porffor 0.57.25 → 0.57.27
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +0 -7
- package/compiler/builtins/regexp.ts +128 -0
- package/compiler/builtins_precompiled.js +411 -344
- package/compiler/codegen.js +26 -78
- package/compiler/prefs.js +1 -6
- package/compiler/wrap.js +8 -0
- package/package.json +1 -1
- package/runtime/index.js +1 -1
- package/rhemyn/README.md +0 -40
- package/rhemyn/compile.js +0 -331
- package/rhemyn/parse.js +0 -323
- package/rhemyn/test/parse.js +0 -59
package/rhemyn/parse.js
DELETED
@@ -1,323 +0,0 @@
|
|
1
|
-
const State = {
|
2
|
-
none: 0,
|
3
|
-
insideSet: 1
|
4
|
-
};
|
5
|
-
|
6
|
-
const Quantifiers = {
|
7
|
-
'*': [ 0 ], // 0 -
|
8
|
-
'+': [ 1 ], // 1 -
|
9
|
-
'?': [ 0, 1 ], // 0 - 1
|
10
|
-
};
|
11
|
-
const QuantifierKeys = Object.keys(Quantifiers);
|
12
|
-
|
13
|
-
const getArg = (name, def) => {
|
14
|
-
const arg = (typeof process !== 'undefined' ? process.argv : Deno.args).find(x => x.startsWith(`--${name}=`));
|
15
|
-
if (arg) return arg.split('=')[0];
|
16
|
-
|
17
|
-
return def;
|
18
|
-
};
|
19
|
-
|
20
|
-
// full is spec-compliant but slower. not needed most of the time. (evil)
|
21
|
-
const DotChars = () => ({
|
22
|
-
full: [ '\n', '\r', '\u2028', '\u2029' ],
|
23
|
-
fast: [ '\n', '\r' ]
|
24
|
-
})[getArg('regex-dot', 'fast')];
|
25
|
-
|
26
|
-
const WordChars = () => ({
|
27
|
-
full: [ [ 'a', 'z' ], [ 'A', 'Z' ], [ '0', '9' ], '_' ],
|
28
|
-
fast: [ [ '_', 'z' ], [ 'A', 'Z' ], [ '0', '9' ] ] // skip individual _ with _-z BUT it also matches '`'
|
29
|
-
})[getArg('regex-word', 'full')];
|
30
|
-
|
31
|
-
const WhitespaceChars = () => ({
|
32
|
-
full: [ ' ', '\t', '\n', '\r', '\u2028', '\u2029' ],
|
33
|
-
fast: [ ' ', '\t', '\n', '\r' ]
|
34
|
-
})[getArg('regex-ws', 'fast')];
|
35
|
-
|
36
|
-
const _Metachars = () => ({
|
37
|
-
unescaped: {
|
38
|
-
'.': [ DotChars(), true ], // dot
|
39
|
-
},
|
40
|
-
escaped: {
|
41
|
-
d: [ [ [ '0', '9' ] ], false ], // digit
|
42
|
-
D: [ [ [ '0', '9' ] ], true ], // not digit
|
43
|
-
w: [ WordChars(), false ], // word
|
44
|
-
W: [ WordChars(), true ], // not word
|
45
|
-
s: [ WhitespaceChars(), false ], // whitespace
|
46
|
-
S: [ WhitespaceChars(), true ], // not whitespace
|
47
|
-
}
|
48
|
-
});
|
49
|
-
|
50
|
-
const EscapeSequences = {
|
51
|
-
f: '\f',
|
52
|
-
n: '\n',
|
53
|
-
r: '\r',
|
54
|
-
t: '\t',
|
55
|
-
v: '\v',
|
56
|
-
'0': '\0'
|
57
|
-
};
|
58
|
-
|
59
|
-
const HexDigit = /[0-9a-fA-F]/;
|
60
|
-
|
61
|
-
export default str => {
|
62
|
-
const Metachars = _Metachars();
|
63
|
-
|
64
|
-
const out = {
|
65
|
-
type: 'Expression',
|
66
|
-
body: []
|
67
|
-
};
|
68
|
-
let node = out, parents = [];
|
69
|
-
|
70
|
-
let state = State.none, setIndex = 0, escape = false;
|
71
|
-
for (let i = 0; i < str.length; i++) {
|
72
|
-
const c = str[i];
|
73
|
-
|
74
|
-
const charNode = char => ({
