forked from googlearchive/code-prettify
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathnode_prettify.js
1715 lines (1591 loc) · 62.8 KB
/
node_prettify.js
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
var IN_GLOBAL_SCOPE = false;
/**
* @license
* Copyright (C) 2006 Google Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @fileoverview
* some functions for browser-side pretty printing of code contained in html.
*
* <p>
* For a fairly comprehensive set of languages see the
* <a href="https://github.com/google/code-prettify#for-which-languages-does-it-work">README</a>
* file that came with this source. At a minimum, the lexer should work on a
* number of languages including C and friends, Java, Python, Bash, SQL, HTML,
* XML, CSS, Javascript, and Makefiles. It works passably on Ruby, PHP and Awk
* and a subset of Perl, but, because of commenting conventions, doesn't work on
* Smalltalk, Lisp-like, or CAML-like languages without an explicit lang class.
* <p>
* Usage: <ol>
* <li> include this source file in an html page via
* {@code <script type="text/javascript" src="/path/to/prettify.js"></script>}
* <li> define style rules. See the example page for examples.
* <li> mark the {@code <pre>} and {@code <code>} tags in your source with
* {@code class=prettyprint.}
* You can also use the (html deprecated) {@code <xmp>} tag, but the pretty
* printer needs to do more substantial DOM manipulations to support that, so
* some css styles may not be preserved.
* </ol>
* That's it. I wanted to keep the API as simple as possible, so there's no
* need to specify which language the code is in, but if you wish, you can add
* another class to the {@code <pre>} or {@code <code>} element to specify the
* language, as in {@code <pre class="prettyprint lang-java">}. Any class that
* starts with "lang-" followed by a file extension, specifies the file type.
* See the "lang-*.js" files in this directory for code that implements
* per-language file handlers.
* <p>
* Change log:<br>
* cbeust, 2006/08/22
* <blockquote>
* Java annotations (start with "@") are now captured as literals ("lit")
* </blockquote>
* @requires console
*/
// JSLint declarations
/*global console, document, navigator, setTimeout, window, define */
/**
* {@type !{
* 'createSimpleLexer': function (Array, Array): (function (JobT)),
* 'registerLangHandler': function (function (JobT), Array.<string>),
* 'PR_ATTRIB_NAME': string,
* 'PR_ATTRIB_NAME': string,
* 'PR_ATTRIB_VALUE': string,
* 'PR_COMMENT': string,
* 'PR_DECLARATION': string,
* 'PR_KEYWORD': string,
* 'PR_LITERAL': string,
* 'PR_NOCODE': string,
* 'PR_PLAIN': string,
* 'PR_PUNCTUATION': string,
* 'PR_SOURCE': string,
* 'PR_STRING': string,
* 'PR_TAG': string,
* 'PR_TYPE': string,
* 'prettyPrintOne': function (string, string, number|boolean),
* 'prettyPrint': function (?function, ?(HTMLElement|HTMLDocument))
* }}
* @const
*/
var PR;
/**
* Split {@code prettyPrint} into multiple timeouts so as not to interfere with
* UI events.
* If set to {@code false}, {@code prettyPrint()} is synchronous.
*/
var PR_SHOULD_USE_CONTINUATION = true
if (typeof window !== 'undefined') {
window['PR_SHOULD_USE_CONTINUATION'] = PR_SHOULD_USE_CONTINUATION;
}
/**
* Pretty print a chunk of code.
* @param {string} sourceCodeHtml The HTML to pretty print.
* @param {string} opt_langExtension The language name to use.
* Typically, a filename extension like 'cpp' or 'java'.
* @param {number|boolean} opt_numberLines True to number lines,
* or the 1-indexed number of the first line in sourceCodeHtml.
* @return {string} code as html, but prettier
*/
var prettyPrintOne;
/**
* Find all the {@code <pre>} and {@code <code>} tags in the DOM with
* {@code class=prettyprint} and prettify them.
*
* @param {Function} opt_whenDone called when prettifying is done.
* @param {HTMLElement|HTMLDocument} opt_root an element or document
* containing all the elements to pretty print.
* Defaults to {@code document.body}.
*/
var prettyPrint;
(function () {
var win = (typeof window !== 'undefined') ? window : {};
// Keyword lists for various languages.
