very-small-parser
Version:
A very small Markdown, HTML, and CSS parser.
250 lines (249 loc) • 8.91 kB
JavaScript
;
Object.defineProperty(exports, "__esModule", { value: true });
exports.parsers = void 0;
const tslib_1 = require("tslib");
const util_1 = require("../../util");
const reg = tslib_1.__importStar(require("../regex"));
const html_1 = require("../../html");
const REG_NEWLINE = /^[\n\r]+/;
const newline = (_, src) => {
const matches = src.match(REG_NEWLINE);
if (matches)
return (0, util_1.token)(matches[0], '');
};
const REG_CODE = /^(\s{4}[^\n]+)+/;
const code = (_, src) => {
const matches = src.match(REG_CODE);
if (!matches)
return;
const subvalue = matches[0];
const overrides = {
value: (0, util_1.rep)(/\n+$/, '', (0, util_1.rep)(/^ {4}/gm, '', subvalue)),
lang: null,
};
return (0, util_1.token)(subvalue, 'code', void 0, overrides, subvalue.length);
};
const REG_FENCES = /^ *(`{3,}|~{3,})([^\s]*) *([^\n]*)\n([\s\S]*?)\s*\1 *(?:\n+|$)/;
const fences = (_, src) => {
const matches = src.match(REG_FENCES);
if (!matches)
return;
const subvalue = matches[0];
const overrides = {
value: matches[4] || matches[3],
lang: matches[2] || '',
meta: matches.length > 4 ? matches[3] : null,
};
return (0, util_1.token)(subvalue, 'code', void 0, overrides);
};
const REG_MATH = /^ *\$\$[ \.]*(\S+)? *\n([\s\S]*?)\s*\$\$ *(?:\n+|$)/;
const math = (_, src) => {
const matches = src.match(REG_MATH);
if (matches)
return (0, util_1.token)(matches[0], 'math', void 0, { value: matches[2] });
};
const thematicBreak = (_, src) => {
const matches = src.match(reg.hr);
if (matches)
return (0, util_1.token)(matches[0], 'thematicBreak', void 0, { value: matches[1] });
};
const REG_HEADING1 = /^ *(#{1,6}) +([^\n]+?) *(?:#+ *)?(?:\n+|$)/;
const REG_HEADING2 = /^([^\n]+)\n *(=|-){2,} *(?:\n+|$)/;
const heading = (parser, src) => {
let matches = src.match(REG_HEADING1);
if (matches) {
const subvalue = matches[2];
return (0, util_1.token)(matches[0], 'heading', parser.parsei(subvalue), { depth: matches[1].length });
}
matches = src.match(REG_HEADING2);
if (matches) {
const subvalue = matches[1];
return (0, util_1.token)(matches[0], 'heading', parser.parsei(subvalue), {
depth: matches[2] === '-' ? 2 : 1,
});
}
};
const REG_BLOCKQUOTE = /^( *>[^\n]+(\n(?!^ *\[([^\]]+)\]: *<?([^\s>]+)>?(?: +["(]([^\n]+)[")])? *(?:\n+|$))[^\n]+)*)+/;
const REG_BLOCKQUOTE_OUTDENT = /^ *> ?/gm;
const REG_BLOCK_SPOILER_CHECK = /^ *!/;
const REG_BLOCK_SPOILER_OUTDENT = /^ *! ?/gm;
const blockquote = (parser, src) => {
const matches = src.match(REG_BLOCKQUOTE);
if (!matches)
return;
const subvalue = matches[0];
let content = (0, util_1.rep)(REG_BLOCKQUOTE_OUTDENT, '', subvalue);
let spoiler = false;
const isPotentiallySpoiler = content.match(REG_BLOCK_SPOILER_CHECK);
if (isPotentiallySpoiler) {
spoiler = content.split('\n').every((line) => line.match(REG_BLOCK_SPOILER_CHECK));
if (spoiler)
content = content.replace(REG_BLOCK_SPOILER_OUTDENT, '');
}
const children = parser.parse(content);
const node = (0, util_1.token)(subvalue, 'blockquote', children);
if (spoiler)
node.spoiler = spoiler;
return node;
};
const REG_BULLET = /^\s{0,3}([*+-]|\d{1,3}\.)\s{1,42}/;
const bull = /(?:[*+-]|\d{1,3}\.)/;
const REG_LOOSE = /\n\s*\n(?!\s*$)/;
const REG_ITEM = reg.replace(/^( {0,333})(bull) [^\n]*(?:\n(?!\1bull )[^\n]*)*/gm, { bull });
const REG_LIST = reg.replace(/^ {0,3}(?:bull) (?:(?!\r?\n\r?\n)[\s\S])+/, {
bull,
hr: reg.hr,
def: reg.def,
});
const getParts = (subvalue) => subvalue.match(REG_ITEM);
const list = (parser, value) => {
const matches = value.match(REG_LIST);
if (!matches)
return;
const subvalue = matches[0];
const parts = getParts(subvalue);
if (!parts)
return;
const length = parts.length;
const children = [];
let start = null;
for (let i = 0; i < length; i++) {
const part = parts[i];
