-
Notifications
You must be signed in to change notification settings - Fork 15
/
Lexer.scala
862 lines (826 loc) · 41.9 KB
/
Lexer.scala
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
/*
* Copyright 2020 Parsley Contributors <https://github.com/j-mie6/Parsley/graphs/contributors>
*
* SPDX-License-Identifier: BSD-3-Clause
*/
package parsley.token
import parsley.Parsley, Parsley.unit
import parsley.combinator.{between, eof, sepBy, sepBy1, skipMany}
import parsley.errors.combinator.{markAsToken, ErrorMethods}
import parsley.registers.Reg
import parsley.token.names.{ConcreteNames, LexemeNames}
import parsley.token.numeric.{LexemeCombined, LexemeInteger, LexemeReal,
SignedCombined, SignedInteger, SignedReal,
UnsignedCombined, UnsignedInteger, UnsignedReal}
import parsley.token.predicate.{Basic, CharPredicate, NotRequired, Unicode}
import parsley.token.symbol.{ConcreteSymbol, LexemeSymbol}
import parsley.token.text.{ConcreteCharacter, ConcreteString, EscapableCharacter, Escape, LexemeCharacter, LexemeString, RawCharacter}
import parsley.unicode.satisfy
import parsley.internal.deepembedding.singletons
/** This class is just to allow for a polymorphic apply for the lexeme versions of each lexing category
*
* It is designed to be extended by `Lexer#lexeme` only (or within the tests!)
*/
private [token] abstract class Lexeme {
def apply[A](p: Parsley[A]): Parsley[A]
}
// TODO: flatten out `numeric` and `text` (and `enclosing` and `separators`) for 5.0.0? wouldn't do much damage,
// we can use documentation tags to group them in the high-level docs again :)
/** This class provides a large selection of functionality concerned
* with lexing.
*
* This class provides lexing functionality to `parsley`, however
* it is guaranteed that nothing in this class is not implementable
* purely using `parsley`'s pre-existing functionality. These are
* regular parsers, but constructed in such a way that they create
* a clear and logical separation from the rest of the parser.
*
* The class is broken up into several internal "modules" that group
* together similar kinds of functionality. Importantly, the `lexemes`
* and `nonlexemes` objects separate the underlying token implementations
* based on whether or not they consume whitespace or not. Functionality
* is broadly duplicated across both of these modules: `lexemes` should
* be used by a wider parser, to ensure whitespace is handled uniformly;
* and `nonlexemes` should be used to define further composite tokens or
* in special circumstances where whitespace should not be consumed.
*
* It is possible that some of the implementations of
* parsers found within this class may have been hand-optimised for
* performance: care '''will''' have been taken to ensure these
* implementations precisely match the semantics of the originals.
*
* @define numeric
* This object contains lexing functionality relevant to the parsing
* of numbers. This is sub-divided into different categories:
*
* - integers (both signed and unsigned)
* - reals (signed only)
* - a combination of the two (signed and unsigned)
*
* These contain relevant functionality for the processing of
* decimal, hexadecimal, octal, and binary literals; or some
* mixed combination thereof (as specified by `desc.numericDesc`).
* Additionally, it is possible to ensure literals represent known
* sizes or precisions.
*
* @define text
* This object contains lexing functionality relevant to the parsing
* of text. This is sub-divided into different categories:
*
* - string literals (both with escapes and raw)
* - multi-line string literals (both with escapes and raw)
* - character literals
*
* These contain the relevant functionality required to specify the
* degree of unicode support for the underlying language, from
* ASCII to full UTF-16.
*
* @define symbol
* This object contains lexing functionality relevant to the parsing
* of atomic symbols.
*
* Symbols are characterised by their "unitness", that is, every parser
* inside returns `Unit`. This is because they all parse a specific
* known entity, and, as such, the result of the parse is irrelevant.
* These can be things such as reserved names, or small symbols like
* parentheses. This object also contains a means of creating new symbols
* as well as implicit conversions to allow for Scala's string literals to serve
* as symbols within a parser.
*
* @define names
* This object contains lexing functionality relevant to the parsing
* of names, which include operators or identifiers.
*
* The parsing of names is mostly concerned with finding the longest
* valid name that is not a reserved name, such as a hard keyword or
* a special operator.
*
* @define natural
* This is a collection of parsers concerned with handling unsigned (positive) integer literals.
