mizzkey/src/mfm/language.ts
2019-04-18 00:40:56 +09:00

191 lines
7.8 KiB
TypeScript
Raw Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

import * as P from 'parsimmon';
import { createLeaf, createTree, urlRegex } from './prelude';
import { takeWhile, cumulativeSum } from '../prelude/array';
import parseAcct from '../misc/acct/parse';
import { toUnicode } from 'punycode';
import { emojiRegex } from '../misc/emoji-regex';
export function removeOrphanedBrackets(s: string): string {
const openBrackets = ['(', '「'];
const closeBrackets = [')', '」'];
const xs = cumulativeSum(s.split('').map(c => {
if (openBrackets.includes(c)) return 1;
if (closeBrackets.includes(c)) return -1;
return 0;
}));
const firstOrphanedCloseBracket = xs.findIndex(x => x < 0);
if (firstOrphanedCloseBracket !== -1) return s.substr(0, firstOrphanedCloseBracket);
const lastMatched = xs.lastIndexOf(0);
return s.substr(0, lastMatched + 1);
}
export const mfmLanguage = P.createLanguage({
root: r => P.alt(r.block, r.inline).atLeast(1),
plain: r => P.alt(r.emoji, r.text).atLeast(1),
block: r => P.alt(
r.title,
r.quote,
r.search,
r.blockCode,
r.mathBlock,
r.center,
),
startOfLine: () => P((input, i) => {
if (i == 0 || input[i] == '\n' || input[i - 1] == '\n') {
return P.makeSuccess(i, null);
} else {
return P.makeFailure(i, 'not newline');
}
}),
title: r => r.startOfLine.then(P((input, i) => {
const text = input.substr(i);
const match = text.match(/^([【\[]([^【\[】\]\n]+?)[】\]])(\n|$)/);
if (!match) return P.makeFailure(i, 'not a title');
const q = match[2].trim();
const contents = r.inline.atLeast(1).tryParse(q);
return P.makeSuccess(i + match[0].length, createTree('title', contents, {}));
})),
quote: r => r.startOfLine.then(P((input, i) => {
const text = input.substr(i);
if (!text.match(/^>[\s\S]+?/)) return P.makeFailure(i, 'not a quote');
const quote = takeWhile(line => line.startsWith('>'), text.split('\n'));
const qInner = quote.join('\n').replace(/^>/gm, '').replace(/^ /gm, '');
if (qInner == '') return P.makeFailure(i, 'not a quote');
const contents = r.root.tryParse(qInner);
return P.makeSuccess(i + quote.join('\n').length + 1, createTree('quote', contents, {}));
})),
search: r => r.startOfLine.then(P((input, i) => {
const text = input.substr(i);
const match = text.match(/^(.+?)( | )(検索|\[検索\]|Search|\[Search\])(\n|$)/i);
if (!match) return P.makeFailure(i, 'not a search');
return P.makeSuccess(i + match[0].length, createLeaf('search', { query: match[1], content: match[0].trim() }));
})),
blockCode: r => r.startOfLine.then(P((input, i) => {
const text = input.substr(i);
const match = text.match(/^```(.+?)?\n([\s\S]+?)\n```(\n|$)/i);
if (!match) return P.makeFailure(i, 'not a blockCode');
return P.makeSuccess(i + match[0].length, createLeaf('blockCode', { code: match[2], lang: match[1] ? match[1].trim() : null }));
})),
inline: r => P.alt(
r.big,
r.bold,
r.small,
r.italic,
r.strike,
r.motion,
r.spin,
r.jump,
r.flip,
r.inlineCode,
r.mathInline,
r.mention,
r.hashtag,
r.url,
r.link,
r.emoji,
r.text
),
big: r => P.regexp(/^\*\*\*([\s\S]+?)\*\*\*/, 1).map(x => createTree('big', r.inline.atLeast(1).tryParse(x), {})),
bold: r => {
const asterisk = P.regexp(/\*\*([\s\S]+?)\*\*/, 1);
const underscore = P.regexp(/__([a-zA-Z0-9\s]+?)__/, 1);
return P.alt(asterisk, underscore).map(x => createTree('bold', r.inline.atLeast(1).tryParse(x), {}));
},
small: r => P.regexp(/<small>([\s\S]+?)<\/small>/, 1).map(x => createTree('small', r.inline.atLeast(1).tryParse(x), {})),
italic: r => {
const xml = P.regexp(/<i>([\s\S]+?)<\/i>/, 1);
const underscore = P((input, i) => {
const text = input.substr(i);
const match = text.match(/^(\*|_)([a-zA-Z0-9]+?[\s\S]*?)\1/);
if (!match) return P.makeFailure(i, 'not a italic');
if (input[i - 1] != null && input[i - 1].match(/[a-z0-9]/i)) return P.makeFailure(i, 'not a italic');
