/
edn.scm
571 lines (536 loc) · 23.6 KB
/
edn.scm
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
;;; Copyright (C) 2021 Rob Browning <rlb@defaultvalue.org>
;;; SPDX-License-Identifier: LGPL-2.1-or-later OR EPL-1.0+
(define-module (lokke scm edn)
;; We could replace this with lookahead-char for full scheme compliance
#:use-module ((ice-9 textual-ports) #:select (unget-char))
#:use-module ((lokke time) #:select (tagged-data->instant))
#:use-module ((lokke uuid) #:select (tagged-data->uuid))
#:use-module ((rnrs io ports)
#:select (eof-object get-char get-string-n lookahead-char))
#:use-module ((srfi srfi-1) #:select (delete-duplicates!))
#:use-module ((srfi srfi-9) #:select (define-record-type))
#:use-module ((srfi srfi-28) #:select (format))
#:use-module ((srfi srfi-43) #:select (reverse-list->vector))
#:use-module ((srfi srfi-88) #:select (string->keyword))
;; Expect imports to use a #:prefix edn/ or similar
#:export (read constructors reader read-string string-reader))
;; Aside from the exceptions specifically mentioned below and in the
;; test suite, we intend to match the edn specification:
;; https://github.com/edn-format/edn
;; Note: the public functions are a bit different from what Clojure
;; specifies both in order to be more idiomatic with respect to
;; Scheme, and to support the common "fixed?" case. i.e. it doesn't
;; seem likely you'd often want to change the tag handlers for each
;; call to read, for example.
;; FIXME: see FIXMEs in test suite
;; FIXME: support rnrs exceptions
;; FIXME: improve eof messages (include file/line and/or partial content)
;; FIXME: forbid unscoped user tags somewhere?
;; FIXME: add strict roundtripping string and port writer functions
;; FIXME: if we add print-length, make sure it's always avoidable for data writes
;; FIXME: ensure any writers always generate valid output or crash
;; FIXME: never dynamically follow *out* in writers
(define-record-type <edn-read-constructors>
(%read-constructors list-init list-add list-finish
vector-init vector-add vector-finish
set-init set-add set-finish
map-init map-add map-finish)
constructors?
(list-init list-init)
(list-add list-add)
(list-finish list-finish)
(vector-init vector-init)
(vector-add vector-add)
(vector-finish vector-finish)
(set-init set-init)
(set-add set-add)
(set-finish set-finish)
(map-init map-init)
(map-add map-add)
(map-finish map-finish))
;; Keep clojure.edn in mind whenever changing any of these
(define* (constructors #:key
(list-init list)
(list-add cons)
(list-finish reverse!)
(vector-init list)
(vector-add cons)
(vector-finish reverse-list->vector)
(set-init list)
(set-add cons)
(set-finish (lambda (x) (delete-duplicates! (reverse! x))))
(map-init list)
(map-add (lambda (s k v) (assoc-set! s k v)))
(map-finish identity))
(%read-constructors list-init list-add list-finish
vector-init vector-add vector-finish
set-init set-add set-finish
map-init map-add map-finish))
(define scheme-constructors (constructors))
(define (scheme-tagged-element-readers tag)
(case tag
((inst) (lambda (tag data) (tagged-data->instant data)))
((uuid) (lambda (tag data) (tagged-data->uuid data)))
(else #f)))
(define edn-digit-set (char-set #\0 #\1 #\2 #\3 #\4 #\5 #\6 #\7 #\8 #\9))
(define edn-blank-set (char-set-adjoin char-set:whitespace #\,))
(define edn-nosep-delimiter-set (char-set #\( #\) #\[ #\] #\{ #\}))
(define symbol-first-set (char-set #\. #\* #\+ #\! #\- #\_ #\? #\$ #\% #\& #\= #\< #\>))
(define symbol-set (char-set-adjoin symbol-first-set #\# #\:))
(define maybe-likely-symbol-letter-set
(char-set