*
* Natural numbers are described generally as follows:
* - '''`desc.numericDesc.literalBreakChar`''': determines whether or not it
* is legal to "break up" the digits within a literal, for example: is `1_000_000` allowed?
* If this is legal, describes what the break character is, and whether it can appear after
* a hexadecimal/octal/binary prefix
* - '''`desc.numericDesc.leadingZerosAllowed`''': determines whether or not it is
* possible to add extraneous zero digits onto the front of a number or not. In some languages,
* like C, this is disallowed, as numbers starting with `0` are octal numbers.
* - '''`desc.numericDesc.integerNumbersCanBe{Hexadecimal/Octal/Binary}`''': these flags
* control what kind of literals can appear within the `number` parser. Each type of literal
* can be individually parsed with its corresponding parser, regardless of the value of the
* flag
* - '''`desc.numericDesc.{hexadecimal/octal/binary}Leads`''': controls what character must
* follow a `0` when starting a number to change it from decimal into another base. This
* set may be empty, in which case the literal is described purely with leading zero (C style
* octals would set `octalLeads` to `Set.empty`)
*
* Additional to the parsing of decimal, hexadecimal, octal, and binary literals, each parser can
* be given a bit-width from 8- to 64-bit: this will check the parsed literal to ensure it is
* a legal literal of that size.
* @define integer
* This is a collection of parsers concerned with handling signed integer literals.
*
* Signed integer literals are an extension of unsigned integer literals with the following
* extra configuration:
* - '''`desc.numericDesc.positiveSign`''': describes whether or not literals are
* allowed to omit `+` for positive literals, must write a `+`, or can never write a `+`.
* @define real
* This is a collection of parsers concerned with handling signed real numbers (like floats and doubles).
*
* These literals consist of a (possibly optional) integer prefix, with at least one of a fractional component (with `.`)
* or an exponential component.
*
* Real numbers are an extension of signed integers with the following additional configuration:
* - '''`desc.numericDesc.leadingDotAllowed`''': determines whether a literal like `.0` would be considered legal
* - '''`desc.numericDesc.trailingDotAllowed`''': determines whether a literal like `0.` would be considered legal
* - '''`desc.numericDesc.realNumbersCanBe{Hexadecimal/Octal/Binary}`''': these flags control
* what kind of literals can appear within the `number` parser. Each type of literal
* may still be individually parsed with its corresponding parser, regardless of the value of
* the flag
* - '''`desc.numericDesc.{decimal/hexadecimal/octal/binary}ExponentDesc`''': describes how the
* exponential syntax works for each kind of base. If the syntax is legal, then this describes:
* which characters start it (classically, this would be `e` or `E` for decimals); whether or
* not it is compulsory for the literal (in Java and C, hexadecimal floats are ''only'' valid
* when they have an exponent attached); and whether or not a `+` sign is mandatory, optional,
* or illegal for positive exponents
*
* Additional to the parsing of decimal, hexadecimal, octal, and binary floating literals, each
* parser can be given a precision of IEEE 754 float or double. This can either be achieved by
* rounding to the nearest representable value, or by ensuring that the literal must be precisely
* representable as one of these numbers (which is defined as being one of binary, decimal
* or exact `float` and `double` values as described by Java)
*
* @define unsignedCombined
* This is a collection of parsers concerned with handling numeric literals that may either be
* unsigned integers ''or'' unsigned reals.
*
* There is no additional configuration offered over that found in `natural` or `real`.
*
* the bit-bounds and precision of the integer or real parts of the result can be specified
* in any pairing.
* @define signedCombined
* This is a collection of parsers concerned with handling numeric literals that may either be
* signed integers ''or'' signed reals.
*
* There is no additional configuration offered over that found in `integer` or `real`.
*
* the bit-bounds and precision of the integer or real parts of the result can be specified
* in any pairing.
*
* @define character
* This is a collection of parsers concerned with handling character literals.
*
* Character literals are described generally as follows:
* - '''`desc.textDesc.characterLiteralEnd`''': the character that starts and ends
* the literal (for example in many languages this is `'`)
* - '''`desc.textDesc.graphicCharacter`''': describes the legal characters that may appear
* in the literal directly. Usually, this excludes control characters and newlines,
* but permits most other things. Escape sequences can represent non-graphic
* characters
* - '''`desc.textDesc.escapeSequences`''': describes the legal escape sequences that
* that can appear in a character literal (for example `\n` or `\u000a`)
*
* Aside from the generic configuration, characters can be parsed in accordance with
* varying levels of unicode support, from ASCII-only to full UTF-16 characters. Parsers
* for each of four different vareties are exposed by this object.
