1 | 'use strict';
|
---|
2 |
|
---|
3 | const openBracket = '{'.charCodeAt(0);
|
---|
4 | const closeBracket = '}'.charCodeAt(0);
|
---|
5 | const openParen = '('.charCodeAt(0);
|
---|
6 | const closeParen = ')'.charCodeAt(0);
|
---|
7 | const singleQuote = '\''.charCodeAt(0);
|
---|
8 | const doubleQuote = '"'.charCodeAt(0);
|
---|
9 | const backslash = '\\'.charCodeAt(0);
|
---|
10 | const slash = '/'.charCodeAt(0);
|
---|
11 | const period = '.'.charCodeAt(0);
|
---|
12 | const comma = ','.charCodeAt(0);
|
---|
13 | const colon = ':'.charCodeAt(0);
|
---|
14 | const asterisk = '*'.charCodeAt(0);
|
---|
15 | const minus = '-'.charCodeAt(0);
|
---|
16 | const plus = '+'.charCodeAt(0);
|
---|
17 | const pound = '#'.charCodeAt(0);
|
---|
18 | const newline = '\n'.charCodeAt(0);
|
---|
19 | const space = ' '.charCodeAt(0);
|
---|
20 | const feed = '\f'.charCodeAt(0);
|
---|
21 | const tab = '\t'.charCodeAt(0);
|
---|
22 | const cr = '\r'.charCodeAt(0);
|
---|
23 | const at = '@'.charCodeAt(0);
|
---|
24 | const lowerE = 'e'.charCodeAt(0);
|
---|
25 | const upperE = 'E'.charCodeAt(0);
|
---|
26 | const digit0 = '0'.charCodeAt(0);
|
---|
27 | const digit9 = '9'.charCodeAt(0);
|
---|
28 | const lowerU = 'u'.charCodeAt(0);
|
---|
29 | const upperU = 'U'.charCodeAt(0);
|
---|
30 | const atEnd = /[ \n\t\r\{\(\)'"\\;,/]/g;
|
---|
31 | const wordEnd = /[ \n\t\r\(\)\{\}\*:;@!&'"\+\|~>,\[\]\\]|\/(?=\*)/g;
|
---|
32 | const wordEndNum = /[ \n\t\r\(\)\{\}\*:;@!&'"\-\+\|~>,\[\]\\]|\//g;
|
---|
33 | const alphaNum = /^[a-z0-9]/i;
|
---|
34 | const unicodeRange = /^[a-f0-9?\-]/i;
|
---|
35 |
|
---|
36 | const util = require('util');
|
---|
37 | const TokenizeError = require('./errors/TokenizeError');
|
---|
38 |
|
---|
39 | module.exports = function tokenize (input, options) {
|
---|
40 |
|
---|
41 | options = options || {};
|
---|
42 |
|
---|
43 | let tokens = [],
|
---|
44 | css = input.valueOf(),
|
---|
45 | length = css.length,
|
---|
46 | offset = -1,
|
---|
47 | line = 1,
|
---|
48 | pos = 0,
|
---|
49 | parentCount = 0,
|
---|
50 | isURLArg = null,
|
---|
51 |
|
---|
52 | code, next, quote, lines, last, content, escape, nextLine, nextOffset,
|
---|
53 | escaped, escapePos, nextChar;
|
---|
54 |
|
---|
55 | function unclosed (what) {
|
---|
56 | let message = util.format('Unclosed %s at line: %d, column: %d, token: %d', what, line, pos - offset, pos);
|
---|
57 | throw new TokenizeError(message);
|
---|
58 | }
|
---|
59 |
|
---|
60 | function tokenizeError () {
|
---|
61 | let message = util.format('Syntax error at line: %d, column: %d, token: %d', line, pos - offset, pos);
|
---|
62 | throw new TokenizeError(message);
|
---|
63 | }
|
---|
64 |
|
---|
65 | while (pos < length) {
|
---|
66 | code = css.charCodeAt(pos);
|
---|
67 |
|
---|
68 | if (code === newline) {
|
---|
69 | offset = pos;
|
---|
70 | line += 1;
|
---|
71 | }
|
---|
72 |
|
---|
73 | switch (code) {
|
---|
74 | case newline:
|
---|
75 | case space:
|
---|
76 | case tab:
|
---|
77 | case cr:
|
---|
78 | case feed:
|
---|
79 | next = pos;
|
---|
80 | do {
|
---|
81 | next += 1;
|
---|
