/
parse.go
973 lines (876 loc) · 21.6 KB
/
parse.go
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
// Package parse implements the elvish parser.
package parse
//go:generate ./boilerplate.py
//go:generate stringer -type=PrimaryType,RedirMode -output=string.go
import (
"bytes"
"errors"
"fmt"
"unicode"
)
// Parse parses Elvish source. If the error is not nil, it always has type
// ParseError.
func Parse(srcname, src string) (*Chunk, error) {
ps := NewParser(srcname, src)
n := ParseChunk(ps)
ps.Done()
return n, ps.Errors()
}
// Errors.
var (
errShouldBeForm = newError("", "form")
errBadLHS = errors.New("bad assignment LHS")
errDuplicateExitusRedir = newError("duplicate exitus redir")
errBadRedirSign = newError("bad redir sign", "'<'", "'>'", "'>>'", "'<>'")
errShouldBeFD = newError("", "a composite term representing fd")
errShouldBeFilename = newError("", "a composite term representing filename")
errShouldBeArray = newError("", "spaced")
errStringUnterminated = newError("string not terminated")
errChainedAssignment = newError("chained assignment not yet supported")
errInvalidEscape = newError("invalid escape sequence")
errInvalidEscapeOct = newError("invalid escape sequence", "octal digit")
errInvalidEscapeHex = newError("invalid escape sequence", "hex digit")
errInvalidEscapeControl = newError("invalid control sequence", "a rune between @ (0x40) and _(0x5F)")
errShouldBePrimary = newError("",
"single-quoted string", "double-quoted string", "bareword")
errShouldBeVariableName = newError("", "variable name")
errShouldBeRBracket = newError("", "']'")
errShouldBeRBrace = newError("", "'}'")
errShouldBeBraceSepOrRBracket = newError("", "','", "'}'")
errShouldBeRParen = newError("", "')'")
errShouldBeCompound = newError("", "compound")
errShouldBeEqual = newError("", "'='")
errBothElementsAndPairs = newError("cannot contain both list elements and map pairs")
errShouldBeEscapeSequence = newError("", "escape sequence")
)
// Chunk = { PipelineSep | Space } { Pipeline { PipelineSep | Space } }
type Chunk struct {
node
Pipelines []*Pipeline
}
func (bn *Chunk) parse(ps *Parser) {
bn.parseSeps(ps)
for startsPipeline(ps.peek()) {
bn.addToPipelines(ParsePipeline(ps))
if bn.parseSeps(ps) == 0 {
break
}
}
}
func isPipelineSep(r rune) bool {
return r == '\n' || r == ';'
}
// parseSeps parses pipeline separators along with whitespaces. It returns the
// number of pipeline separators parsed.
func (bn *Chunk) parseSeps(ps *Parser) int {
nseps := 0
for {
r := ps.peek()
if isPipelineSep(r) {
// parse as a Sep
parseSep(bn, ps, r)
nseps++
} else if IsSpace(r) {
// parse a run of spaces as a Sep
parseSpaces(bn, ps)
} else if r == '#' {
// parse a comment as a Sep
for {
r := ps.peek()
if r == eof || r == '\n' {
break
}
ps.next()
}
addSep(bn, ps)
nseps++
} else {
break
}
}
return nseps
}
// Pipeline = Form { '|' Form }
type Pipeline struct {
node
Forms []*Form
Background bool
}
func (pn *Pipeline) parse(ps *Parser) {
pn.addToForms(ParseForm(ps))
for parseSep(pn, ps, '|') {
parseSpacesAndNewlines(pn, ps)
if !startsForm(ps.peek()) {
ps.error(errShouldBeForm)
return
}
pn.addToForms(ParseForm(ps))
}
parseSpaces(pn, ps)
if ps.peek() == '&' {
ps.next()
addSep(pn, ps)
pn.Background = true
parseSpaces(pn, ps)
}
}
func startsPipeline(r rune) bool {
return startsForm(r)
}
// Form = { Space } { { Assignment } { Space } }
// { Compound } { Space } { ( Compound | MapPair | Redir | ExitusRedir ) { Space } }
type Form struct {
node
Assignments []*Assignment
Head *Compound
// Left-hand-sides for the spacey assignment. Right-hand-sides are in Args.