|
75
|
-
type: 'Character',
|
76
|
-
char
|
77
|
-
});
|
78
|
-
|
79
|
-
const rangeNode = (from, to) => ({
|
80
|
-
type: 'Range',
|
81
|
-
from,
|
82
|
-
to
|
83
|
-
});
|
84
|
-
|
85
|
-
const addChar = (char = c) => {
|
86
|
-
node.body.push(charNode(char));
|
87
|
-
};
|
88
|
-
|
89
|
-
const addSet = (matches, negated = false) => {
|
90
|
-
let body = matches.map(x => x[1] ? rangeNode(x[0], x[1]) : charNode(x));
|
91
|
-
if (state === State.insideSet) {
|
92
|
-
// if negated, mark each node as negated for merge
|
93
|
-
if (negated) body = body.map(x => {
|
94
|
-
x.negated = true;
|
95
|
-
return x;
|
96
|
-
});
|
97
|
-
|
98
|
-
// already in set, merge bodies
|
99
|
-
node.body.push(...body);
|
100
|
-
return;
|
101
|
-
}
|
102
|
-
|
103
|
-
node.body.push({
|
104
|
-
type: 'Set',
|
105
|
-
body,
|
106
|
-
negated
|
107
|
-
});
|
108
|
-
};
|
109
|
-
|
110
|
-
const addMetachar = meta => {
|
111
|
-
const [ matches, negated = false ] = meta;
|
112
|
-
return addSet(matches, negated);
|
113
|
-
};
|
114
|
-
|
115
|
-
// get next char and consume it
|
116
|
-
const seek = (allowEscaped = true) => {
|
117
|
-
const cNext = str[++i];
|
118
|
-
|
119
|
-
if (cNext === '\\') return !allowEscaped ? undefined : [ str[++i], true ];
|
120
|
-
return !allowEscaped ? cNext : [ cNext, false ];
|
121
|
-
};
|
122
|
-
|
123
|
-
// get next char without consuming
|
124
|
-
const peek = (allowEscaped = true, offset = 0) => {
|
125
|
-
const cNext = str[i + 1 + offset];
|
126
|
-
|
127
|
-
if (cNext === '\\') return !allowEscaped ? undefined : [ str[i + 2 + offset], true ];
|
128
|
-
return !allowEscaped ? cNext : [ cNext, false ];
|
129
|
-
};
|
130
|
-
|
131
|
-
if (escape) {
|
132
|
-
escape = false;
|
133
|
-
if (EscapeSequences[c]) {
|
134
|
-
addChar(EscapeSequences[c]);
|
135
|
-
continue;
|
136
|
-
}
|
137
|
-
|
138
|
-
if (Metachars.escaped[c]) {
|
139
|
-
addMetachar(Metachars.escaped[c]);
|
140
|
-
continue;
|
141
|
-
}
|
142
|
-
|
143
|
-
if (c === 'c') {
|
144
|
-
// \c (not [A-Za-z] ...) = literal \c... (WHY)
|
145
|
-
const next = peek(false);
|
146
|
-
if (next == null || /[^a-zA-Z]/.test(next)) {
|
147
|
-
addChar('\\');
|
148
|
-
addChar('c');
|
149
|
-
continue;
|
150
|
-
}
|
151
|
-
|
152
|
-
// \c[A-Za-z]
|
153
|
-
const code = seek(false).charCodeAt(0);
|
154
|
-
addChar(String.fromCharCode(code % 32));
|
155
|
-
continue;
|
156
|
-
}
|
157
|
-
|
158
|
-
if (c === 'x') {
|
159
|
-
// \x = x
|
160
|
-
// \xH = xH
|
161
|
-
// \x[0-9a-zA-Z][0-9a-zA-Z] = \xAB
|
162
|
-
const next1 = peek(false);
|
163
|
-
const next2 = peek(false, 1);
|
164
|
-
|
165
|
-
// missing a char or invalid hex digit
|
166
|
-
if (next1 == null || next2 == null || !HexDigit.test(next1) || !HexDigit.test(next2)) {
|
167
|
-
addChar('x');
|
168
|
-
continue;
|
169
|
-
}
|
170
|
-
|
171
|
-
const code = parseInt(seek(false) + seek(false), 16);
