mirror of https://github.com/jkjoy/sunpeiwen.git
282 lines
11 KiB
JavaScript
282 lines
11 KiB
JavaScript
import { createLexer } from 'leac';
|
|
import * as p from 'peberminta';
|
|
|
|
var ast = /*#__PURE__*/Object.freeze({
|
|
__proto__: null
|
|
});
|
|
|
|
const ws = `(?:[ \\t\\r\\n\\f]*)`;
|
|
const nl = `(?:\\n|\\r\\n|\\r|\\f)`;
|
|
const nonascii = `[^\\x00-\\x7F]`;
|
|
const unicode = `(?:\\\\[0-9a-f]{1,6}(?:\\r\\n|[ \\n\\r\\t\\f])?)`;
|
|
const escape = `(?:\\\\[^\\n\\r\\f0-9a-f])`;
|
|
const nmstart = `(?:[_a-z]|${nonascii}|${unicode}|${escape})`;
|
|
const nmchar = `(?:[_a-z0-9-]|${nonascii}|${unicode}|${escape})`;
|
|
const name = `(?:${nmchar}+)`;
|
|
const ident = `(?:[-]?${nmstart}${nmchar}*)`;
|
|
const string1 = `'([^\\n\\r\\f\\\\']|\\\\${nl}|${nonascii}|${unicode}|${escape})*'`;
|
|
const string2 = `"([^\\n\\r\\f\\\\"]|\\\\${nl}|${nonascii}|${unicode}|${escape})*"`;
|
|
const lexSelector = createLexer([
|
|
{ name: 'ws', regex: new RegExp(ws) },
|
|
{ name: 'hash', regex: new RegExp(`#${name}`, 'i') },
|
|
{ name: 'ident', regex: new RegExp(ident, 'i') },
|
|
{ name: 'str1', regex: new RegExp(string1, 'i') },
|
|
{ name: 'str2', regex: new RegExp(string2, 'i') },
|
|
{ name: '*' },
|
|
{ name: '.' },
|
|
{ name: ',' },
|
|
{ name: '[' },
|
|
{ name: ']' },
|
|
{ name: '=' },
|
|
{ name: '>' },
|
|
{ name: '|' },
|
|
{ name: '+' },
|
|
{ name: '~' },
|
|
{ name: '^' },
|
|
{ name: '$' },
|
|
]);
|
|
const lexEscapedString = createLexer([
|
|
{ name: 'unicode', regex: new RegExp(unicode, 'i') },
|
|
{ name: 'escape', regex: new RegExp(escape, 'i') },
|
|
{ name: 'any', regex: new RegExp('[\\s\\S]', 'i') }
|
|
]);
|
|
function sumSpec([a0, a1, a2], [b0, b1, b2]) {
|
|
return [a0 + b0, a1 + b1, a2 + b2];
|
|
}
|
|
function sumAllSpec(ss) {
|
|
return ss.reduce(sumSpec, [0, 0, 0]);
|
|
}
|
|
const unicodeEscapedSequence_ = p.token((t) => t.name === 'unicode' ? String.fromCodePoint(parseInt(t.text.slice(1), 16)) : undefined);
|
|
const escapedSequence_ = p.token((t) => t.name === 'escape' ? t.text.slice(1) : undefined);
|
|
const anyChar_ = p.token((t) => t.name === 'any' ? t.text : undefined);
|
|
const escapedString_ = p.map(p.many(p.or(unicodeEscapedSequence_, escapedSequence_, anyChar_)), (cs) => cs.join(''));
|
|
function unescape(escapedString) {
|
|
const lexerResult = lexEscapedString(escapedString);
|
|
const result = escapedString_({ tokens: lexerResult.tokens, options: undefined }, 0);
|
|
return result.value;
|
|
}
|
|
function literal(name) {
|
|
return p.token((t) => t.name === name ? true : undefined);
|
|
}
|
|
const whitespace_ = p.token((t) => t.name === 'ws' ? null : undefined);
|
|
const optionalWhitespace_ = p.option(whitespace_, null);
|
|
function optionallySpaced(parser) {
|
|
return p.middle(optionalWhitespace_, parser, optionalWhitespace_);
|
|
}
|
|
const identifier_ = p.token((t) => t.name === 'ident' ? unescape(t.text) : undefined);