Vars []*Compound
Args []*Compound
Opts []*MapPair
Redirs []*Redir
ExitusRedir *ExitusRedir
}
func (fn *Form) parse(ps *Parser) {
parseSpaces(fn, ps)
for fn.tryAssignment(ps) {
parseSpaces(fn, ps)
}
// Parse head.
if !startsCompound(ps.peek(), CmdExpr) {
if len(fn.Assignments) > 0 {
// Assignment-only form.
return
}
// Bad form.
ps.error(fmt.Errorf("bad rune at form head: %q", ps.peek()))
}
fn.setHead(ParseCompound(ps, CmdExpr))
parseSpaces(fn, ps)
for {
r := ps.peek()
switch {
case r == '&':
ps.next()
hasMapPair := startsCompound(ps.peek(), LHSExpr)
ps.backup()
if !hasMapPair {
// background indicator
return
}
fn.addToOpts(ParseMapPair(ps))
case startsCompound(r, NormalExpr):
if ps.hasPrefix("?>") {
if fn.ExitusRedir != nil {
ps.error(errDuplicateExitusRedir)
// Parse the duplicate redir anyway.
addChild(fn, ParseExitusRedir(ps))
} else {
fn.setExitusRedir(ParseExitusRedir(ps))
}
continue
}
cn := ParseCompound(ps, NormalExpr)
if isRedirSign(ps.peek()) {
// Redir
fn.addToRedirs(ParseRedir(ps, cn))
} else if cn.sourceText == "=" {
// Spacey assignment.
// Turn the equal sign into a Sep.
addChild(fn, NewSep(ps.src, cn.begin, cn.end))
// Turn the head and preceding arguments into LHSs.
addLHS := func(cn *Compound) {
if len(cn.Indexings) == 1 && checkVariableInAssignment(cn.Indexings[0].Head, ps) {
fn.Vars = append(fn.Vars, cn)
} else {
ps.errorp(cn.begin, cn.end, errBadLHS)
}
}
if fn.Head != nil {
addLHS(fn.Head)
} else {
ps.error(errChainedAssignment)
}
fn.Head = nil
for _, cn := range fn.Args {
addLHS(cn)
}
fn.Args = nil
} else {
fn.addToArgs(cn)
}
case isRedirSign(r):
fn.addToRedirs(ParseRedir(ps, nil))
default:
return
}
parseSpaces(fn, ps)
}
}
// tryAssignment tries to parse an assignment. If succeeded, it adds the parsed
// assignment to fn.Assignments and returns true. Otherwise it rewinds the
// parser and returns false.
func (fn *Form) tryAssignment(ps *Parser) bool {
if !startsIndexing(ps.peek(), LHSExpr) {
return false
}
pos := ps.pos
errorEntries := ps.errors.Entries
an := ParseAssignment(ps)
// If errors were added, revert
if len(ps.errors.Entries) > len(errorEntries) {
ps.errors.Entries = errorEntries
ps.pos = pos
return false
}
fn.addToAssignments(an)
return true
}
func startsForm(r rune) bool {
return IsSpace(r) || startsCompound(r, CmdExpr)
}
// Assignment = Indexing '=' Compound
type Assignment struct {
node
Left *Indexing
Right *Compound
}
func (an *Assignment) parse(ps *Parser) {
an.setLeft(ParseIndexing(ps, LHSExpr))
head := an.Left.Head
if !checkVariableInAssignment(head, ps) {
ps.errorp(head.Begin(), head.End(), errShouldBeVariableName)
}
if !parseSep(an, ps, '=') {
ps.error(errShouldBeEqual)
}
an.setRight(ParseCompound(ps, NormalExpr))
}
func checkVariableInAssignment(p *Primary, ps *Parser) bool {
if p.Type == Braced {
// XXX don't check further inside braced expression
return true
}
if p.Type != Bareword && p.Type != SingleQuoted && p.Type != DoubleQuoted {
return false
}
if p.Value == "" {
return false
}
for _, r := range p.Value {
// XXX special case '&' and '@'.
if !allowedInVariableName(r) && r != '&' && r != '@' {
return false
}
}
return true
}
// ExitusRedir = '?' '>' { Space } Compound
type ExitusRedir struct {
node
Dest *Compound
}
func (ern *ExitusRedir) parse(ps *Parser) {
ps.next()
ps.next()
addSep(ern, ps)
parseSpaces(ern, ps)
ern.setDest(ParseCompound(ps, NormalExpr))
}
// Redir = { Compound } { '<'|'>'|'<>'|'>>' } { Space } ( '&'? Compound )
type Redir struct {
node
Left *Compound
Mode RedirMode
RightIsFd bool
Right *Compound
}
func (rn *Redir) parse(ps *Parser, dest *Compound) {
// The parsing of the Left part is done in Form.parse.