#\a #\b #\c #\d #\e #\f #\g #\h #\i #\j #\k #\l #\m #\n #\o #\p #\q #\r #\s #\t #\u #\v #\w #\x #\y #\z
#\A #\B #\C #\D #\E #\F #\G #\H #\I #\J #\K #\L #\M #\N #\O #\P #\Q #\R #\S #\T #\U #\V #\W #\X #\Y #\Z))
(define remaining-symbol-letter-set
(char-set-difference char-set:letter maybe-likely-symbol-letter-set))
(define-inlinable (blank? c) (char-set-contains? edn-blank-set c))
(define-inlinable (delimiter? c) (char-set-contains? edn-nosep-delimiter-set c))
(define-inlinable (must-end port message)
(let ((c (lookahead-char port)))
(unless (or (eof-object? c)
(blank? c)
(delimiter? c))
(error (message c)))))
(define (require-char port name remaining result)
(let ((s (get-string-n port (string-length remaining))))
(cond
((eof-object? s)
(error (string-append "Unexpected end of file while reading " name)))
((< (string-length s) (string-length remaining))
(error (string-append "Unexpected end of file while reading " name)))
((string=? s remaining)
(must-end port (lambda (c)
(format "Unrecognized edn character ~a~a..." name c)))
result))))
(define (read-unicode-point port)
;; For now, we only support 4 hex digits, matching clojure/jvm
(let ((s (get-string-n port 4)))
(cond
((or (eof-object? s) (not (= 4 (string-length s))))
(error "Unexpected end of file while reading \\uNNNN character"))
(else
(must-end port (lambda (c)
(format "Unrecognized edn character \\u~a~a..." s c)))
(integer->char (string->number s 16))))))
(define (read-character-remainder port)
(let ((c (get-char port)))
(cond
((eof-object? c)
(error "Unexpected end of file while reading edn character"))
(else
(case c
((#\n) (require-char port "\\newline" "ewline" #\newline))
((#\t) (require-char port "\\tab" "ab" #\tab))
((#\s) (require-char port "\\space" "pace" #\space))
((#\r) (require-char port "\\return" "eturn" #\return))
((#\u) (read-unicode-point port))
(else c))))))
(define (read-string-remainder port pending backslashed?)
(let ((c (get-char port)))
(cond
((eof-object? c) (error "Unexpected end of file while reading edn string"))
(backslashed?
(case c
((#\n) (read-string-remainder port (cons #\newline pending) #f))
((#\" #\\) (read-string-remainder port (cons c pending) #f))
((#\t) (read-string-remainder port (cons #\tab pending) #f))
((#\r) (read-string-remainder port (cons #\return pending) #f))
(else
(error "Invalid edn string escape:" (string #\\ c)))))
(else
(case c
((#\") (reverse-list->string pending))
((#\\) (read-string-remainder port pending #t))
(else (read-string-remainder port (cons c pending) #f)))))))
(define (read-symbolic-str-remainder port pending name slash?)
;; By this point, we know it can't be a number, and has at least one char
(define (validate-symbol-name s slash?)
(unless (or (not slash?)
(string=? "/" s)
(and (> (string-length s) 2)
(not (string-prefix? "/" s))
(not (string-suffix? "/" s))))
(error (string-append "Invalid " name " name:") s))
s)
(let ((c (get-char port)))
(cond
((or (eof-object? c) (blank? c))
(validate-symbol-name (reverse-list->string pending) slash?))
((delimiter? c)
(unget-char port c)
(validate-symbol-name (reverse-list->string pending) slash?))
((char-set-contains? char-set:letter+digit c)
(read-symbolic-str-remainder port (cons c pending) name slash?))
(else
(case c
((#\/)
(let ((pending (cons c pending)))
(when slash?
(error (string-append "Encountered multiple / characters in edn "
name ":")
(reverse-list->string pending)))
(read-symbolic-str-remainder port pending name #t)))
;; Assumes that : and # were already forbidden as the first character
((#\. #\* #\+ #\! #\- #\_ #\? #\$ #\% #\& #\= #\< #\> #\# #\:)
(read-symbolic-str-remainder port (cons c pending) name slash?))