|
|
const hashId_ = p.token((t) => t.name === 'hash' ? unescape(t.text.slice(1)) : undefined);
|
|
const string_ = p.token((t) => t.name.startsWith('str') ? unescape(t.text.slice(1, -1)) : undefined);
|
|
const namespace_ = p.left(p.option(identifier_, ''), literal('|'));
|
|
const qualifiedName_ = p.eitherOr(p.ab(namespace_, identifier_, (ns, name) => ({ name: name, namespace: ns })), p.map(identifier_, (name) => ({ name: name, namespace: null })));
|
|
const uniSelector_ = p.eitherOr(p.ab(namespace_, literal('*'), (ns) => ({ type: 'universal', namespace: ns, specificity: [0, 0, 0] })), p.map(literal('*'), () => ({ type: 'universal', namespace: null, specificity: [0, 0, 0] })));
|
|
const tagSelector_ = p.map(qualifiedName_, ({ name, namespace }) => ({
|
|
type: 'tag',
|
|
name: name,
|
|
namespace: namespace,
|
|
specificity: [0, 0, 1]
|
|
}));
|
|
const classSelector_ = p.ab(literal('.'), identifier_, (fullstop, name) => ({
|
|
type: 'class',
|
|
name: name,
|
|
specificity: [0, 1, 0]
|
|
}));
|
|
const idSelector_ = p.map(hashId_, (name) => ({
|
|
type: 'id',
|
|
name: name,
|
|
specificity: [1, 0, 0]
|
|
}));
|
|
const attrModifier_ = p.token((t) => {
|
|
if (t.name === 'ident') {
|
|
if (t.text === 'i' || t.text === 'I') {
|
|
return 'i';
|
|
}
|
|
if (t.text === 's' || t.text === 'S') {
|
|
return 's';
|
|
}
|
|
}
|
|
return undefined;
|
|
});
|
|
const attrValue_ = p.eitherOr(p.ab(string_, p.option(p.right(optionalWhitespace_, attrModifier_), null), (v, mod) => ({ value: v, modifier: mod })), p.ab(identifier_, p.option(p.right(whitespace_, attrModifier_), null), (v, mod) => ({ value: v, modifier: mod })));
|
|
const attrMatcher_ = p.choice(p.map(literal('='), () => '='), p.ab(literal('~'), literal('='), () => '~='), p.ab(literal('|'), literal('='), () => '|='), p.ab(literal('^'), literal('='), () => '^='), p.ab(literal('$'), literal('='), () => '$='), p.ab(literal('*'), literal('='), () => '*='));
|
|
const attrPresenceSelector_ = p.abc(literal('['), optionallySpaced(qualifiedName_), literal(']'), (lbr, { name, namespace }) => ({
|
|
type: 'attrPresence',
|
|
name: name,
|
|
namespace: namespace,
|
|
specificity: [0, 1, 0]
|
|
}));
|
|
const attrValueSelector_ = p.middle(literal('['), p.abc(optionallySpaced(qualifiedName_), attrMatcher_, optionallySpaced(attrValue_), ({ name, namespace }, matcher, { value, modifier }) => ({
|
|
type: 'attrValue',
|
|
name: name,
|
|
namespace: namespace,
|
|
matcher: matcher,
|
|
value: value,
|
|
modifier: modifier,
|
|
specificity: [0, 1, 0]
|
|
})), literal(']'));
|
|
const attrSelector_ = p.eitherOr(attrPresenceSelector_, attrValueSelector_);
|
|
const typeSelector_ = p.eitherOr(uniSelector_, tagSelector_);
|
|
const subclassSelector_ = p.choice(idSelector_, classSelector_, attrSelector_);
|
|
const compoundSelector_ = p.map(p.eitherOr(p.flatten(typeSelector_, p.many(subclassSelector_)), p.many1(subclassSelector_)), (ss) => {