if dest != nil {
rn.setLeft(dest)
rn.begin = dest.begin
}
begin := ps.pos
for isRedirSign(ps.peek()) {
ps.next()
}
sign := ps.src[begin:ps.pos]
switch sign {
case "<":
rn.Mode = Read
case ">":
rn.Mode = Write
case ">>":
rn.Mode = Append
case "<>":
rn.Mode = ReadWrite
default:
ps.error(errBadRedirSign)
}
addSep(rn, ps)
parseSpaces(rn, ps)
if parseSep(rn, ps, '&') {
rn.RightIsFd = true
}
rn.setRight(ParseCompound(ps, NormalExpr))
if len(rn.Right.Indexings) == 0 {
if rn.RightIsFd {
ps.error(errShouldBeFD)
} else {
ps.error(errShouldBeFilename)
}
return
}
}
func isRedirSign(r rune) bool {
return r == '<' || r == '>'
}
// RedirMode records the mode of an IO redirection.
type RedirMode int
// Possible values for RedirMode.
const (
BadRedirMode RedirMode = iota
Read
Write
ReadWrite
Append
)
// Compound = { Indexing }
type Compound struct {
node
Indexings []*Indexing
}
// ExprCtx represents special contexts of expression parsing.
type ExprCtx int
const (
// NormalExpr represents a normal expression, namely none of the special
// ones below.
NormalExpr ExprCtx = iota
// CmdExpr represents an expression used as the command in a form. In this
// context, unquoted <>*^ are treated as bareword characters.
CmdExpr
// LHSExpr represents an expression used as the left-hand-side in either
// assignments or map pairs. In this context, an unquoted = serves as an
// expression terminator and is thus not treated as a bareword character.
LHSExpr
// BracedElemExpr represents an expression used as an element in a braced
// expression. In this context, an unquoted , serves as an expression
// terminator and is thus not treated as a bareword character.
BracedElemExpr
// strictExpr is only meaningful to allowedInBareword.
strictExpr
)
func (cn *Compound) parse(ps *Parser, ctx ExprCtx) {
cn.tilde(ps)
for startsIndexing(ps.peek(), ctx) {
cn.addToIndexings(ParseIndexing(ps, ctx))
}
}
// tilde parses a tilde if there is one. It is implemented here instead of
// within Primary since a tilde can only appear as the first part of a
// Compound. Elsewhere tildes are barewords.
func (cn *Compound) tilde(ps *Parser) {
if ps.peek() == '~' {
ps.next()
base := node{nil, ps.pos - 1, ps.pos, "~", nil}
pn := &Primary{node: base, Type: Tilde, Value: "~"}
in := &Indexing{node: base}
in.setHead(pn)
cn.addToIndexings(in)
}
}
func startsCompound(r rune, ctx ExprCtx) bool {
return startsIndexing(r, ctx)
}
// Indexing = Primary { '[' Array ']' }
type Indexing struct {
node
Head *Primary
Indicies []*Array
}
func (in *Indexing) parse(ps *Parser, ctx ExprCtx) {
in.setHead(ParsePrimary(ps, ctx))
for parseSep(in, ps, '[') {
if !startsArray(ps.peek()) {
ps.error(errShouldBeArray)
}
in.addToIndicies(ParseArray(ps, false))
if !parseSep(in, ps, ']') {
ps.error(errShouldBeRBracket)
return
}
}
}
func startsIndexing(r rune, ctx ExprCtx) bool {
return startsPrimary(r, ctx)
}
// Array = { Space | '\n' } { Compound { Space | '\n' } }
type Array struct {
node
Compounds []*Compound
// When non-empty, records the occurrences of semicolons by the indices of
// the compounds they appear before. For instance, [; ; a b; c d;] results
// in Semicolons={0 0 2 4}.