// We use things that coerce to strings to make them compact when minified
// and to defeat aggressive optimizers that fold large string constants.
var FLOW_CONTROL_KEYWORDS = ["break,continue,do,else,for,if,return,while"];
var C_KEYWORDS = [FLOW_CONTROL_KEYWORDS,"auto,case,char,const,default," +
"double,enum,extern,float,goto,inline,int,long,register,restrict,short,signed," +
"sizeof,static,struct,switch,typedef,union,unsigned,void,volatile"];
var COMMON_KEYWORDS = [C_KEYWORDS,"catch,class,delete,false,import," +
"new,operator,private,protected,public,this,throw,true,try,typeof"];
var CPP_KEYWORDS = [COMMON_KEYWORDS,"alignas,alignof,align_union,asm,axiom,bool," +
"concept,concept_map,const_cast,constexpr,decltype,delegate," +
"dynamic_cast,explicit,export,friend,generic,late_check," +
"mutable,namespace,noexcept,noreturn,nullptr,property,reinterpret_cast,static_assert," +
"static_cast,template,typeid,typename,using,virtual,where"];
var JAVA_KEYWORDS = [COMMON_KEYWORDS,
"abstract,assert,boolean,byte,extends,finally,final,implements,import," +
"instanceof,interface,null,native,package,strictfp,super,synchronized," +
"throws,transient"];
var CSHARP_KEYWORDS = [COMMON_KEYWORDS,
"abstract,add,alias,as,ascending,async,await,base,bool,by,byte,checked,decimal,delegate,descending," +
"dynamic,event,finally,fixed,foreach,from,get,global,group,implicit,in,interface," +
"internal,into,is,join,let,lock,null,object,out,override,orderby,params," +
"partial,readonly,ref,remove,sbyte,sealed,select,set,stackalloc,string,select,uint,ulong," +
"unchecked,unsafe,ushort,value,var,virtual,where,yield"];
var COFFEE_KEYWORDS = "all,and,by,catch,class,else,extends,false,finally," +
"for,if,in,is,isnt,loop,new,no,not,null,of,off,on,or,return,super,then," +
"throw,true,try,unless,until,when,while,yes";
var JSCRIPT_KEYWORDS = [COMMON_KEYWORDS,
"abstract,async,await,constructor,debugger,enum,eval,export,from,function," +
"get,import,implements,instanceof,interface,let,null,of,set,undefined," +
"var,with,yield,Infinity,NaN"];
var PERL_KEYWORDS = "caller,delete,die,do,dump,elsif,eval,exit,foreach,for," +
"goto,if,import,last,local,my,next,no,our,print,package,redo,require," +
"sub,undef,unless,until,use,wantarray,while,BEGIN,END";
var PYTHON_KEYWORDS = [FLOW_CONTROL_KEYWORDS, "and,as,assert,class,def,del," +
"elif,except,exec,finally,from,global,import,in,is,lambda," +
"nonlocal,not,or,pass,print,raise,try,with,yield," +
"False,True,None"];
var RUBY_KEYWORDS = [FLOW_CONTROL_KEYWORDS, "alias,and,begin,case,class," +
"def,defined,elsif,end,ensure,false,in,module,next,nil,not,or,redo," +
"rescue,retry,self,super,then,true,undef,unless,until,when,yield," +
"BEGIN,END"];
var SH_KEYWORDS = [FLOW_CONTROL_KEYWORDS, "case,done,elif,esac,eval,fi," +
"function,in,local,set,then,until"];
var ALL_KEYWORDS = [
CPP_KEYWORDS, CSHARP_KEYWORDS, JAVA_KEYWORDS, JSCRIPT_KEYWORDS,
PERL_KEYWORDS, PYTHON_KEYWORDS, RUBY_KEYWORDS, SH_KEYWORDS];
var C_TYPES = /^(DIR|FILE|array|vector|(de|priority_)?queue|(forward_)?list|stack|(const_)?(reverse_)?iterator|(unordered_)?(multi)?(set|map)|bitset|u?(int|float)\d*)\b/;
// token style names. correspond to css classes
/**
* token style for a string literal
* @const
*/
var PR_STRING = 'str';
/**
* token style for a keyword
* @const
*/
var PR_KEYWORD = 'kwd';
/**
* token style for a comment
* @const
*/
var PR_COMMENT = 'com';
/**
* token style for a type
* @const
*/
var PR_TYPE = 'typ';
/**
* token style for a literal value. e.g. 1, null, true.
* @const
*/
var PR_LITERAL = 'lit';
/**
* token style for a punctuation string.
* @const
*/
var PR_PUNCTUATION = 'pun';
/**
* token style for plain text.
* @const
*/
var PR_PLAIN = 'pln';
/**
* token style for an sgml tag.
* @const
*/
var PR_TAG = 'tag';
/**
* token style for a markup declaration such as a DOCTYPE.
* @const
*/
var PR_DECLARATION = 'dec';
/**
* token style for embedded source.