|
|
return {
|
|
type: 'compound',
|
|
list: ss,
|
|
specificity: sumAllSpec(ss.map(s => s.specificity))
|
|
};
|
|
});
|
|
const combinator_ = p.choice(p.map(literal('>'), () => '>'), p.map(literal('+'), () => '+'), p.map(literal('~'), () => '~'), p.ab(literal('|'), literal('|'), () => '||'));
|
|
const combinatorSeparator_ = p.eitherOr(optionallySpaced(combinator_), p.map(whitespace_, () => ' '));
|
|
const complexSelector_ = p.leftAssoc2(compoundSelector_, p.map(combinatorSeparator_, (c) => (left, right) => ({
|
|
type: 'compound',
|
|
list: [...right.list, { type: 'combinator', combinator: c, left: left, specificity: left.specificity }],
|
|
specificity: sumSpec(left.specificity, right.specificity)
|
|
})), compoundSelector_);
|
|
const listSelector_ = p.leftAssoc2(p.map(complexSelector_, (s) => ({ type: 'list', list: [s] })), p.map(optionallySpaced(literal(',')), () => (acc, next) => ({ type: 'list', list: [...acc.list, next] })), complexSelector_);
|
|
function parse_(parser, str) {
|
|
if (!(typeof str === 'string' || str instanceof String)) {
|
|
throw new Error('Expected a selector string. Actual input is not a string!');
|
|
}
|
|
const lexerResult = lexSelector(str);
|
|
if (!lexerResult.complete) {
|
|
throw new Error(`The input "${str}" was only partially tokenized, stopped at offset ${lexerResult.offset}!\n` +
|
|
prettyPrintPosition(str, lexerResult.offset));
|
|
}
|
|
const result = optionallySpaced(parser)({ tokens: lexerResult.tokens, options: undefined }, 0);
|
|
if (!result.matched) {
|
|
throw new Error(`No match for "${str}" input!`);
|
|
}
|
|
if (result.position < lexerResult.tokens.length) {
|
|
const token = lexerResult.tokens[result.position];
|
|
throw new Error(`The input "${str}" was only partially parsed, stopped at offset ${token.offset}!\n` +
|
|
prettyPrintPosition(str, token.offset, token.len));
|
|
}
|
|
return result.value;
|
|
}
|
|
function prettyPrintPosition(str, offset, len = 1) {
|
|
return `${str.replace(/(\t)|(\r)|(\n)/g, (m, t, r) => t ? '\u2409' : r ? '\u240d' : '\u240a')}\n${''.padEnd(offset)}${'^'.repeat(len)}`;
|
|
}
|
|
function parse(str) {
|
|
return parse_(listSelector_, str);
|
|
}
|
|
function parse1(str) {
|
|
return parse_(complexSelector_, str);
|
|
}
|
|
|
|
function serialize(selector) {
|
|
if (!selector.type) {
|
|
throw new Error('This is not an AST node.');
|
|
}
|
|
switch (selector.type) {
|
|
case 'universal':
|
|
return _serNs(selector.namespace) + '*';
|
|
case 'tag':
|
|
return _serNs(selector.namespace) + _serIdent(selector.name);
|
|
case 'class':
|
|
return '.' + _serIdent(selector.name);
|
|
case 'id':
|
|
return '#' + _serIdent(selector.name);
|
|
case 'attrPresence':
|
|
return `[${_serNs(selector.namespace)}${_serIdent(selector.name)}]`;
|
|
case 'attrValue':
|
|