|
172
|
-
addChar(String.fromCodePoint(code));
|
173
|
-
continue;
|
174
|
-
}
|
175
|
-
|
176
|
-
if (c === 'u') {
|
177
|
-
// '\u' = u
|
178
|
-
// '\uHHH' = uHHH
|
179
|
-
// '\uABCD' = \uABCD
|
180
|
-
const next1 = peek(false);
|
181
|
-
const next2 = peek(false, 1);
|
182
|
-
const next3 = peek(false, 2);
|
183
|
-
const next4 = peek(false, 3);
|
184
|
-
|
185
|
-
// missing a char or invalid hex digit
|
186
|
-
if (next1 == null || next2 == null || next3 == null || next4 == null || !HexDigit.test(next1) || !HexDigit.test(next2) || !HexDigit.test(next3) || !HexDigit.test(next4)) {
|
187
|
-
addChar('u');
|
188
|
-
continue;
|
189
|
-
}
|
190
|
-
|
191
|
-
const code = parseInt(seek(false) + seek(false) + seek(false) + seek(false), 16);
|
192
|
-
addChar(String.fromCodePoint(code));
|
193
|
-
continue;
|
194
|
-
}
|
195
|
-
|
196
|
-
addChar();
|
197
|
-
continue;
|
198
|
-
}
|
199
|
-
|
200
|
-
if (c === '\\') {
|
201
|
-
escape = true;
|
202
|
-
continue;
|
203
|
-
}
|
204
|
-
|
205
|
-
switch (state) {
|
206
|
-
case State.none:
|
207
|
-
if (c === '[') {
|
208
|
-
parents.push(node);
|
209
|
-
node = {
|
210
|
-
type: 'Set',
|
211
|
-
body: [],
|
212
|
-
negated: false
|
213
|
-
};
|
214
|
-
|
215
|
-
parents.at(-1).body.push(node);
|
216
|
-
|
217
|
-
state = State.insideSet;
|
218
|
-
setIndex = 0;
|
219
|
-
continue;
|
220
|
-
}
|
221
|
-
|
222
|
-
if (c === '(') {
|
223
|
-
parents.push(node);
|
224
|
-
node = {
|
225
|
-
type: 'Group',
|
226
|
-
body: []
|
227
|
-
};
|
228
|
-
|
229
|
-
parents.at(-1).body.push(node);
|
230
|
-
continue;
|
231
|
-
}
|
232
|
-
|
233
|
-
if (c === ')') {
|
234
|
-
if (node.type !== 'Group') throw new SyntaxError('Unmatched closing parenthesis');
|
235
|
-
|
236
|
-
node = parents.pop();
|
237
|
-
continue;
|
238
|
-
}
|
239
|
-
|
240
|
-
if (QuantifierKeys.includes(c)) {
|
241
|
-
const last = node.body.at(-1);
|
242
|
-
if (!last) continue; // ignore, maybe lookahead
|
243
|
-
|
244
|
-
last.quantifier = Quantifiers[c];
|
245
|
-
|
246
|
-
// lazy modifier
|
247
|
-
if (peek(false) === '?') last.lazy = true;
|
248
|
-
|
249
|
-
continue;
|
250
|
-
}
|
251
|
-
|
252
|
-
if (Metachars.unescaped[c]) {
|
253
|
-
addMetachar(Metachars.unescaped[c]);
|
254
|
-
continue;
|
255
|
-
}
|
256
|
-
|
257
|
-
addChar();
|
258
|
-
break;
|
259
|
-
|
260
|
-
case State.insideSet:
|
261
|
-
setIndex++;
|
262
|
-
if (setIndex === 1) {
|
263
|
-
// first char in set
|
264
|
-
if (c === '^') {
|
265
|
-
node.negated = true;
|
266
|
-
continue;
|
267
|
-
}
|
268
|
-
}
|
269
|
-
|
270
|
-
if (c === ']') {
|
271
|
-
state = State.none;
|
272
|
-
node = parents.pop();
|
273
|
-
|
274
|
-
continue;
|
275
|
-
}
|
276
|
-
|
277
|
-
// range
|
278
|
-
if (c === '-') {
|
279
|
-
// start of set (or not char), just literal -
|
280
|
-
if (node.body.at(-1)?.char == null) {
|
281