* @const
*/
var PR_SOURCE = 'src';
/**
* token style for an sgml attribute name.
* @const
*/
var PR_ATTRIB_NAME = 'atn';
/**
* token style for an sgml attribute value.
* @const
*/
var PR_ATTRIB_VALUE = 'atv';
/**
* A class that indicates a section of markup that is not code, e.g. to allow
* embedding of line numbers within code listings.
* @const
*/
var PR_NOCODE = 'nocode';
// Regex pattern below is automatically generated by regexpPrecederPatterns.pl
// Do not modify, your changes will be erased.
// CAVEAT: this does not properly handle the case where a regular
// expression immediately follows another since a regular expression may
// have flags for case-sensitivity and the like. Having regexp tokens
// adjacent is not valid in any language I'm aware of, so I'm punting.
// TODO: maybe style special characters inside a regexp as punctuation.
/**
* A set of tokens that can precede a regular expression literal in
* javascript
* http://web.archive.org/web/20070717142515/http://www.mozilla.org/js/language/js20/rationale/syntax.html
* has the full list, but I've removed ones that might be problematic when
* seen in languages that don't support regular expression literals.
*
* Specifically, I've removed any keywords that can't precede a regexp
* literal in a syntactically legal javascript program, and I've removed the
* "in" keyword since it's not a keyword in many languages, and might be used
* as a count of inches.
*
* The link above does not accurately describe EcmaScript rules since
* it fails to distinguish between (a=++/b/i) and (a++/b/i) but it works
* very well in practice.
*
* @private
* @const
*/
var REGEXP_PRECEDER_PATTERN = '(?:^^\\.?|[+-]|[!=]=?=?|\\#|%=?|&&?=?|\\(|\\*=?|[+\\-]=|->|\\/=?|::?|<<?=?|>>?>?=?|,|;|\\?|@|\\[|~|{|\\^\\^?=?|\\|\\|?=?|break|case|continue|delete|do|else|finally|instanceof|return|throw|try|typeof)\\s*';
/**
* Given a group of {@link RegExp}s, returns a {@code RegExp} that globally
* matches the union of the sets of strings matched by the input RegExp.
* Since it matches globally, if the input strings have a start-of-input
* anchor (/^.../), it is ignored for the purposes of unioning.
* @param {Array.<RegExp>} regexs non multiline, non-global regexs.
* @return {RegExp} a global regex.
*/
function combinePrefixPatterns(regexs) {
var capturedGroupIndex = 0;
var needToFoldCase = false;
var ignoreCase = false;
for (var i = 0, n = regexs.length; i < n; ++i) {
var regex = regexs[i];
if (regex.ignoreCase) {
ignoreCase = true;
} else if (/[a-z]/i.test(regex.source.replace(
/\\u[0-9a-f]{4}|\\x[0-9a-f]{2}|\\[^ux]/gi, ''))) {
needToFoldCase = true;
ignoreCase = false;
break;
}
}
var escapeCharToCodeUnit = {
'b': 8,
't': 9,
'n': 0xa,
'v': 0xb,
'f': 0xc,
'r': 0xd
};
function decodeEscape(charsetPart) {
var cc0 = charsetPart.charCodeAt(0);
if (cc0 !== 92 /* \\ */) {
return cc0;
}
var c1 = charsetPart.charAt(1);
cc0 = escapeCharToCodeUnit[c1];
if (cc0) {
return cc0;
} else if ('0' <= c1 && c1 <= '7') {
return parseInt(charsetPart.substring(1), 8);
} else if (c1 === 'u' || c1 === 'x') {
return parseInt(charsetPart.substring(2), 16);
} else {
return charsetPart.charCodeAt(1);
}
}
function encodeEscape(charCode) {
if (charCode < 0x20) {
return (charCode < 0x10 ? '\\x0' : '\\x') + charCode.toString(16);
}
var ch = String.fromCharCode(charCode);
return (ch === '\\' || ch === '-' || ch === ']' || ch === '^')
? "\\" + ch : ch;
}
function caseFoldCharset(charSet) {
var charsetParts = charSet.substring(1, charSet.length - 1).match(
new RegExp(
'\\\\u[0-9A-Fa-f]{4}'
+ '|\\\\x[0-9A-Fa-f]{2}'
+ '|\\\\[0-3][0-7]{0,2}'
+ '|\\\\[0-7]{1,2}'
+ '|\\\\[\\s\\S]'
+ '|-'
+ '|[^-\\\\]',
'g'));
var ranges = [];
var inverse = charsetParts[0] === '^';
var out = ['['];
if (inverse) { out.push('^'); }
for (var i = inverse ? 1 : 0, n = charsetParts.length; i < n; ++i) {
var p = charsetParts[i];
if (/\\[bdsw]/i.test(p)) { // Don't muck with named groups.
out.push(p);
} else {
var start = decodeEscape(p);
var end;
if (i + 2 < n && '-' === charsetParts[i + 1]) {
end = decodeEscape(charsetParts[i + 2]);
i += 2;
} else {
end = start;
}
ranges.push([start, end]);
// If the range might intersect letters, then expand it.