return P.makeSuccess(i + match[0].length, match[2]);
});
return P.alt(xml, underscore).map(x => createTree('italic', r.inline.atLeast(1).tryParse(x), {}));
},
strike: r => P.regexp(/~~(.+?)~~/, 1).map(x => createTree('strike', r.inline.atLeast(1).tryParse(x), {})),
motion: r => {
const paren = P.regexp(/\(\(\(([\s\S]+?)\)\)\)/, 1);
const xml = P.regexp(/<motion>(.+?)<\/motion>/, 1);
return P.alt(paren, xml).map(x => createTree('motion', r.inline.atLeast(1).tryParse(x), {}));
},
spin: r => {
return P((input, i) => {
const text = input.substr(i);
const match = text.match(/^<spin(\s[a-z]+?)?>(.+?)<\/spin>/i);
if (!match) return P.makeFailure(i, 'not a spin');
return P.makeSuccess(i + match[0].length, {
content: match[2], attr: match[1] ? match[1].trim() : null
});
}).map(x => createTree('spin', r.inline.atLeast(1).tryParse(x.content), { attr: x.attr }));
},
jump: r => P.regexp(/<jump>(.+?)<\/jump>/, 1).map(x => createTree('jump', r.inline.atLeast(1).tryParse(x), {})),
flip: r => P.regexp(/<flip>(.+?)<\/flip>/, 1).map(x => createTree('flip', r.inline.atLeast(1).tryParse(x), {})),
center: r => r.startOfLine.then(P.regexp(/<center>([\s\S]+?)<\/center>/, 1).map(x => createTree('center', r.inline.atLeast(1).tryParse(x), {}))),
inlineCode: () => P.regexp(/`([^´\n]+?)`/, 1).map(x => createLeaf('inlineCode', { code: x })),
mathBlock: r => r.startOfLine.then(P.regexp(/\\\[([\s\S]+?)\\\]/, 1).map(x => createLeaf('mathBlock', { formula: x.trim() }))),
mathInline: () => P.regexp(/\\\((.+?)\\\)/, 1).map(x => createLeaf('mathInline', { formula: x })),
mention: () => {
return P((input, i) => {
const text = input.substr(i);
const match = text.match(/^@\w([\w-]*\w)?(?:@[\w\.\-]+\w)?/);
if (!match) return P.makeFailure(i, 'not a mention');
if (input[i - 1] != null && input[i - 1].match(/[a-z0-9]/i)) return P.makeFailure(i, 'not a mention');
return P.makeSuccess(i + match[0].length, match[0]);
}).map(x => {
const { username, host } = parseAcct(x.substr(1));
const canonical = host != null ? `@${username}@${toUnicode(host)}` : x;
return createLeaf('mention', { canonical, username, host, acct: x });
});
},
hashtag: () => P((input, i) => {
const text = input.substr(i);
const match = text.match(/^#([^\s\.,!\?'"#:\/\[\]【】]+)/i);
if (!match) return P.makeFailure(i, 'not a hashtag');
let hashtag = match[1];
hashtag = removeOrphanedBrackets(hashtag);
if (hashtag.match(/^[0-9]+$/)) return P.makeFailure(i, 'not a hashtag');
if (input[i - 1] != null && input[i - 1].match(/[a-z0-9]/i)) return P.makeFailure(i, 'not a hashtag');
if (hashtag.length > 50) return P.makeFailure(i, 'not a hashtag');
return P.makeSuccess(i + ('#' + hashtag).length, createLeaf('hashtag', { hashtag: hashtag }));
}),
url: () => {
return P((input, i) => {
const text = input.substr(i);
const match = text.match(urlRegex);
let url: string;
if (!match) {
const match = text.match(/^<(https?:\/\/.*?)>/);
if (!match)
return P.makeFailure(i, 'not a url');
url = match[1];
i += 2;
} else
url = match[0];
url = removeOrphanedBrackets(url);
if (url.endsWith('.')) url = url.substr(0, url.lastIndexOf('.'));
if (url.endsWith(',')) url = url.substr(0, url.lastIndexOf(','));
return P.makeSuccess(i + url.length, url);
}).map(x => createLeaf('url', { url: x }));
},
link: r => {
return P.seqObj(
['silent', P.string('?').fallback(null).map(x => x != null)] as any,
P.string('['), ['text', P.regexp(/[^\n\[\]]+/)] as any, P.string(']'),
P.string('('), ['url', r.url] as any, P.string(')'),
).map((x: any) => {
return createTree('link', r.inline.atLeast(1).tryParse(x.text), {
silent: x.silent,
url: x.url.node.props.url
});
});
},
emoji: () => {
const name = P.regexp(/:([a-z0-9_+-]+):/i, 1).map(x => createLeaf('emoji', { name: x }));
const code = P.regexp(emojiRegex).map(x => createLeaf('emoji', { emoji: x }));
return P.alt(name, code);
},
text: () => P.any.map(x => createLeaf('text', { text: x }))
});