return `[${_serNs(selector.namespace)}${_serIdent(selector.name)}${selector.matcher}"${_serStr(selector.value)}"${(selector.modifier ? selector.modifier : '')}]`;
|
|
case 'combinator':
|
|
return serialize(selector.left) + selector.combinator;
|
|
case 'compound':
|
|
return selector.list.reduce((acc, node) => {
|
|
if (node.type === 'combinator') {
|
|
return serialize(node) + acc;
|
|
}
|
|
else {
|
|
return acc + serialize(node);
|
|
}
|
|
}, '');
|
|
case 'list':
|
|
return selector.list.map(serialize).join(',');
|
|
}
|
|
}
|
|
function _serNs(ns) {
|
|
return (ns || ns === '')
|
|
? _serIdent(ns) + '|'
|
|
: '';
|
|
}
|
|
function _codePoint(char) {
|
|
return `\\${char.codePointAt(0).toString(16)} `;
|
|
}
|
|
function _serIdent(str) {
|
|
return str.replace(
|
|
/(^[0-9])|(^-[0-9])|(^-$)|([-0-9a-zA-Z_]|[^\x00-\x7F])|(\x00)|([\x01-\x1f]|\x7f)|([\s\S])/g, (m, d1, d2, hy, safe, nl, ctrl, other) => d1 ? _codePoint(d1) :
|
|
d2 ? '-' + _codePoint(d2.slice(1)) :
|
|
hy ? '\\-' :
|
|
safe ? safe :
|
|
nl ? '\ufffd' :
|
|
ctrl ? _codePoint(ctrl) :
|
|
'\\' + other);
|
|
}
|
|
function _serStr(str) {
|
|
return str.replace(
|
|
/(")|(\\)|(\x00)|([\x01-\x1f]|\x7f)/g, (m, dq, bs, nl, ctrl) => dq ? '\\"' :
|
|
bs ? '\\\\' :
|
|
nl ? '\ufffd' :
|
|
_codePoint(ctrl));
|
|
}
|
|
function normalize(selector) {
|
|
if (!selector.type) {
|
|
throw new Error('This is not an AST node.');
|
|
}
|
|
switch (selector.type) {
|
|
case 'compound': {
|
|
selector.list.forEach(normalize);
|
|
selector.list.sort((a, b) => _compareArrays(_getSelectorPriority(a), _getSelectorPriority(b)));
|
|
break;
|
|
}
|
|
case 'combinator': {
|
|
normalize(selector.left);
|
|
break;
|
|
}
|
|
case 'list': {
|
|
selector.list.forEach(normalize);
|
|
selector.list.sort((a, b) => (serialize(a) < serialize(b)) ? -1 : 1);
|
|
break;
|
|
}
|
|
}
|
|
return selector;
|
|
}
|
|
function _getSelectorPriority(selector) {
|
|
switch (selector.type) {
|
|
case 'universal':
|
|
return [1];
|
|
case 'tag':
|
|
return [1];
|
|
case 'id':
|
|
return [2];
|
|
case 'class':
|
|
return [3, selector.name];
|
|
case 'attrPresence':
|
|
return [4, serialize(selector)];
|
|
case 'attrValue':
|
|
return [5, serialize(selector)];
|
|
case 'combinator':
|
|
return [15, serialize(selector)];
|
|
}
|
|
}
|
|
function compareSelectors(a, b) {
|
|
return _compareArrays(a.specificity, b.specificity);
|
|
}
|
|
function compareSpecificity(a, b) {
|
|
return _compareArrays(a, b);
|
|
}
|
|
function _compareArrays(a, b) {
|
|
if (!Array.isArray(a) || !Array.isArray(b)) {
|
|
throw new Error('Arguments must be arrays.');
|
|
}
|
|
const shorter = (a.length < b.length) ? a.length : b.length;
|
|
for (let i = 0; i < shorter; i++) {
|
|
if (a[i] === b[i]) {
|
|
continue;
|
|
}
|
|
return (a[i] < b[i]) ? -1 : 1;
|
|
}
|
|
return a.length - b.length;
|
|
}
|
|
|
|
export { ast as Ast, compareSelectors, compareSpecificity, normalize, parse, parse1, serialize };
|