// This case handling is too simplistic.
// It does not deal with non-latin case folding.
// It works for latin source code identifiers though.
if (!(end < 65 || start > 122)) {
if (!(end < 65 || start > 90)) {
ranges.push([Math.max(65, start) | 32, Math.min(end, 90) | 32]);
}
if (!(end < 97 || start > 122)) {
ranges.push([Math.max(97, start) & ~32, Math.min(end, 122) & ~32]);
}
}
}
}
// [[1, 10], [3, 4], [8, 12], [14, 14], [16, 16], [17, 17]]
// -> [[1, 12], [14, 14], [16, 17]]
ranges.sort(function (a, b) { return (a[0] - b[0]) || (b[1] - a[1]); });
var consolidatedRanges = [];
var lastRange = [];
for (var i = 0; i < ranges.length; ++i) {
var range = ranges[i];
if (range[0] <= lastRange[1] + 1) {
lastRange[1] = Math.max(lastRange[1], range[1]);
} else {
consolidatedRanges.push(lastRange = range);
}
}
for (var i = 0; i < consolidatedRanges.length; ++i) {
var range = consolidatedRanges[i];
out.push(encodeEscape(range[0]));
if (range[1] > range[0]) {
if (range[1] + 1 > range[0]) { out.push('-'); }
out.push(encodeEscape(range[1]));
}
}
out.push(']');
return out.join('');
}
function allowAnywhereFoldCaseAndRenumberGroups(regex) {
// Split into character sets, escape sequences, punctuation strings
// like ('(', '(?:', ')', '^'), and runs of characters that do not
// include any of the above.
var parts = regex.source.match(
new RegExp(
'(?:'
+ '\\[(?:[^\\x5C\\x5D]|\\\\[\\s\\S])*\\]' // a character set
+ '|\\\\u[A-Fa-f0-9]{4}' // a unicode escape
+ '|\\\\x[A-Fa-f0-9]{2}' // a hex escape
+ '|\\\\[0-9]+' // a back-reference or octal escape
+ '|\\\\[^ux0-9]' // other escape sequence
+ '|\\(\\?[:!=]' // start of a non-capturing group
+ '|[\\(\\)\\^]' // start/end of a group, or line start
+ '|[^\\x5B\\x5C\\(\\)\\^]+' // run of other characters
+ ')',
'g'));
var n = parts.length;
// Maps captured group numbers to the number they will occupy in
// the output or to -1 if that has not been determined, or to
// undefined if they need not be capturing in the output.
var capturedGroups = [];
// Walk over and identify back references to build the capturedGroups
// mapping.
for (var i = 0, groupIndex = 0; i < n; ++i) {
var p = parts[i];
if (p === '(') {
// groups are 1-indexed, so max group index is count of '('
++groupIndex;
} else if ('\\' === p.charAt(0)) {
var decimalValue = +p.substring(1);
if (decimalValue) {
if (decimalValue <= groupIndex) {
capturedGroups[decimalValue] = -1;
} else {
// Replace with an unambiguous escape sequence so that
// an octal escape sequence does not turn into a backreference
// to a capturing group from an earlier regex.
parts[i] = encodeEscape(decimalValue);
}
}
}
}
// Renumber groups and reduce capturing groups to non-capturing groups
// where possible.
for (var i = 1; i < capturedGroups.length; ++i) {
if (-1 === capturedGroups[i]) {
capturedGroups[i] = ++capturedGroupIndex;
}
}
for (var i = 0, groupIndex = 0; i < n; ++i) {
var p = parts[i];
if (p === '(') {
++groupIndex;
if (!capturedGroups[groupIndex]) {
parts[i] = '(?:';
}
} else if ('\\' === p.charAt(0)) {
var decimalValue = +p.substring(1);
if (decimalValue && decimalValue <= groupIndex) {
parts[i] = '\\' + capturedGroups[decimalValue];
}
}
}
// Remove any prefix anchors so that the output will match anywhere.