* @define string
* This is a collection of parsers concerned with handling single-line string literals.
*
* String literals are described generally as follows:
* - '''`desc.textDesc.stringEnds`''': the sequence of characters that can begin or
* end a string literal. Regardless of which of these is used for a specific literal,
* the end of the literal ''must'' use the same sequence
* - '''`desc.textDesc.graphicCharacter`''': describes the legal characters that may appear
* in the literal directly. Usually, this excludes control characters and newlines,
* but permits most other things. Escape sequences can represent non-graphic
* characters for non-raw strings
* - '''`desc.textDesc.escapeSequences`''': describes the legal escape sequences that
* that can appear in a string literal (for example `\n` or `\u000a`)
* @define multiString
* This is a collection of parsers concerned with handling multi-line string literals.
*
* String literals are described generally as follows:
* - '''`desc.textDesc.multiStringEnds`''': the sequence of characters that can begin or
* end a multi-line string literal. Regardless of which of these is used for a specific literal,
* the end of the literal ''must'' use the same sequence
* - '''`desc.textDesc.graphicCharacter`''': describes the legal characters that may appear
* in the literal directly. Usually, this excludes control characters and newlines,
* but permits most other things. Escape sequences can represent non-graphic
* characters for non-raw strings
* - '''`desc.textDesc.escapeSequences`''': describes the legal escape sequences that
* that can appear in a string literal (for example `\n` or `\u000a`)
* @define raw this will be parsed without handling any escape sequences,
* this includes literal-end characters and the escape prefix
* (often `"` and `\` respectively)
*
* @constructor Builds a new lexer with a given description for the lexical structure as
* well as how error messages should be specialised.
* @param desc the configuration for the lexer, specifying the lexical
* rules of the grammar being parsed.
* @param errConfig the configuration for error messages generated within
* the lexer.
* @since 4.0.0
*/
@deprecatedInheritance("this class will be made final in 5.0.0", since = "4.2.0")
class Lexer(desc: descriptions.LexicalDesc, errConfig: errors.ErrorConfig) {
/** Builds a new lexer with a given description for the lexical structure of the language.
*
* @param desc the configuration for the lexer, specifying the lexical
* rules of the grammar/language being parsed.
* @since 4.0.0
*/
def this(desc: descriptions.LexicalDesc) = this(desc, new errors.ErrorConfig)
private val generic = new numeric.Generic(errConfig)
/** This object is concerned with ''lexemes'': these are tokens that are
* treated as "words", such that whitespace will be consumed after each
* has been parsed.
*
* Ideally, a wider parser should not be concerned with
* handling whitespace, as it is responsible for dealing with a stream
* of tokens. With parser combinators, however, it is usually not the
* case that there is a separate distinction between the parsing phase
* and the lexing phase. That said, it is good practice to establish
* a logical separation between the two worlds. As such, this object
* contains parsers that parse tokens, and these are whitespace-aware.
* This means that whitespace will be consumed '''after''' any of these
* parsers are parsed. It is not, however, required that whitespace be
* present.
*
* @since 4.0.0
*/
object lexeme extends Lexeme {
/** This combinator turns a non-lexeme parser into a lexeme one by
* ensuring whitespace is consumed after the parser.
*
* When using parser combinators, it is important to establish a
* consistent whitespace consumption scheme: ideally, there is no
* wasteful parsing, and whitespace consumption should not impact
* backtracking. This leads to a convention that whitespace must
* only be consumed ''after'' a token, and only once at the very
* start of the parser (see [[fully `fully`]]). When manually
* constructing tokens that are not supported by this lexer, use
* this combinator to ensure it also follows the whitespace convention.
*
* @param p the token parser to ensure consumes trailing whitespace.
* @since 4.0.0
*/
def apply[A](p: Parsley[A]): Parsley[A] = markAsToken(p) <* space.whiteSpace
/** $names
*
* @since 4.0.0
*/
val names: parsley.token.names.Names = new LexemeNames(nonlexeme.names, this)
/** $numeric
*
* @since 4.0.0
*/
object numeric {
private [Lexer] val _natural = new LexemeInteger(nonlexeme.numeric.natural, lexeme)
private [Lexer] val _integer = new LexemeInteger(nonlexeme.numeric.integer, lexeme)
private [Lexer] val _positiveReal = new LexemeReal(nonlexeme.numeric._positiveReal, lexeme, errConfig)
private [Lexer] val _real = new LexemeReal(nonlexeme.numeric.real, lexeme, errConfig)
private [Lexer] val _unsignedCombined = new LexemeCombined(nonlexeme.numeric.unsignedCombined, lexeme, errConfig)
private [Lexer] val _signedCombined = new LexemeCombined(nonlexeme.numeric.signedCombined, lexeme, errConfig)
/** $natural
*
* @since 4.0.0
* @note alias for [[natural `natural`]].