Semicolons []int
}
func (sn *Array) parse(ps *Parser, allowSemicolon bool) {
parseSep := func() {
parseSpacesAndNewlines(sn, ps)
if allowSemicolon {
for parseSep(sn, ps, ';') {
sn.Semicolons = append(sn.Semicolons, len(sn.Compounds))
}
parseSpacesAndNewlines(sn, ps)
}
}
parseSep()
for startsCompound(ps.peek(), NormalExpr) {
sn.addToCompounds(ParseCompound(ps, NormalExpr))
parseSep()
}
}
func IsSpace(r rune) bool {
return r == ' ' || r == '\t'
}
func startsArray(r rune) bool {
return IsSpaceOrNewline(r) || startsIndexing(r, NormalExpr)
}
// Primary is the smallest expression unit.
type Primary struct {
node
Type PrimaryType
// The unquoted string value. Valid for Bareword, SingleQuoted,
// DoubleQuoted, Variable, Wildcard and Tilde.
Value string
Elements []*Compound // Valid for List and Labda
Chunk *Chunk // Valid for OutputCapture, ExitusCapture and Lambda
MapPairs []*MapPair // Valid for Map and Lambda
Braced []*Compound // Valid for Braced
}
// PrimaryType is the type of a Primary.
type PrimaryType int
// Possible values for PrimaryType.
const (
BadPrimary PrimaryType = iota
Bareword
SingleQuoted
DoubleQuoted
Variable
Wildcard
Tilde
ExceptionCapture
OutputCapture
List
Lambda
Map
Braced
)
func (pn *Primary) parse(ps *Parser, ctx ExprCtx) {
r := ps.peek()
if !startsPrimary(r, ctx) {
ps.error(errShouldBePrimary)
return
}
// Try bareword early, since it has precedence over wildcard on *
// when ctx = commandExpr.
if allowedInBareword(r, ctx) {
pn.bareword(ps, ctx)
return
}
switch r {
case '\'':
pn.singleQuoted(ps)
case '"':
pn.doubleQuoted(ps)
case '$':
pn.variable(ps)
case '*':
pn.wildcard(ps)
case '?':
if ps.hasPrefix("?(") {
pn.exitusCapture(ps)
} else {
pn.wildcard(ps)
}
case '(':
pn.outputCapture(ps)
case '[':
pn.lbracket(ps)
case '{':
pn.lbrace(ps)
default:
// Parse an empty bareword.
pn.Type = Bareword
}
}
func (pn *Primary) singleQuoted(ps *Parser) {
pn.Type = SingleQuoted
ps.next()
var buf bytes.Buffer
defer func() { pn.Value = buf.String() }()
for {
switch r := ps.next(); r {
case eof:
ps.error(errStringUnterminated)
return
case '\'':
if ps.peek() == '\'' {
// Two consecutive single quotes
ps.next()
buf.WriteByte('\'')
} else {
// End of string
return
}
default:
buf.WriteRune(r)
}
}
}
func (pn *Primary) doubleQuoted(ps *Parser) {
pn.Type = DoubleQuoted
ps.next()
var buf bytes.Buffer
defer func() { pn.Value = buf.String() }()
for {
switch r := ps.next(); r {
case eof:
ps.error(errStringUnterminated)
return
case '"':
return
case '\\':
switch r := ps.next(); r {
case 'c', '^':
// Control sequence
r := ps.next()
if r < 0x40 || r >= 0x60 {
ps.backup()
ps.error(errInvalidEscapeControl)
ps.next()
}
buf.WriteByte(byte(r - 0x40))
case 'x', 'u', 'U':
var n int
switch r {
case 'x':
n = 2
case 'u':
n = 4
case 'U':
n = 8
}
var rr rune
for i := 0; i < n; i++ {
d, ok := hexToDigit(ps.next())
if !ok {
ps.backup()
ps.error(errInvalidEscapeHex)
break
}
rr = rr*16 + d
}
buf.WriteRune(rr)
case '0', '1', '2', '3', '4', '5', '6', '7':
// 2 more octal digits
rr := r - '0'
for i := 0; i < 2; i++ {
r := ps.next()
if r < '0' || r > '7' {
ps.backup()
ps.error(errInvalidEscapeOct)
break
}
rr = rr*8 + (r - '0')
}
buf.WriteRune(rr)
default:
if rr, ok := doubleEscape[r]; ok {
buf.WriteRune(rr)
} else {
ps.backup()
ps.error(errInvalidEscape)
ps.next()
}
}
default:
buf.WriteRune(r)
}
}
}
// a table for the simple double-quote escape sequences.