|
-
addChar(); // add -
|
282
|
-
continue;
|
283
|
-
}
|
284
|
-
|
285
|
-
const from = node.body.pop().char;
|
286
|
-
const [ to, escaped ] = seek();
|
287
|
-
|
288
|
-
// end of set, just literal -
|
289
|
-
if (to == null || (!escaped && to === ']')) {
|
290
|
-
addChar(from); // add from char back
|
291
|
-
i--; // rollback seek
|
292
|
-
|
293
|
-
addChar(); // add -
|
294
|
-
continue;
|
295
|
-
}
|
296
|
-
|
297
|
-
// next char was escaped and a metachar, just literal -
|
298
|
-
if (escaped && Metachars.escaped[to] != null) {
|
299
|
-
i -= 2; // rollback seek
|
300
|
-
|
301
|
-
addChar(); // add -
|
302
|
-
continue;
|
303
|
-
}
|
304
|
-
|
305
|
-
if (to < from) throw new SyntaxError('Range out of order');
|
306
|
-
|
307
|
-
node.body.push(rangeNode(from, to));
|
308
|
-
continue;
|
309
|
-
}
|
310
|
-
|
311
|
-
addChar();
|
312
|
-
break;
|
313
|
-
}
|
314
|
-
}
|
315
|
-
|
316
|
-
// still in a group by the end
|
317
|
-
if (node.type !== 'Expression') throw new SyntaxError('Unmatched opening parenthesis');
|
318
|
-
|
319
|
-
// still in a set by the end
|
320
|
-
if (state === State.insideSet) throw new SyntaxError('Unmatched opening square bracket');
|
321
|
-
|
322
|
-
return out;
|
323
|
-
};
|
package/rhemyn/test/parse.js
DELETED
@@ -1,59 +0,0 @@
|
|
1
|
-
import util from 'node:util';
|
2
|
-
|
3
|
-
import parse from '../parse.js';
|
4
|
-
|
5
|
-
const tests = {
|
6
|
-
'a': {},
|
7
|
-
'a(b)': {},
|
8
|
-
'a(b(c))': {},
|
9
|
-
'ab': {},
|
10
|
-
'[ab]': {},
|
11
|
-
'[a-z]': {},
|
12
|
-
'a*': {},
|
13
|
-
'a+': {},
|
14
|
-
'a?': {},
|
15
|
-
'a(b)+': {},
|
16
|
-
'[^a]': {},
|
17
|
-
'[a^]': {},
|
18
|
-
'[^ab]': {},
|
19
|
-
'.': {},
|
20
|
-
|
21
|
-
// not range
|
22
|
-
'[-]': {},
|
23
|
-
'[0-]': {},
|
24
|
-
'[-0]': {},
|
25
|
-
'[\\s-\\S]': {},
|
26
|
-
'[\\s-.]': {},
|
27
|
-
|
28
|
-
'[\\S]': {},
|
29
|
-
|
30
|
-
'\\c': {},
|
31
|
-
'\\c0': {},
|
32
|
-
'\\cJ': {},
|
33
|
-
|
34
|
-
'\\x': {},
|
35
|
-
'\\x0': {},
|
36
|
-
'\\x0g': {},
|
37
|
-
'\\x0a': {},
|
38
|
-
|
39
|
-
'\\u': {},
|
40
|
-
'\\u0': {},
|
41
|
-
'\\u000': {},
|
42
|
-
'\\u000g': {},
|
43
|
-
'\\u000a': {},
|
44
|
-
|
45
|
-
/*
|
46
|
-
// email regexes
|
47
|
-
'^[a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+$': {},
|
48
|
-
|
49
|
-
// input type=email from HTML spec
|
50
|
-
// https://html.spec.whatwg.org/multipage/input.html#email-state-(type=email)
|
51
|
-
// simpler form
|
52
|
-
'^[a-zA-Z0-9.!#$%&’*+/=?^_`{|}~-]+@[a-zA-Z0-9-]+(?:\\.[a-zA-Z0-9-]+)*$': {},
|
53
|
-
// full/complex form
|
54
|
-
'^[a-zA-Z0-9.!#$%&\'*+\\/=?^_`{|}~-]+@[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(?:\\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*$': {}*/
|
55
|
-
};
|
56
|
-
|
57
|
-
for (const str in tests) {
|
58
|
-
console.log(str, util.inspect(parse(str), false, null, true));
|
59
|
-
}
|