*/
// $COVERAGE-OFF$
def unsigned: parsley.token.numeric.Integer = natural
// $COVERAGE-ON$
/** $natural
*
* @since 4.0.0
*/
def natural: parsley.token.numeric.Integer = _natural
/** $integer
*
* @since 4.0.0
* @note alias for [[integer `integer`]]
* @see [[unsigned `unsigned`]] for a full description of signed integer configuration
*/
// $COVERAGE-OFF$
def signed: parsley.token.numeric.Integer = integer
// $COVERAGE-ON$
/** $integer
*
* @since 4.0.0
* @see [[natural `natural`]] for a full description of integer configuration
*/
def integer: parsley.token.numeric.Integer = _integer
/** $real
*
* @since 4.0.0
* @note alias for [[real `real`]]
* @see [[natural `natural`]] and [[integer `integer`]] for a full description of the configuration for the start of a real number
*/
// $COVERAGE-OFF$
def floating: parsley.token.numeric.Real = real
// $COVERAGE-ON$
/** $real
*
* @since 4.0.0
* @see [[natural `natural`]] and [[integer `integer`]] for a full description of the configuration for the start of a real number
*/
def real: parsley.token.numeric.Real = _real
/** $unsignedCombined
*
* @since 4.0.0
*/
def unsignedCombined: parsley.token.numeric.Combined = _unsignedCombined
/** $signedCombined
*
* @since 4.0.0
*/
def signedCombined: parsley.token.numeric.Combined = _signedCombined
}
/** $text
*
* @since 4.0.0
*/
object text {
private [Lexer] val _character = new LexemeCharacter(nonlexeme.text._character, lexeme)
private [Lexer] val _string = new LexemeString(nonlexeme.text._string, lexeme)
private [Lexer] val _rawString = new LexemeString(nonlexeme.text._rawString, lexeme)
private [Lexer] val _multiString = new LexemeString(nonlexeme.text._multiString, lexeme)
private [Lexer] val _rawMultiString = new LexemeString(nonlexeme.text._rawMultiString, lexeme)
/** $character
*
* @since 4.0.0
*/
def character: parsley.token.text.Character = _character
/** $string
*
* @since 4.0.0
*/
def string: parsley.token.text.String = _string
/** $string
*
* @note $raw
* @since 4.0.0
*/
def rawString: parsley.token.text.String = _rawString
/** $multiString
*
* @since 4.0.0
*/
def multiString: parsley.token.text.String = _multiString
/** $multiString
*
* @note $raw
* @since 4.0.0
*/
def rawMultiString: parsley.token.text.String = _rawMultiString
}
/** $symbol
*
* @since 4.0.0
*/
val symbol: parsley.token.symbol.Symbol = new LexemeSymbol(nonlexeme.symbol, this, errConfig)
/** This object contains helper combinators for parsing terms separated by
* common symbols.
*
* @since 4.0.0
*/
object separators {
/** This combinator parses '''zero''' or more occurrences of `p`, separated by semi-colons.
*
* Behaves just like `semiSep1`, except does not require an initial `p`, returning the empty list instead.
*
* @example {{{
* scala> ...
* scala> val stmts = lexer.lexeme.separators.semiSep(int)
* scala> stmts.parse("7; 3;2")
* val res0 = Success(List(7; 3; 2))
* scala> stmts.parse("")
* val res1 = Success(Nil)
* scala> stmts.parse("1")
* val res2 = Success(List(1))
* scala> stmts.parse("1; 2; ")
* val res3 = Failure(..) // no trailing semi-colon allowed
* }}}
*
* @param p the parser whose results are collected into a list.
* @return a parser that parses `p` delimited by semi-colons, returning the list of `p`'s results.
* @since 4.0.0
*/
def semiSep[A](p: Parsley[A]): Parsley[List[A]] = sepBy(p, symbol.semi)
/** This combinator parses '''one''' or more occurrences of `p`, separated by semi-colons.