// ^^ really does mean an anchored match though.
for (var i = 0; i < n; ++i) {
if ('^' === parts[i] && '^' !== parts[i + 1]) { parts[i] = ''; }
}
// Expand letters to groups to handle mixing of case-sensitive and
// case-insensitive patterns if necessary.
if (regex.ignoreCase && needToFoldCase) {
for (var i = 0; i < n; ++i) {
var p = parts[i];
var ch0 = p.charAt(0);
if (p.length >= 2 && ch0 === '[') {
parts[i] = caseFoldCharset(p);
} else if (ch0 !== '\\') {
// TODO: handle letters in numeric escapes.
parts[i] = p.replace(
/[a-zA-Z]/g,
function (ch) {
var cc = ch.charCodeAt(0);
return '[' + String.fromCharCode(cc & ~32, cc | 32) + ']';
});
}
}
}
return parts.join('');
}
var rewritten = [];
for (var i = 0, n = regexs.length; i < n; ++i) {
var regex = regexs[i];
if (regex.global || regex.multiline) { throw new Error('' + regex); }
rewritten.push(
'(?:' + allowAnywhereFoldCaseAndRenumberGroups(regex) + ')');
}
return new RegExp(rewritten.join('|'), ignoreCase ? 'gi' : 'g');
}
/**
* Split markup into a string of source code and an array mapping ranges in
* that string to the text nodes in which they appear.
*
* <p>
* The HTML DOM structure:</p>
* <pre>
* (Element "p"
* (Element "b"
* (Text "print ")) ; #1
* (Text "'Hello '") ; #2
* (Element "br") ; #3
* (Text " + 'World';")) ; #4
* </pre>
* <p>
* corresponds to the HTML
* {@code <p><b>print </b>'Hello '<br> + 'World';</p>}.</p>
*
* <p>
* It will produce the output:</p>
* <pre>
* {
* sourceCode: "print 'Hello '\n + 'World';",
* // 1 2
* // 012345678901234 5678901234567
* spans: [0, #1, 6, #2, 14, #3, 15, #4]
* }
* </pre>
* <p>
* where #1 is a reference to the {@code "print "} text node above, and so
* on for the other text nodes.
* </p>
*
* <p>
* The {@code} spans array is an array of pairs. Even elements are the start
* indices of substrings, and odd elements are the text nodes (or BR elements)
* that contain the text for those substrings.
* Substrings continue until the next index or the end of the source.
* </p>
*
* @param {Node} node an HTML DOM subtree containing source-code.
* @param {boolean|number} isPreformatted truthy if white-space in
* text nodes should be considered significant.
* @return {SourceSpansT} source code and the nodes in which they occur.
*/
function extractSourceSpans(node, isPreformatted) {
var nocode = /(?:^|\s)nocode(?:\s|$)/;
var chunks = [];
var length = 0;
var spans = [];
var k = 0;
function walk(node) {
var type = node.nodeType;
if (type == 1) { // Element
if (nocode.test(node.className)) { return; }
for (var child = node.firstChild; child; child = child.nextSibling) {
walk(child);
}
var nodeName = node.nodeName.toLowerCase();
if ('br' === nodeName || 'li' === nodeName) {
chunks[k] = '\n';
spans[k << 1] = length++;
spans[(k++ << 1) | 1] = node;
}
} else if (type == 3 || type == 4) { // Text
var text = node.nodeValue;
if (text.length) {
if (!isPreformatted) {
text = text.replace(/[ \t\r\n]+/g, ' ');
} else {
text = text.replace(/\r\n?/g, '\n'); // Normalize newlines.
}
// TODO: handle tabs here?
chunks[k] = text;
spans[k << 1] = length;
length += text.length;
spans[(k++ << 1) | 1] = node;
}
}
}
walk(node);
return {
sourceCode: chunks.join('').replace(/\n$/, ''),
spans: spans
};
}
/**
* Apply the given language handler to sourceCode and add the resulting
* decorations to out.
* @param {!Element} sourceNode
* @param {number} basePos the index of sourceCode within the chunk of source
* whose decorations are already present on out.
* @param {string} sourceCode
* @param {function(JobT)} langHandler
* @param {DecorationsT} out
*/
function appendDecorations(
sourceNode, basePos, sourceCode, langHandler, out) {
if (!sourceCode) { return; }
/** @type {JobT} */
var job = {
sourceNode: sourceNode,
pre: 1,
langExtension: null,
numberLines: null,
sourceCode: sourceCode,
spans: null,
basePos: basePos,
decorations: null
};
langHandler(job);
out.push.apply(out, job.decorations);
}
var notWs = /\S/;
/**
* Given an element, if it contains only one child element and any text nodes
* it contains contain only space characters, return the sole child element.
* Otherwise returns undefined.
