}