var doubleEscape = map[rune]rune{
// same as golang
'a': '\a', 'b': '\b', 'f': '\f', 'n': '\n', 'r': '\r',
't': '\t', 'v': '\v', '\\': '\\', '"': '"',
// additional
'e': '\033',
}
var doubleUnescape = map[rune]rune{}
func init() {
for k, v := range doubleEscape {
doubleUnescape[v] = k
}
}
func hexToDigit(r rune) (rune, bool) {
switch {
case '0' <= r && r <= '9':
return r - '0', true
case 'a' <= r && r <= 'f':
return r - 'a' + 10, true
case 'A' <= r && r <= 'F':
return r - 'A' + 10, true
default:
return -1, false
}
}
func (pn *Primary) variable(ps *Parser) {
pn.Type = Variable
defer func() { pn.Value = ps.src[pn.begin+1 : ps.pos] }()
ps.next()
// The character of the variable name can be anything.
if ps.next() == eof {
ps.backup()
ps.error(errShouldBeVariableName)
ps.next()
}
for allowedInVariableName(ps.peek()) {
ps.next()
}
}
// The following are allowed in variable names:
// * Anything beyond ASCII that is printable
// * Letters and numbers
// * The symbols "-_:~"
func allowedInVariableName(r rune) bool {
return (r >= 0x80 && unicode.IsPrint(r)) ||
('0' <= r && r <= '9') ||
('a' <= r && r <= 'z') ||
('A' <= r && r <= 'Z') ||
r == '-' || r == '_' || r == ':' || r == '~'
}
func (pn *Primary) wildcard(ps *Parser) {
pn.Type = Wildcard
for isWildcard(ps.peek()) {
ps.next()
}
pn.Value = ps.src[pn.begin:ps.pos]
}
func isWildcard(r rune) bool {
return r == '*' || r == '?'
}
func (pn *Primary) exitusCapture(ps *Parser) {
ps.next()
ps.next()
addSep(pn, ps)
pn.Type = ExceptionCapture
pn.setChunk(ParseChunk(ps))
if !parseSep(pn, ps, ')') {
ps.error(errShouldBeRParen)
}
}
func (pn *Primary) outputCapture(ps *Parser) {
pn.Type = OutputCapture
parseSep(pn, ps, '(')
pn.setChunk(ParseChunk(ps))
if !parseSep(pn, ps, ')') {
ps.error(errShouldBeRParen)
}
}
// List = '[' { Space } { Compound } ']'
// = '[' { Space } { MapPair { Space } } ']'
// Map = '[' { Space } '&' { Space } ']'
// Lambda = '[' { Space } { (Compound | MapPair) { Space } } ']' '{' Chunk '}'
func (pn *Primary) lbracket(ps *Parser) {
parseSep(pn, ps, '[')
parseSpacesAndNewlines(pn, ps)
loneAmpersand := false
items:
for {
r := ps.peek()
switch {
case r == '&':
ps.next()
hasMapPair := startsCompound(ps.peek(), LHSExpr)
if !hasMapPair {
loneAmpersand = true
addSep(pn, ps)
parseSpacesAndNewlines(pn, ps)
break items
}
ps.backup()
pn.addToMapPairs(ParseMapPair(ps))
case startsCompound(r, NormalExpr):
pn.addToElements(ParseCompound(ps, NormalExpr))
default:
break items
}
parseSpacesAndNewlines(pn, ps)
}
if !parseSep(pn, ps, ']') {
ps.error(errShouldBeRBracket)
}
if parseSep(pn, ps, '{') {
pn.lambda(ps)
} else {
if loneAmpersand || len(pn.MapPairs) > 0 {
if len(pn.Elements) > 0 {
ps.error(errBothElementsAndPairs)
}
pn.Type = Map
} else {
pn.Type = List
}
}
}
// lambda parses a lambda expression. The opening brace has been seen.
func (pn *Primary) lambda(ps *Parser) {
pn.Type = Lambda
pn.setChunk(ParseChunk(ps))
if !parseSep(pn, ps, '}') {
ps.error(errShouldBeRBrace)
}
}
// Braced = '{' Compound { BracedSep Compounds } '}'
// BracedSep = { Space | '\n' } [ ',' ] { Space | '\n' }
func (pn *Primary) lbrace(ps *Parser) {
parseSep(pn, ps, '{')
if r := ps.peek(); r == ';' || r == '\n' || IsSpace(r) {
pn.lambda(ps)
return
}
pn.Type = Braced
// XXX: The compound can be empty, which allows us to parse {,foo}.