* <p>
* This is meant to return the CODE element in {@code <pre><code ...>} when
* there is a single child element that contains all the non-space textual
* content, but not to return anything where there are multiple child elements
* as in {@code <pre><code>...</code><code>...</code></pre>} or when there
* is textual content.
*/
function childContentWrapper(element) {
var wrapper = undefined;
for (var c = element.firstChild; c; c = c.nextSibling) {
var type = c.nodeType;
wrapper = (type === 1) // Element Node
? (wrapper ? element : c)
: (type === 3) // Text Node
? (notWs.test(c.nodeValue) ? element : wrapper)
: wrapper;
}
return wrapper === element ? undefined : wrapper;
}
/** Given triples of [style, pattern, context] returns a lexing function,
* The lexing function interprets the patterns to find token boundaries and
* returns a decoration list of the form
* [index_0, style_0, index_1, style_1, ..., index_n, style_n]
* where index_n is an index into the sourceCode, and style_n is a style
* constant like PR_PLAIN. index_n-1 <= index_n, and style_n-1 applies to
* all characters in sourceCode[index_n-1:index_n].
*
* The stylePatterns is a list whose elements have the form
* [style : string, pattern : RegExp, DEPRECATED, shortcut : string].
*
* Style is a style constant like PR_PLAIN, or can be a string of the
* form 'lang-FOO', where FOO is a language extension describing the
* language of the portion of the token in $1 after pattern executes.
* E.g., if style is 'lang-lisp', and group 1 contains the text
* '(hello (world))', then that portion of the token will be passed to the
* registered lisp handler for formatting.
* The text before and after group 1 will be restyled using this decorator
* so decorators should take care that this doesn't result in infinite
* recursion. For example, the HTML lexer rule for SCRIPT elements looks
* something like ['lang-js', /<[s]cript>(.+?)<\/script>/]. This may match
* '<script>foo()<\/script>', which would cause the current decorator to
* be called with '<script>' which would not match the same rule since
* group 1 must not be empty, so it would be instead styled as PR_TAG by
* the generic tag rule. The handler registered for the 'js' extension would
* then be called with 'foo()', and finally, the current decorator would
* be called with '<\/script>' which would not match the original rule and
* so the generic tag rule would identify it as a tag.
*
* Pattern must only match prefixes, and if it matches a prefix, then that
* match is considered a token with the same style.
*
* Context is applied to the last non-whitespace, non-comment token
* recognized.
*
* Shortcut is an optional string of characters, any of which, if the first
* character, gurantee that this pattern and only this pattern matches.
*
* @param {Array} shortcutStylePatterns patterns that always start with
* a known character. Must have a shortcut string.
* @param {Array} fallthroughStylePatterns patterns that will be tried in
* order if the shortcut ones fail. May have shortcuts.
*
* @return {function (JobT)} a function that takes an undecorated job and
* attaches a list of decorations.
*/
function createSimpleLexer(shortcutStylePatterns, fallthroughStylePatterns) {
var shortcuts = {};
var tokenizer;
(function () {
var allPatterns = shortcutStylePatterns.concat(fallthroughStylePatterns);
var allRegexs = [];
var regexKeys = {};
for (var i = 0, n = allPatterns.length; i < n; ++i) {
var patternParts = allPatterns[i];
var shortcutChars = patternParts[3];
if (shortcutChars) {
for (var c = shortcutChars.length; --c >= 0;) {
shortcuts[shortcutChars.charAt(c)] = patternParts;
}
}
var regex = patternParts[1];
var k = '' + regex;
if (!regexKeys.hasOwnProperty(k)) {
allRegexs.push(regex);
regexKeys[k] = null;
}
}
allRegexs.push(/[\0-\uffff]/);
tokenizer = combinePrefixPatterns(allRegexs);
})();
var nPatterns = fallthroughStylePatterns.length;
/**
* Lexes job.sourceCode and attaches an output array job.decorations of
* style classes preceded by the position at which they start in
* job.sourceCode in order.
*
* @type{function (JobT)}
*/
var decorate = function (job) {
var sourceCode = job.sourceCode, basePos = job.basePos;
var sourceNode = job.sourceNode;
/** Even entries are positions in source in ascending order. Odd enties
* are style markers (e.g., PR_COMMENT) that run from that position until
* the end.