82 | code = css.charCodeAt(next);
|
---|
83 | if (code === newline) {
|
---|
84 | offset = next;
|
---|
85 | line += 1;
|
---|
86 | }
|
---|
87 | } while (code === space ||
|
---|
88 | code === newline ||
|
---|
89 | code === tab ||
|
---|
90 | code === cr ||
|
---|
91 | code === feed);
|
---|
92 |
|
---|
93 | tokens.push(['space', css.slice(pos, next),
|
---|
94 | line, pos - offset,
|
---|
95 | line, next - offset,
|
---|
96 | pos
|
---|
97 | ]);
|
---|
98 |
|
---|
99 | pos = next - 1;
|
---|
100 | break;
|
---|
101 |
|
---|
102 | case colon:
|
---|
103 | next = pos + 1;
|
---|
104 | tokens.push(['colon', css.slice(pos, next),
|
---|
105 | line, pos - offset,
|
---|
106 | line, next - offset,
|
---|
107 | pos
|
---|
108 | ]);
|
---|
109 |
|
---|
110 | pos = next - 1;
|
---|
111 | break;
|
---|
112 |
|
---|
113 | case comma:
|
---|
114 | next = pos + 1;
|
---|
115 | tokens.push(['comma', css.slice(pos, next),
|
---|
116 | line, pos - offset,
|
---|
117 | line, next - offset,
|
---|
118 | pos
|
---|
119 | ]);
|
---|
120 |
|
---|
121 | pos = next - 1;
|
---|
122 | break;
|
---|
123 |
|
---|
124 | case openBracket:
|
---|
125 | tokens.push(['{', '{',
|
---|
126 | line, pos - offset,
|
---|
127 | line, next - offset,
|
---|
128 | pos
|
---|
129 | ]);
|
---|
130 | break;
|
---|
131 |
|
---|
132 | case closeBracket:
|
---|
133 | tokens.push(['}', '}',
|
---|
134 | line, pos - offset,
|
---|
135 | line, next - offset,
|
---|
136 | pos
|
---|
137 | ]);
|
---|
138 | break;
|
---|
139 |
|
---|
140 | case openParen:
|
---|
141 | parentCount++;
|
---|
142 | isURLArg = !isURLArg && parentCount === 1 &&
|
---|
143 | tokens.length > 0 &&
|
---|
144 | tokens[tokens.length - 1][0] === "word" &&
|
---|
145 | tokens[tokens.length - 1][1] === "url";
|
---|
146 | tokens.push(['(', '(',
|
---|
147 | line, pos - offset,
|
---|
148 | line, next - offset,
|
---|
149 | pos
|
---|
150 | ]);
|
---|
151 | break;
|
---|
152 |
|
---|
153 | case closeParen:
|
---|
154 | parentCount--;
|
---|
155 | isURLArg = isURLArg && parentCount > 0;
|
---|
156 | tokens.push([')', ')',
|
---|
157 | line, pos - offset,
|
---|
158 | line, next - offset,
|
---|
159 | pos
|
---|
160 | ]);
|
---|
161 | break;
|
---|
162 |
|
---|
163 | case singleQuote:
|
---|
164 | case doubleQuote:
|
---|
165 | quote = code === singleQuote ? '\'' : '"';
|
---|
166 | next = pos;
|
---|
167 | do {
|
---|
168 | escaped = false;
|
---|
169 | next = css.indexOf(quote, next + 1);
|
---|
170 | if (next === -1) {
|
---|
171 | unclosed('quote', quote);
|
---|
172 | }
|
---|
173 | escapePos = next;
|
---|
174 | while (css.charCodeAt(escapePos - 1) === backslash) {
|
---|
175 | escapePos -= 1;
|
---|
176 | escaped = !escaped;
|
---|
177 | }
|
---|
178 | } while (escaped);
|
---|
179 |
|
---|
180 | tokens.push(['string', css.slice(pos, next + 1),
|
---|
181 | line, pos - offset,
|
---|
182 | line, next - offset,
|
---|
183 | pos
|
---|
184 | ]);
|
---|
185 | pos = next;
|
---|
186 | break;
|
---|
187 |
|
---|
188 | case at:
|
---|
189 | atEnd.lastIndex = pos + 1;
|
---|
190 | atEnd.test(css);
|
---|
191 |
|
---|
192 | if (atEnd.lastIndex === 0) {
|