// Allowing compounds to be empty can be fragile in other cases.
pn.addToBraced(ParseCompound(ps, BracedElemExpr))
for isBracedSep(ps.peek()) {
parseSpacesAndNewlines(pn, ps)
// optional, so ignore the return value
parseSep(pn, ps, ',')
parseSpacesAndNewlines(pn, ps)
pn.addToBraced(ParseCompound(ps, BracedElemExpr))
}
if !parseSep(pn, ps, '}') {
ps.error(errShouldBeBraceSepOrRBracket)
}
}
func isBracedSep(r rune) bool {
return r == ',' || IsSpaceOrNewline(r)
}
func (pn *Primary) bareword(ps *Parser, ctx ExprCtx) {
pn.Type = Bareword
defer func() { pn.Value = ps.src[pn.begin:ps.pos] }()
for allowedInBareword(ps.peek(), ctx) {
ps.next()
}
}
// allowedInBareword returns where a rune is allowed in barewords in the given
// expression context. The special strictExpr context queries whether the rune
// is allowed in all contexts.
//
// The following are allowed in barewords:
//
// * Anything allowed in variable names
// * The symbols "./@%+!"
// * The symbol "=", if ctx != lhsExpr && ctx != strictExpr
// * The symbol ",", if ctx != bracedExpr && ctx != strictExpr
// * The symbols "<>*^", if ctx = commandExpr
//
// The seemingly weird inclusion of \ is for easier path manipulation in
// Windows.
func allowedInBareword(r rune, ctx ExprCtx) bool {
return allowedInVariableName(r) || r == '.' || r == '/' ||
r == '@' || r == '%' || r == '+' || r == '!' ||
(ctx != LHSExpr && ctx != strictExpr && r == '=') ||
(ctx != BracedElemExpr && ctx != strictExpr && r == ',') ||
(ctx == CmdExpr && (r == '<' || r == '>' || r == '*' || r == '^'))
}
func startsPrimary(r rune, ctx ExprCtx) bool {
return r == '\'' || r == '"' || r == '$' || allowedInBareword(r, ctx) ||
r == '?' || r == '*' || r == '(' || r == '[' || r == '{'
}
// MapPair = '&' { Space } Compound { Space } Compound
type MapPair struct {
node
Key, Value *Compound
}
func (mpn *MapPair) parse(ps *Parser) {
parseSep(mpn, ps, '&')
mpn.setKey(ParseCompound(ps, LHSExpr))
if len(mpn.Key.Indexings) == 0 {
ps.error(errShouldBeCompound)
}
if parseSep(mpn, ps, '=') {
parseSpacesAndNewlines(mpn, ps)
// Parse value part.
mpn.setValue(ParseCompound(ps, NormalExpr))
// The value part can be empty.
}
}
// Sep is the catch-all node type for leaf nodes that lack internal structures
// and semantics, and serve solely for syntactic purposes. The parsing of
// separators depend on the Parent node; as such it lacks a genuine parse
// method.
type Sep struct {
node
}
func NewSep(src string, begin, end int) *Sep {
return &Sep{node{nil, begin, end, src[begin:end], nil}}
}
func addSep(n Node, ps *Parser) {
var begin int
ch := n.Children()
if len(ch) > 0 {
begin = ch[len(ch)-1].End()
} else {
begin = n.Begin()
}
if begin < ps.pos {
addChild(n, NewSep(ps.src, begin, ps.pos))
}
}
func parseSep(n Node, ps *Parser, sep rune) bool {
if ps.peek() == sep {
ps.next()
addSep(n, ps)
return true
}
return false
}
func parseSpaces(n Node, ps *Parser) {
parseSpacesInner(n, ps, IsSpace)
}
func parseSpacesAndNewlines(n Node, ps *Parser) {
parseSpacesInner(n, ps, IsSpaceOrNewline)
}
func parseSpacesInner(n Node, ps *Parser, isSpace func(rune) bool) {
spaces:
for {
r := ps.peek()
switch {
case isSpace(r):
ps.next()
case r == '\\': // line continuation
ps.next()
switch ps.peek() {
case '\n':
ps.next()
case eof:
ps.error(errShouldBeEscapeSequence)
default:
ps.backup()
break spaces
}
default:
break spaces
}
}
addSep(n, ps)
}
func IsSpaceOrNewline(r rune) bool {
return IsSpace(r) || r == '\n'
}
func addChild(p Node, ch Node) {
p.n().children = append(p.n().children, ch)
ch.n().parent = p
}