*
* First parses a `p`. Then parses a semi-colon followed by `p` until there are no more semi-colons.
* The results of the `p`'s, `x,,1,,` through `x,,n,,`, are returned as `List(x,,1,,, .., x,,n,,)`.
* If `p` fails having consumed input, the whole parser fails. Requires at least
* one `p` to have been parsed.
*
* @example {{{
* scala> ...
* scala> val stmts = lexer.lexeme.separators.semiSep1(int)
* scala> stmts.parse("7; 3;2")
* val res0 = Success(List(7; 3; 2))
* scala> stmts.parse("")
* val res1 = Failure(..)
* scala> stmts.parse("1")
* val res2 = Success(List(1))
* scala> stmts.parse("1; 2; ")
* val res3 = Failure(..) // no trailing semi-colon allowed
* }}}
*
* @param p the parser whose results are collected into a list.
* @return a parser that parses `p` delimited by semi-colons, returning the list of `p`'s results.
* @since 4.0.0
*/
def semiSep1[A](p: Parsley[A]): Parsley[List[A]] = sepBy1(p, symbol.semi)
/** This combinator parses '''zero''' or more occurrences of `p`, separated by commas.
*
* Behaves just like `commaSep1`, except does not require an initial `p`, returning the empty list instead.
*
* @example {{{
* scala> ...
* scala> val stmts = lexer.lexeme.separators.commaSep(int)
* scala> stmts.parse("7, 3,2")
* val res0 = Success(List(7, 3, 2))
* scala> stmts.parse("")
* val res1 = Success(Nil)
* scala> stmts.parse("1")
* val res2 = Success(List(1))
* scala> stmts.parse("1, 2, ")
* val res3 = Failure(..) // no trailing comma allowed
* }}}
*
* @param p the parser whose results are collected into a list.
* @return a parser that parses `p` delimited by commas, returning the list of `p`'s results.
* @since 4.0.0
*/
def commaSep[A](p: Parsley[A]): Parsley[List[A]] = sepBy(p, symbol.comma)
/** This combinator parses '''one''' or more occurrences of `p`, separated by commas.
*
* First parses a `p`. Then parses a comma followed by `p` until there are no more commas.
* The results of the `p`'s, `x,,1,,` through `x,,n,,`, are returned as `List(x,,1,,, .., x,,n,,)`.
* If `p` fails having consumed input, the whole parser fails. Requires at least
* one `p` to have been parsed.
*
* @example {{{
* scala> ...
* scala> val stmts = lexer.lexeme.separators.commaSep1(int)
* scala> stmts.parse("7, 3,2")
* val res0 = Success(List(7, 3, 2))
* scala> stmts.parse("")
* val res1 = Failure(..)
* scala> stmts.parse("1")
* val res2 = Success(List(1))
* scala> stmts.parse("1, 2, ")
* val res3 = Failure(..) // no trailing comma allowed
* }}}
*
* @param p the parser whose results are collected into a list.
* @return a parser that parses `p` delimited by commas, returning the list of `p`'s results.
* @since 4.0.0
*/
def commaSep1[A](p: Parsley[A]): Parsley[List[A]] = sepBy1(p, symbol.comma)
}
/** This object contains helper combinators for parsing terms enclosed by
* common symbols.
*
* @since 4.0.0
*/
object enclosing {
/** This combinator parses a `p` enclosed within parentheses.
*
* First parse an open parenthesis, any whitespace, then parse, `p`, producing `x`. Finally, parse a closing parenthesis and any whitespace.
* If all three parts succeeded, then return `x`. If any of them failed, this combinator fails.
*
* @example {{{
* scala> ...
* scala> val p = lexer.nonlexeme.enclosing.parens(int)
* scala> p.parse("( 5)")
* val res0 = Success(5)
* scala> p.parse("(5")
* val res1 = Failure(...)
* scala> p.parse("5)")
* val res2 = Failure(...)
* }}}
*
* @param p the parser to parse between parentheses.
* @return a parser that reads an open parenthesis, then `p`, then a closing parenthesis and returns the result of `p`.
* @since 4.0.0
*/
def parens[A](p: =>Parsley[A]): Parsley[A] = enclosing(p, symbol.openParen, symbol.closingParen, "parentheses")
/** This combinator parses a `p` enclosed within braces.
*
* First parse an open brace, any whitespace, then parse, `p`, producing `x`. Finally, parse a closing brace and any whitespace.