---|
193 | next = css.length - 1;
|
---|
194 | }
|
---|
195 | else {
|
---|
196 | next = atEnd.lastIndex - 2;
|
---|
197 | }
|
---|
198 |
|
---|
199 | tokens.push(['atword', css.slice(pos, next + 1),
|
---|
200 | line, pos - offset,
|
---|
201 | line, next - offset,
|
---|
202 | pos
|
---|
203 | ]);
|
---|
204 | pos = next;
|
---|
205 | break;
|
---|
206 |
|
---|
207 | case backslash:
|
---|
208 | next = pos;
|
---|
209 | code = css.charCodeAt(next + 1);
|
---|
210 |
|
---|
211 | if (escape && (code !== slash && code !== space &&
|
---|
212 | code !== newline && code !== tab &&
|
---|
213 | code !== cr && code !== feed)) {
|
---|
214 | next += 1;
|
---|
215 | }
|
---|
216 |
|
---|
217 | tokens.push(['word', css.slice(pos, next + 1),
|
---|
218 | line, pos - offset,
|
---|
219 | line, next - offset,
|
---|
220 | pos
|
---|
221 | ]);
|
---|
222 |
|
---|
223 | pos = next;
|
---|
224 | break;
|
---|
225 |
|
---|
226 | case plus:
|
---|
227 | case minus:
|
---|
228 | case asterisk:
|
---|
229 | next = pos + 1;
|
---|
230 | nextChar = css.slice(pos + 1, next + 1);
|
---|
231 |
|
---|
232 | let prevChar = css.slice(pos - 1, pos);
|
---|
233 |
|
---|
234 | // if the operator is immediately followed by a word character, then we
|
---|
235 | // have a prefix of some kind, and should fall-through. eg. -webkit
|
---|
236 |
|
---|
237 | // look for --* for custom variables
|
---|
238 | if (code === minus && nextChar.charCodeAt(0) === minus) {
|
---|
239 | next++;
|
---|
240 |
|
---|
241 | tokens.push(['word', css.slice(pos, next),
|
---|
242 | line, pos - offset,
|
---|
243 | line, next - offset,
|
---|
244 | pos
|
---|
245 | ]);
|
---|
246 |
|
---|
247 | pos = next - 1;
|
---|
248 | break;
|
---|
249 | }
|
---|
250 |
|
---|
251 | tokens.push(['operator', css.slice(pos, next),
|
---|
252 | line, pos - offset,
|
---|
253 | line, next - offset,
|
---|
254 | pos
|
---|
255 | ]);
|
---|
256 |
|
---|
257 | pos = next - 1;
|
---|
258 | break;
|
---|
259 |
|
---|
260 | default:
|
---|
261 | if (code === slash && (css.charCodeAt(pos + 1) === asterisk || (options.loose && !isURLArg && css.charCodeAt(pos + 1) === slash))) {
|
---|
262 | const isStandardComment = css.charCodeAt(pos + 1) === asterisk;
|
---|
263 |
|
---|
264 | if (isStandardComment) {
|
---|
265 | next = css.indexOf('*/', pos + 2) + 1;
|
---|
266 | if (next === 0) {
|
---|
267 | unclosed('comment', '*/');
|
---|
268 | }
|
---|
269 | }
|
---|
270 | else {
|
---|
271 | const newlinePos = css.indexOf('\n', pos + 2);
|
---|
272 |
|
---|
273 | next = newlinePos !== -1 ? newlinePos - 1 : length;
|
---|
274 | }
|
---|
275 |
|
---|
276 | content = css.slice(pos, next + 1);
|
---|
277 | lines = content.split('\n');
|
---|
278 | last = lines.length - 1;
|
---|
279 |
|
---|
280 | if (last > 0) {
|
---|
281 | nextLine = line + last;
|
---|
282 | nextOffset = next - lines[last].length;
|
---|
283 | }
|
---|
284 | else {
|
---|
285 | nextLine = line;
|
---|
286 | nextOffset = offset;
|
---|
287 | }
|
---|
288 |
|
---|
289 | tokens.push(['comment', content,
|
---|
290 | line, pos - offset,
|
---|
291 | nextLine, next - nextOffset,