// if (part[part.length - 1] === '\n') part = part.trimEnd();
const bulletMatch = part.match(REG_BULLET);
if (!bulletMatch)
return;
const [bulletWithWhitespace, bullet] = bulletMatch;
let content = part.slice(bulletWithWhitespace.length);
if (i === 0 && bullet.length > 1)
start = Number(bullet);
let checked = null;
if (content[0] === '[' && content[2] === ']' && content[3] === ' ') {
checked = content[1] !== ' ';
content = content.slice(4);
}
const newLinePos = content.indexOf('\n');
if (newLinePos > 0) {
let outdentSize = 0;
for (let pos = newLinePos + 1; pos < newLinePos + 4; pos++)
if (content[pos] === ' ')
outdentSize++;
else
break;
if (outdentSize)
content = (0, util_1.rep)(new RegExp('^ {1,' + outdentSize + '}', 'gm'), '', content);
}
children.push({
type: 'listItem',
spread: REG_LOOSE.test(content),
checked,
children: parser.parse(content),
});
}
return (0, util_1.token)(subvalue, 'list', children, { ordered: start !== null, start });
};
const splitCells = (tableRow, count) => {
const cells = (0, util_1.rep)(/([^\\])\|/g, '$1 |', tableRow).split(/ +\| */);
if (count !== void 0) {
if (cells.length > count)
cells.splice(count);
else
while (cells.length < count)
cells.push('');
}
for (let i = 0; i < cells.length; i++)
cells[i] = (0, util_1.rep)(/\\\|/g, '|', cells[i]);
return cells;
};
const REG_TABLE = /^ *\|?(.+)\n *\|?( *[-:]+[-| :]*)(?:\n((?: *[^>\n ].*(?:\n|$))*)\n*|$)/;
const table = (parser, value) => {
const matches = value.match(REG_TABLE);
if (!matches)
return;
const subvalue = matches[0];
const header = matches[1];
const align = (0, util_1.rep)(/^ *|\| *$/g, '', matches[2])
.split(/ *\| */)
.map((spec) => {
spec = spec.trim();
return spec[0] === ':'
? spec[spec.length - 1] === ':'
? 'center'
: 'left'
: spec[spec.length - 1] === ':'
? 'right'
: null;
});
const rows = matches[3] ? (0, util_1.rep)(/(?: *\| *)?\n$/, '', matches[3]).split('\n') : [];
const children = [];
const headers = splitCells((0, util_1.rep)(/^ *| *\| *$/g, '', header)).map((headerText) => ({
type: 'tableCell',
children: parser.parsei(headerText),
}));
children.push({
type: 'tableRow',
children: headers,
});
if (rows && rows.length) {
for (let i = 0; i < rows.length; i++) {
const row = rows[i];
const cells = splitCells((0, util_1.rep)(/^ *\| *| *\| *$/g, '', row), headers.length);
children.push({
type: 'tableRow',
children: cells.map((cellRawValue) => ({
type: 'tableCell',
children: parser.parsei(cellRawValue),
})),
});
}
}
return (0, util_1.token)(subvalue, 'table', children, { align });
};
const REG_FOOTNOTE_DEF = /^\[\^([a-zA-Z0-9\-_]+)\]: *([^\n]*(\n?(( {2}([^\n]*)\n?)|\n(?!\n))*)?)/;
const footnoteDefinition = (parser, value) => {
const matches = value.match(REG_FOOTNOTE_DEF);
if (!matches)
return;
const subvalue = matches[0];
const label = matches[1];
const identifier = label.toLowerCase();
const outdented = (0, util_1.rep)(/^ {1,4}/gm, '', matches[2]);
const children = parser.parse(outdented);
return (0, util_1.token)(subvalue, 'footnoteDefinition', children, { label, identifier });
};
const definition = (_, value) => {
const matches = value.match(reg.def);
if (!matches)
return;
const subvalue = matches[0];
const label = matches[1];
const title = matches[3];
return (0, util_1.token)(subvalue, 'definition', void 0, {
label: label,
identifier: label.toLowerCase(),
title: title ? title.slice(1, -1) : null,
url: matches[2],
});
};
const html = (_, src) => html_1.html.el(src);
const REG_PARAGRAPH = reg.replace(/^((?:[^\n]+(\n(?!\s{0,3}bull))?)+)\n*/, { bull });
const paragraph = (parser, value) => {
const matches = value.match(REG_PARAGRAPH);
if (matches)
return (0, util_1.token)(matches[0], 'paragraph', parser.parsei(matches[1].trim()));
};
exports.parsers = [
newline,
code,
fences,
math,
thematicBreak,
heading,
blockquote,
list,
table,
footnoteDefinition,
definition,
html,
paragraph,
];