* If all three parts succeeded, then return `x`. If any of them failed, this combinator fails.
*
* @example {{{
* scala> ...
* scala> val p = lexer.nonlexeme.enclosing.braces(int)
* scala> p.parse("{ 5}")
* val res0 = Success(5)
* scala> p.parse("{5")
* val res1 = Failure(...)
* scala> p.parse("5}")
* val res2 = Failure(...)
* }}}
*
* @param p the parser to parse between parentheses.
* @return a parser that reads an open brace, then `p`, then a closing brace and returns the result of `p`.
* @since 4.0.0
*/
def braces[A](p: =>Parsley[A]): Parsley[A] = enclosing(p, symbol.openBrace, symbol.closingBrace, "braces")
/** This combinator parses a `p` enclosed within angle brackets.
*
* First parse an open bracket, any whitespace, then parse, `p`, producing `x`. Finally, parse a closing bracket and any whitespace.
* If all three parts succeeded, then return `x`. If any of them failed, this combinator fails.
*
* @example {{{
* scala> ...
* scala> val p = lexer.nonlexeme.enclosing.brackets(int)
* scala> p.parse("< 5>")
* val res0 = Success(5)
* scala> p.parse("<5")
* val res1 = Failure(...)
* scala> p.parse("5>")
* val res2 = Failure(...)
* }}}
*
* @param p the parser to parse between parentheses.
* @return a parser that reads an open bracket, then `p`, then a closing bracket and returns the result of `p`.
* @since 4.0.0
*/
def angles[A](p: =>Parsley[A]): Parsley[A] = enclosing(p, symbol.openAngle, symbol.closingAngle, "angle brackets")
/** This combinator parses a `p` enclosed within square brackets.
*
* First parse an open bracket, any whitespace, then parse, `p`, producing `x`. Finally, parse a closing bracket and any whitespace.
* If all three parts succeeded, then return `x`. If any of them failed, this combinator fails.
*
* @example {{{
* scala> ...
* scala> val p = lexer.nonlexeme.enclosing.brackets(int)
* scala> p.parse("[ 5]")
* val res0 = Success(5)
* scala> p.parse("[5")
* val res1 = Failure(...)
* scala> p.parse("5]")
* val res2 = Failure(...)
* }}}
*
* @param p the parser to parse between parentheses.
* @return a parser that reads an open bracket, then `p`, then a closing bracket and returns the result of `p`.
* @since 4.0.0
*/
def brackets[A](p: =>Parsley[A]): Parsley[A] = enclosing(p, symbol.openSquare, symbol.closingSquare, "square brackets")
private def enclosing[A](p: =>Parsley[A], open: Parsley[Unit], close: Parsley[Unit], plural: String) =
between(open, close.explain(s"unclosed $plural"), p)
}
}
/** This object is concerned with ''non-lexemes'': these are tokens that
* do not give any special treatment to whitespace.
*
* Whilst the functionality in `lexeme` is ''strongly'' recommended for
* wider use in a parser, the functionality here may be useful for more
* specialised use-cases. In particular, these may for the building blocks
* for more complex tokens (where whitespace is not allowed between them, say),
* in which case these compound tokens can be turned into lexemes manually.
* For example, the lexer does not have configuration for trailing specifiers
* on numeric literals (like, `1024L` in Scala, say): the desired numeric
* literal parser could be extended with this functionality ''before'' whitespace
* is consumed by using the variant found in this object.
*
* Alternatively, these tokens can be used for ''lexical extraction'', which
* can be performed by the [[parsley.errors.ErrorBuilder `ErrorBuilder`]]
* typeclass: this can be used to try and extract tokens from the input stream
* when an error happens, to provide a more informative error. In this case,
* it is desirable to ''not'' consume whitespace after the token to keep the
* error tight and precise.
*
* @since 4.0.0
*/
object nonlexeme {
/** $names
*
* @since 4.0.0
*/
val names: parsley.token.names.Names = new ConcreteNames(desc.nameDesc, desc.symbolDesc, errConfig)
/** $numeric
*
* @since 4.0.0
*/
object numeric {
private [Lexer] val _natural = new UnsignedInteger(desc.numericDesc, errConfig, generic)
private [Lexer] val _integer = new SignedInteger(desc.numericDesc, _natural, errConfig)
private [Lexer] val _positiveReal = new UnsignedReal(desc.numericDesc, _natural, errConfig, generic)
private [Lexer] val _real = new SignedReal(desc.numericDesc, _positiveReal, errConfig)
private [Lexer] val _unsignedCombined = new UnsignedCombined(desc.numericDesc, _integer, _positiveReal, errConfig)
private [Lexer] val _signedCombined = new SignedCombined(desc.numericDesc, _unsignedCombined, errConfig)
/** $natural
*
* @since 4.0.0
* @note alias for [[natural `natural`]].