* @type {DecorationsT}
*/
var decorations = [basePos, PR_PLAIN];
var pos = 0; // index into sourceCode
var tokens = sourceCode.match(tokenizer) || [];
var styleCache = {};
for (var ti = 0, nTokens = tokens.length; ti < nTokens; ++ti) {
var token = tokens[ti];
var style = styleCache[token];
var match = void 0;
var isEmbedded;
if (typeof style === 'string') {
isEmbedded = false;
} else {
var patternParts = shortcuts[token.charAt(0)];
if (patternParts) {
match = token.match(patternParts[1]);
style = patternParts[0];
} else {
for (var i = 0; i < nPatterns; ++i) {
patternParts = fallthroughStylePatterns[i];
match = token.match(patternParts[1]);
if (match) {
style = patternParts[0];
break;
}
}
if (!match) { // make sure that we make progress
style = PR_PLAIN;
}
}
isEmbedded = style.length >= 5 && 'lang-' === style.substring(0, 5);
if (isEmbedded && !(match && typeof match[1] === 'string')) {
isEmbedded = false;
style = PR_SOURCE;
}
if (!isEmbedded) { styleCache[token] = style; }
}
var tokenStart = pos;
pos += token.length;
if (!isEmbedded) {
decorations.push(basePos + tokenStart, style);
} else { // Treat group 1 as an embedded block of source code.
var embeddedSource = match[1];
var embeddedSourceStart = token.indexOf(embeddedSource);
var embeddedSourceEnd = embeddedSourceStart + embeddedSource.length;
if (match[2]) {
// If embeddedSource can be blank, then it would match at the
// beginning which would cause us to infinitely recurse on the
// entire token, so we catch the right context in match[2].
embeddedSourceEnd = token.length - match[2].length;
embeddedSourceStart = embeddedSourceEnd - embeddedSource.length;
}
var lang = style.substring(5);
// Decorate the left of the embedded source
appendDecorations(
sourceNode,
basePos + tokenStart,
token.substring(0, embeddedSourceStart),
decorate, decorations);
// Decorate the embedded source
appendDecorations(
sourceNode,
basePos + tokenStart + embeddedSourceStart,
embeddedSource,
langHandlerForExtension(lang, embeddedSource),
decorations);
// Decorate the right of the embedded section
appendDecorations(
sourceNode,
basePos + tokenStart + embeddedSourceEnd,
token.substring(embeddedSourceEnd),
decorate, decorations);
}
}
job.decorations = decorations;
};
return decorate;
}
/** returns a function that produces a list of decorations from source text.
*
* This code treats ", ', and ` as string delimiters, and \ as a string
* escape. It does not recognize perl's qq() style strings.
* It has no special handling for double delimiter escapes as in basic, or
* the tripled delimiters used in python, but should work on those regardless
* although in those cases a single string literal may be broken up into
* multiple adjacent string literals.
*
* It recognizes C, C++, and shell style comments.
*
* @param {Object} options a set of optional parameters.
* @return {function (JobT)} a function that examines the source code
* in the input job and builds a decoration list which it attaches to
* the job.
*/
function sourceDecorator(options) {
var shortcutStylePatterns = [], fallthroughStylePatterns = [];
if (options['tripleQuotedStrings']) {
// '''multi-line-string''', 'single-line-string', and double-quoted
shortcutStylePatterns.push(
[PR_STRING, /^(?:\'\'\'(?:[^\'\\]|\\[\s\S]|\'{1,2}(?=[^\']))*(?:\'\'\'|$)|\"\"\"(?:[^\"\\]|\\[\s\S]|\"{1,2}(?=[^\"]))*(?:\"\"\"|$)|\'(?:[^\\\']|\\[\s\S])*(?:\'|$)|\"(?:[^\\\"]|\\[\s\S])*(?:\"|$))/,
null, '\'"']);
} else if (options['multiLineStrings']) {
// 'multi-line-string', "multi-line-string"
shortcutStylePatterns.push(
[PR_STRING, /^(?:\'(?:[^\\\']|\\[\s\S])*(?:\'|$)|\"(?:[^\\\"]|\\[\s\S])*(?:\"|$)|\`(?:[^\\\`]|\\[\s\S])*(?:\`|$))/,
null, '\'"`']);
} else {
// 'single-line-string', "single-line-string"
shortcutStylePatterns.push(
[PR_STRING,
/^(?:\'(?:[^\\\'\r\n]|\\.)*(?:\'|$)|\"(?:[^\\\"\r\n]|\\.)*(?:\"|$))/,
null, '"\'']);
}
if (options['verbatimStrings']) {
// verbatim-string-literal production from the C# grammar. See issue 93.