(else
(error (string-append "Invalid " name " prefix:")
(reverse-list->string (cons c pending)))))))))
(define-inlinable (read-symbolic-remainder port pending name slash?)
(let ((x (read-symbolic-str-remainder port pending name slash?)))
(if (string? x)
(string->symbol x)
x)))
(define (read-symbol-or-literal-remainder port pending expect literal)
;; Try to match the given literal, and at the first mismatched char,
;; defer to the normal symbol reader.
(let loop ((pending pending)
(i (1- (vector-length expect))))
(let ((c (get-char port)))
(cond
((or (eof-object? c) (blank? c))
(let ((s (reverse-list->string pending)))
(if (= -1 i)
literal
;; Must be a prefix of one of the literals, which is a valid symbol
(string->symbol (reverse-list->string pending)))))
((delimiter? c)
(unget-char port c)
(if (= -1 i)
literal
;; Must be a prefix of one of the literals, which is a valid symbol
(string->symbol (reverse-list->string pending))))
((>= i 0)
(if (eqv? c (vector-ref expect i))
(loop (cons c pending) (1- i))
(begin
(unget-char port c)
(read-symbolic-remainder port pending "symbol" #f))))
(else
(unget-char port c)
(read-symbolic-remainder port pending "symbol" #f))))))
(define (parse-float s)
(let ((n (string->number s)))
(unless n
(error "Invalid edn floating point syntax" s))
n))
(define (read-exp-magnitude port pending valid?)
(let ((c (get-char port)))
(cond
((eof-object? c)
(unless valid?
(error "Invalid edn floating point syntax" (reverse-list->string pending)))
(parse-float (reverse-list->string pending)))
((char-set-contains? edn-digit-set c)
(read-exp-magnitude port (cons c pending) #t))
((or (blank? c) (delimiter? c))
(unless valid?
(error "Invalid edn floating point syntax" (reverse-list->string pending)))
(parse-float (reverse-list->string pending)))
((eqv? c #\M)
(unless valid?
(error (string-append "Invalid edn floating point syntax")
(reverse-list->string pending) "M"))
(must-end port
(lambda (c)
(format "Invalid edn floating point syntax ~aM~a..."
(reverse-list->string pending) c)))
(parse-float (string-append "#e" (reverse-list->string pending))))
(else
(error "Invalid edn floating point syntax"
(reverse-list->string (cons c pending)))))))
(define (read-exp-remainder port pending)
(let ((c (get-char port)))
(cond
((eof-object? c)
(error "Invalid edn floating point syntax" (reverse-list->string pending)))
(else
(case c
((#\- #\+) (read-exp-magnitude port (cons c pending) #f))
(else
(unget-char port c)
(read-exp-magnitude port pending #f)))))))
(define (read-float-remainder port pending dot?)
;; Possibilities upon first entry:
;; dot? \.[0-9]
;; dot? [0-9]+\.
;;
;; Note that the edn spec's "grammar" doesn't appear to allow M
;; after anything but [0-9]+, but the prose says it can follow any
;; float, so support that.
(let ((c (get-char port)))
(cond
((or (eof-object? c) (blank? c) (delimiter? c))
(parse-float (reverse-list->string pending)))
((char-set-contains? edn-digit-set c)
(read-float-remainder port (cons c pending) dot?))
(else
(case c
((#\.)
(let ((pending (cons c pending)))
(when dot?
(error "Invalid edn floating point syntax"
(reverse-list->string pending)))
(read-float-remainder port pending #t)))
((#\e #\E)
(read-exp-remainder port (cons c pending)))
((#\M)
(must-end port
(lambda (c)
(format "Invalid edn floating point syntax ~aM~a..."
(reverse-list->string pending) c)))
(parse-float (string-append "#e" (reverse-list->string pending))))
(else
(error "Invalid edn floating point syntax"
(reverse-list->string (cons c pending)))))))))
(define (read-numeric-remainder port pending n?)