*/
// $COVERAGE-OFF$
def unsigned: parsley.token.numeric.Integer = _natural
// $COVERAGE-ON$
/** $natural
*
* @since 4.0.0
*/
def natural: parsley.token.numeric.Integer = _natural
/** $integer
*
* @since 4.0.0
* @note alias for [[integer `integer`]]
* @see [[unsigned `unsigned`]] for a full description of signed integer configuration
*/
// $COVERAGE-OFF$
def signed: parsley.token.numeric.Integer = _integer
// $COVERAGE-ON$
/** $integer
*
* @since 4.0.0
* @see [[natural `natural`]] for a full description of integer configuration
*/
def integer: parsley.token.numeric.Integer = _integer
/** $real
*
* @since 4.0.0
* @note alias for [[real `real`]]
* @see [[natural `natural`]] and [[integer `integer`]] for a full description of the configuration for the start of a real number
*/
// $COVERAGE-OFF$
def floating: parsley.token.numeric.Real = real
// $COVERAGE-ON$
/** $real
*
* @since 4.0.0
* @see [[natural `natural`]] and [[integer `integer`]] for a full description of the configuration for the start of a real number
*/
def real: parsley.token.numeric.Real = _real
/** $unsignedCombined
*
* @since 4.0.0
*/
def unsignedCombined: parsley.token.numeric.Combined = _unsignedCombined
/** $signedCombined
*
* @since 4.0.0
*/
def signedCombined: parsley.token.numeric.Combined = _signedCombined
}
/** $text
*
* @since 4.0.0
*/
object text {
private val escapes = new Escape(desc.textDesc.escapeSequences, errConfig, generic)
private val escapeChar = new EscapableCharacter(desc.textDesc.escapeSequences, escapes, space.space, errConfig)
private val rawChar = new RawCharacter(errConfig)
private [Lexer] val _character: parsley.token.text.Character = new ConcreteCharacter(desc.textDesc, escapes, errConfig)
private [Lexer] val _string = new ConcreteString(desc.textDesc.stringEnds, escapeChar, desc.textDesc.graphicCharacter, false, errConfig)
private [Lexer] val _rawString = new ConcreteString(desc.textDesc.stringEnds, rawChar, desc.textDesc.graphicCharacter, false, errConfig)
private [Lexer] val _multiString = new ConcreteString(desc.textDesc.multiStringEnds, escapeChar, desc.textDesc.graphicCharacter, true, errConfig)
private [Lexer] val _rawMultiString = new ConcreteString(desc.textDesc.multiStringEnds, rawChar, desc.textDesc.graphicCharacter, true, errConfig)
/** $character
*
* @since 4.0.0
*/
def character: parsley.token.text.Character = _character
/** $string
*
* @since 4.0.0
*/
def string: parsley.token.text.String = _string
/** $string
*
* @note $raw
* @since 4.0.0
*/
def rawString: parsley.token.text.String = _rawString
/** $multiString
*
* @since 4.0.0
*/
def multiString: parsley.token.text.String = _multiString
/** $multiString
*
* @note $raw
* @since 4.0.0
*/
def rawMultiString: parsley.token.text.String = _rawMultiString
}
/** $symbol
*
* @since 4.0.0
*/
val symbol: parsley.token.symbol.Symbol = new ConcreteSymbol(desc.nameDesc, desc.symbolDesc, errConfig)
}
/** This combinator ensures a parser fully parses all available input, and consumes whitespace
* at the start.
*
* This combinator should be used ''once'' as the outermost combinator in a parser. It is the
* only combinator that should consume ''leading'' whitespace, and this must be the first
* thing a parser does. It will ensure that, after the parser is complete, the end of the
* input stream has been reached.
*
* @since 4.0.0
*/
def fully[A](p: Parsley[A]): Parsley[A] = {
val init = if (desc.spaceDesc.whitespaceIsContextDependent) space.init else unit
init *> space.whiteSpace *> p <* eof
}
/** This object is concerned with special treatment of whitespace.