|
---|
292 | pos
|
---|
293 | ]);
|
---|
294 |
|
---|
295 | offset = nextOffset;
|
---|
296 | line = nextLine;
|
---|
297 | pos = next;
|
---|
298 |
|
---|
299 | }
|
---|
300 | else if (code === pound && !alphaNum.test(css.slice(pos + 1, pos + 2))) {
|
---|
301 | next = pos + 1;
|
---|
302 |
|
---|
303 | tokens.push(['#', css.slice(pos, next),
|
---|
304 | line, pos - offset,
|
---|
305 | line, next - offset,
|
---|
306 | pos
|
---|
307 | ]);
|
---|
308 |
|
---|
309 | pos = next - 1;
|
---|
310 | }
|
---|
311 | else if ((code === lowerU || code === upperU) && css.charCodeAt(pos + 1) === plus) {
|
---|
312 | next = pos + 2;
|
---|
313 |
|
---|
314 | do {
|
---|
315 | next += 1;
|
---|
316 | code = css.charCodeAt(next);
|
---|
317 | } while (next < length && unicodeRange.test(css.slice(next, next + 1)));
|
---|
318 |
|
---|
319 | tokens.push(['unicoderange', css.slice(pos, next),
|
---|
320 | line, pos - offset,
|
---|
321 | line, next - offset,
|
---|
322 | pos
|
---|
323 | ]);
|
---|
324 | pos = next - 1;
|
---|
325 | }
|
---|
326 | // catch a regular slash, that isn't a comment
|
---|
327 | else if (code === slash) {
|
---|
328 | next = pos + 1;
|
---|
329 |
|
---|
330 | tokens.push(['operator', css.slice(pos, next),
|
---|
331 | line, pos - offset,
|
---|
332 | line, next - offset,
|
---|
333 | pos
|
---|
334 | ]);
|
---|
335 |
|
---|
336 | pos = next - 1;
|
---|
337 | }
|
---|
338 | else {
|
---|
339 | let regex = wordEnd;
|
---|
340 |
|
---|
341 | // we're dealing with a word that starts with a number
|
---|
342 | // those get treated differently
|
---|
343 | if (code >= digit0 && code <= digit9) {
|
---|
344 | regex = wordEndNum;
|
---|
345 | }
|
---|
346 |
|
---|
347 | regex.lastIndex = pos + 1;
|
---|
348 | regex.test(css);
|
---|
349 |
|
---|
350 | if (regex.lastIndex === 0) {
|
---|
351 | next = css.length - 1;
|
---|
352 | }
|
---|
353 | else {
|
---|
354 | next = regex.lastIndex - 2;
|
---|
355 | }
|
---|
356 |
|
---|
357 | // Exponential number notation with minus or plus: 1e-10, 1e+10
|
---|
358 | if (regex === wordEndNum || code === period) {
|
---|
359 | let ncode = css.charCodeAt(next),
|
---|
360 | ncode1 = css.charCodeAt(next + 1),
|
---|
361 | ncode2 = css.charCodeAt(next + 2);
|
---|
362 |
|
---|
363 | if (
|
---|
364 | (ncode === lowerE || ncode === upperE) &&
|
---|
365 | (ncode1 === minus || ncode1 === plus) &&
|
---|
366 | (ncode2 >= digit0 && ncode2 <= digit9)
|
---|
367 | ) {
|
---|
368 | wordEndNum.lastIndex = next + 2;
|
---|
369 | wordEndNum.test(css);
|
---|
370 |
|
---|
371 | if (wordEndNum.lastIndex === 0) {
|
---|
372 | next = css.length - 1;
|
---|
373 | }
|
---|
374 | else {
|
---|
375 | next = wordEndNum.lastIndex - 2;
|
---|
376 | }
|
---|
377 | }
|
---|
378 | }
|
---|
379 |
|
---|
380 | tokens.push(['word', css.slice(pos, next + 1),
|
---|
381 | line, pos - offset,
|
---|
382 | line, next - offset,
|
---|
383 | pos
|
---|
384 | ]);
|
---|
385 | pos = next;
|
---|
386 | }
|
---|
387 | break;
|
---|
388 | }
|
---|
389 |
|
---|
390 | pos ++;
|
---|
391 | }
|
---|
392 |
|
---|
393 | return tokens;
|
---|
394 | };
|
---|