2021-02-06 04:44:46 -08:00
|
|
|
import * as parse5 from 'parse5';
|
|
|
|
import treeAdapter = require('parse5/lib/tree-adapters/default');
|
|
|
|
import { URL } from 'url';
|
2021-04-01 18:36:11 -07:00
|
|
|
|
|
|
|
const urlRegex = /^https?:\/\/[\w\/:%#@$&?!()\[\]~.,=+\-]+/;
|
|
|
|
const urlRegexFull = /^https?:\/\/[\w\/:%#@$&?!()\[\]~.,=+\-]+$/;
|
2018-06-20 09:21:57 -07:00
|
|
|
|
2020-04-03 06:51:38 -07:00
|
|
|
export function fromHtml(html: string, hashtagNames?: string[]): string {
|
2021-02-06 04:44:46 -08:00
|
|
|
const dom = parse5.parseFragment(html);
|
2018-06-20 09:21:57 -07:00
|
|
|
|
|
|
|
let text = '';
|
|
|
|
|
2018-12-11 03:36:55 -08:00
|
|
|
for (const n of dom.childNodes) {
|
|
|
|
analyze(n);
|
|
|
|
}
|
2018-06-20 09:21:57 -07:00
|
|
|
|
|
|
|
return text.trim();
|
|
|
|
|
2021-02-06 04:44:46 -08:00
|
|
|
function getText(node: parse5.Node): string {
|
|
|
|
if (treeAdapter.isTextNode(node)) return node.value;
|
|
|
|
if (!treeAdapter.isElementNode(node)) return '';
|
2018-06-20 09:21:57 -07:00
|
|
|
|
|
|
|
if (node.childNodes) {
|
2021-02-06 04:44:46 -08:00
|
|
|
return node.childNodes.map(n => getText(n)).join('');
|
2018-06-20 09:21:57 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
return '';
|
|
|
|
}
|
|
|
|
|
2021-02-06 04:44:46 -08:00
|
|
|
function analyze(node: parse5.Node) {
|
|
|
|
if (treeAdapter.isTextNode(node)) {
|
|
|
|
text += node.value;
|
|
|
|
return;
|
|
|
|
}
|
2018-06-20 09:21:57 -07:00
|
|
|
|
2021-02-06 04:44:46 -08:00
|
|
|
// Skip comment or document type node
|
|
|
|
if (!treeAdapter.isElementNode(node)) return;
|
|
|
|
|
|
|
|
switch (node.nodeName) {
|
2018-06-20 09:21:57 -07:00
|
|
|
case 'br':
|
|
|
|
text += '\n';
|
|
|
|
break;
|
|
|
|
|
|
|
|
case 'a':
|
|
|
|
const txt = getText(node);
|
2021-02-06 04:44:46 -08:00
|
|
|
const rel = node.attrs.find(x => x.name === 'rel');
|
|
|
|
const href = node.attrs.find(x => x.name === 'href');
|
2018-06-20 09:21:57 -07:00
|
|
|
|
2020-04-03 06:51:38 -07:00
|
|
|
// ハッシュタグ
|
|
|
|
if (hashtagNames && href && hashtagNames.map(x => x.toLowerCase()).includes(txt.toLowerCase())) {
|
|
|
|
text += txt;
|
2018-06-20 09:21:57 -07:00
|
|
|
// メンション
|
2018-12-11 18:47:07 -08:00
|
|
|
} else if (txt.startsWith('@') && !(rel && rel.value.match(/^me /))) {
|
2018-06-20 09:21:57 -07:00
|
|
|
const part = txt.split('@');
|
|
|
|
|
2021-02-06 04:44:46 -08:00
|
|
|
if (part.length === 2 && href) {
|
2018-06-20 09:21:57 -07:00
|
|
|
//#region ホスト名部分が省略されているので復元する
|
2018-09-01 07:12:51 -07:00
|
|
|
const acct = `${txt}@${(new URL(href.value)).hostname}`;
|
2018-06-20 09:21:57 -07:00
|
|
|
text += acct;
|
|
|
|
//#endregion
|
2020-04-03 16:46:54 -07:00
|
|
|
} else if (part.length === 3) {
|
2018-06-20 09:21:57 -07:00
|
|
|
text += txt;
|
|
|
|
}
|
2018-09-01 06:45:27 -07:00
|
|
|
// その他
|
|
|
|
} else {
|
2021-02-06 04:44:46 -08:00
|
|
|
const generateLink = () => {
|
|
|
|
if (!href && !txt) {
|
|
|
|
return '';
|
|
|
|
}
|
|
|
|
if (!href) {
|
|
|
|
return txt;
|
|
|
|
}
|
|
|
|
if (!txt || txt === href.value) { // #6383: Missing text node
|
|
|
|
if (href.value.match(urlRegexFull)) {
|
|
|
|
return href.value;
|
|
|
|
} else {
|
|
|
|
return `<${href.value}>`;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (href.value.match(urlRegex) && !href.value.match(urlRegexFull)) {
|
|
|
|
return `[${txt}](<${href.value}>)`; // #6846
|
|
|
|
} else {
|
|
|
|
return `[${txt}](${href.value})`;
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
text += generateLink();
|
2018-06-20 09:21:57 -07:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case 'p':
|
|
|
|
text += '\n\n';
|
|
|
|
if (node.childNodes) {
|
2018-12-11 03:36:55 -08:00
|
|
|
for (const n of node.childNodes) {
|
|
|
|
analyze(n);
|
|
|
|
}
|
2018-06-20 09:21:57 -07:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
default:
|
|
|
|
if (node.childNodes) {
|
2018-12-11 03:36:55 -08:00
|
|
|
for (const n of node.childNodes) {
|
|
|
|
analyze(n);
|
|
|
|
}
|
2018-06-20 09:21:57 -07:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|