*
* For the vast majority of cases, the functionality within this
* object shouldn't be needed, as whitespace is consistently handled
* by `lexeme` and `fully`. However, for grammars where whitespace
* is significant (like indentation-sensitive languages), this object
* provides some more fine-grained control over how whitespace is
* consumed by the parsers within `lexeme`.
*
* @since 4.0.0
*/
object space {
private [Lexer] lazy val space = desc.spaceDesc.space.toNative
private lazy val wsImpl = Reg.make[Parsley[Unit]]
/** This parser initialises the whitespace used by the lexer when
* `spaceDesc.whiteSpaceIsContextDependent` is set to `true`.
*
* The whitespace is set to the implementation given by the lexical description.
* This parser '''must''' be used, by `fully` or otherwise, as the first thing
* the global parser does or an `UnfilledRegisterException` will occur.
*
* @note this parser is automatically invoked by the [[fully `fully`]] combinator when applicable.
* @see [[alter `alter`]] for how to change whitespace during a parse.
* @since 4.0.0
*/
def init: Parsley[Unit] = {
if (!desc.spaceDesc.whitespaceIsContextDependent) {
throw new UnsupportedOperationException( // scalastyle:ignore throw
"Whitespace cannot be initialised unless `spaceDesc.whitespaceIsContextDependent` is true"
)
}
wsImpl.put(configuredWhiteSpace)
}
/** This combinator changes how whitespace is parsed by lexemes for the duration of
* a given parser.
*
* So long as `spaceDesc.whiteSpaceIsContextDependent` is set to `true`, this combinator
* will be able to locally change the definition of whitespace during the given parser.
*
* @example
* In indentation sensitive languages, the indentation sensitivity is often ignored
* within parentheses or braces. In these cases `lexeme.enclosing.parens(space.alter(withNewline)(p))`
* would allow unrestricted newlines within parentheses.
*
* @param newSpace the new implementation of whitespace to be used during the execution of `within`.
* @param within the parser that should be parsed using the updated whitespace.
* @note the whitespace will not be restored to its original implementation if the
* given parser fails having consumed input.
* @since 4.0.0
*/
def alter[A](newSpace: CharPredicate)(within: =>Parsley[A]): Parsley[A] = {
if (!desc.spaceDesc.whitespaceIsContextDependent) {
throw new UnsupportedOperationException( // scalastyle:ignore throw
"Whitespace cannot be altered unless `spaceDesc.whitespaceIsContextDependent` is true"
)
}
wsImpl.rollback(wsImpl.local(whiteSpace(newSpace))(within))
}
/** This parser skips '''zero''' or more (insignificant) whitespace characters as well as comments.
*
* The implementation of this parser depends on whether `whitespaceIsContextDependent` is
* set: when it is, this parser may change based on the use of the `alter` combinator.
* This parser will always use the `hide` combinator as to not appear as a valid alternative
* in an error message: it's likely always the case whitespace can be added at any given time,
* but that doesn't make it a ''useful'' suggestion unless it is significant.
*
* @since 4.0.0
*/
val whiteSpace: Parsley[Unit] = {
if (desc.spaceDesc.whitespaceIsContextDependent) wsImpl.get.flatten
else configuredWhiteSpace
}
/** This parser skips '''zero''' or more comments.
*
* The implementation of this combinator does not vary with `whitespaceIsContextDependent`.
* It will use the `hide` combinator as to not appear as a valid alternative in an error
* message: adding a comment is often legal, but not a ''useful'' solution for how to make
* the input syntactically valid.
*
* @since 4.0.0
*/
lazy val skipComments: Parsley[Unit] = {
if (!desc.spaceDesc.supportsComments) unit
else new Parsley(new singletons.SkipComments(desc.spaceDesc, errConfig))
}
private def configuredWhiteSpace: Parsley[Unit] = whiteSpace(desc.spaceDesc.space)
private def whiteSpace(impl: CharPredicate): Parsley[Unit] = impl match {
case NotRequired => skipComments
case Basic(ws) => new Parsley(new singletons.WhiteSpace(ws, desc.spaceDesc, errConfig))
// satisfyUtf16 is effectively hidden, and so is Comment
case Unicode(ws) if desc.spaceDesc.supportsComments =>
skipMany(new Parsley(new singletons.Comment(desc.spaceDesc, errConfig)) <|> satisfy(ws))
case Unicode(ws) => skipMany(satisfy(ws))
}
}
}