fallthroughStylePatterns.push(
[PR_STRING, /^@\"(?:[^\"]|\"\")*(?:\"|$)/, null]);
}
var hc = options['hashComments'];
if (hc) {
if (options['cStyleComments']) {
if (hc > 1) { // multiline hash comments
shortcutStylePatterns.push(
[PR_COMMENT, /^#(?:##(?:[^#]|#(?!##))*(?:###|$)|.*)/, null, '#']);
} else {
// Stop C preprocessor declarations at an unclosed open comment
shortcutStylePatterns.push(
[PR_COMMENT, /^#(?:(?:define|e(?:l|nd)if|else|error|ifn?def|include|line|pragma|undef|warning)\b|[^\r\n]*)/,
null, '#']);
}
// #include <stdio.h>
fallthroughStylePatterns.push(
[PR_STRING,
/^<(?:(?:(?:\.\.\/)*|\/?)(?:[\w-]+(?:\/[\w-]+)+)?[\w-]+\.h(?:h|pp|\+\+)?|[a-z]\w*)>/,
null]);
} else {
shortcutStylePatterns.push([PR_COMMENT, /^#[^\r\n]*/, null, '#']);
}
}
if (options['cStyleComments']) {
fallthroughStylePatterns.push([PR_COMMENT, /^\/\/[^\r\n]*/, null]);
fallthroughStylePatterns.push(
[PR_COMMENT, /^\/\*[\s\S]*?(?:\*\/|$)/, null]);
}
var regexLiterals = options['regexLiterals'];
if (regexLiterals) {
/**
* @const
*/
var regexExcls = regexLiterals > 1
? '' // Multiline regex literals
: '\n\r';
/**
* @const
*/
var regexAny = regexExcls ? '.' : '[\\S\\s]';
/**
* @const
*/
var REGEX_LITERAL = (
// A regular expression literal starts with a slash that is
// not followed by * or / so that it is not confused with
// comments.
'/(?=[^/*' + regexExcls + '])'
// and then contains any number of raw characters,
+ '(?:[^/\\x5B\\x5C' + regexExcls + ']'
// escape sequences (\x5C),
+ '|\\x5C' + regexAny
// or non-nesting character sets (\x5B\x5D);
+ '|\\x5B(?:[^\\x5C\\x5D' + regexExcls + ']'
+ '|\\x5C' + regexAny + ')*(?:\\x5D|$))+'
// finally closed by a /.
+ '/');
fallthroughStylePatterns.push(
['lang-regex',
RegExp('^' + REGEXP_PRECEDER_PATTERN + '(' + REGEX_LITERAL + ')')
]);
}
var types = options['types'];
if (types) {
fallthroughStylePatterns.push([PR_TYPE, types]);
}
var keywords = ("" + options['keywords']).replace(/^ | $/g, '');
if (keywords.length) {
fallthroughStylePatterns.push(
[PR_KEYWORD,
new RegExp('^(?:' + keywords.replace(/[\s,]+/g, '|') + ')\\b'),
null]);
}
shortcutStylePatterns.push([PR_PLAIN, /^\s+/, null, ' \r\n\t\xA0']);
var punctuation =
// The Bash man page says
// A word is a sequence of characters considered as a single
// unit by GRUB. Words are separated by metacharacters,
// which are the following plus space, tab, and newline: { }
// | & $ ; < >
// ...
// A word beginning with # causes that word and all remaining
// characters on that line to be ignored.
// which means that only a '#' after /(?:^|[{}|&$;<>\s])/ starts a
// comment but empirically
// $ echo {#}
// {#}
// $ echo \$#
// $#
// $ echo }#
// }#
// so /(?:^|[|&;<>\s])/ is more appropriate.
// http://gcc.gnu.org/onlinedocs/gcc-2.95.3/cpp_1.html#SEC3
// suggests that this definition is compatible with a
// default mode that tries to use a single token definition
// to recognize both bash/python style comments and C
// preprocessor directives.
// This definition of punctuation does not include # in the list of
// follow-on exclusions, so # will not be broken before if preceeded
// by a punctuation character. We could try to exclude # after
// [|&;<>] but that doesn't seem to cause many major problems.
// If that does turn out to be a problem, we should change the below
// when hc is truthy to include # in the run of punctuation characters
// only when not followint [|&;<>].
'^.[^\\s\\w.$@\'"`/\\\\]*';
if (options['regexLiterals']) {
punctuation += '(?!\s*\/)';
}
fallthroughStylePatterns.push(
// TODO(mikesamuel): recognize non-latin letters and numerals in idents
[PR_LITERAL, /^@[a-z_$][a-z_$@0-9]*/i, null],
[PR_TYPE, /^(?:[@_]?[A-Z]+[a-z][A-Za-z_$@0-9]*|\w+_t\b)/, null],
[PR_PLAIN, /^[a-z_$][a-z_$@0-9]*/i, null],
[PR_LITERAL,
new RegExp(
'^(?:'