;; pending already has at least one [0-9]
(let ((c (get-char port)))
(cond
((or (eof-object? c) (blank? c)) ;; could only be integer
(let ((s (reverse-list->string pending)))
;; Treat leading zero as octal for now, matching the JVM, even
;; though the edn spec forbids it.
(case (string-ref s 0)
((#\- #\+) (case (string-ref s 1)
((#\0) (string->number (reverse-list->string pending) 8))
(else (string->number (reverse-list->string pending)))))
((#\0) (string->number (reverse-list->string pending) 8))
(else (string->number (reverse-list->string pending))))))
((delimiter? c) ;; could only be integer
(unget-char port c)
(string->number (reverse-list->string pending)))
((char-set-contains? edn-digit-set c)
(read-numeric-remainder port (cons c pending) n?))
((eqv? c #\.)
(let ((pending (cons c pending)))
(when n?
(error "Invalid edn number prefix:" (reverse-list->string pending)))
(read-float-remainder port pending #t)))
;; char set?
((memv c '(#\e #\E))
(let ((pending (cons c pending)))
(when n?
(error "Invalid edn number prefix:" (reverse-list->string pending)))
(read-exp-remainder port pending)))
((eqv? c #\N)
(when n?
(error "Invalid edn number prefix:"
(reverse-list->string (cons c pending))))
(read-numeric-remainder port pending #t))
((eqv? c #\M)
(must-end port
(lambda (c)
(format "Invalid edn nummeric syntax ~aM~a..."
(reverse-list->string pending) c)))
(string->number (reverse-list->string pending)))
((char-set-contains? char-set:digit c)
(error "Invalid edn numeric digit" c))
(else
(error "Invalid edn numeric prefix:" (reverse-list->string pending))))))
(define (read-symbol-or-number port pending frac?)
;; The JVM implementation treats anything starting with a "digit"
;; (Character/isDigit) as a number, but then only allows [0-9].
;; However, anything that starts with [-.+] is taken as a symbol
;; unless the second char is [0-9]. Match that behavior for now.
(let ((c (get-char port)))
(cond
((or (eof-object? c) (blank? c)) ;; must be + - .
(string->symbol (reverse-list->string pending)))
((delimiter? c) ;; must be + - .
(unget-char port c)
(string->symbol (reverse-list->string pending)))
((char-set-contains? edn-digit-set c)
(let ((pending (cons c pending)))
(if frac?
(read-float-remainder port pending #t)
(read-numeric-remainder port pending #f))))
((char-set-contains? char-set:letter c)
(read-symbolic-remainder port (cons c pending) "symbol" #f))
((char-set-contains? symbol-set c)
(read-symbolic-remainder port (cons c pending) "symbol" #f))
((eqv? c #\/)
(read-symbolic-remainder port (cons c pending) "symbol" #t))
((char-set-contains? char-set:digit c) ; Should only match non [0-9] digits
(read-symbolic-remainder port (cons c pending) "symbol" #f))
(else
(error "Invalid edn element prefix:" (reverse-list->string pending))))))
(define (read-keyword-remainder port)
(let ((c (get-char port)))
(cond
((eof-object? c)
(error "Unexpected end of file while reading edn keyword"))
((blank? c)
(error "Unexpected blank space after start of edn keyword"))
((delimiter? c)
(error (format "Unexpected delimiter ~s after start of edn keyword"
(string c))))
((eqv? c #\:) (error "Read :: at start of edn keyword"))
(else
(let ((s (read-symbolic-str-remainder port (list c) "keyword"
(eqv? c #\/))))
(if (equal? s "/")
(error "Invalid edn keyword :/")
(string->keyword s)))))))
(define (read-container-content port name terminator pending
tag-rdr default-tag-rdr constructors
add finish)
(let ((c (get-char port)))
(cond
((eof-object? c)
(error (string-append "Unexpected end of file while reading " name)))
((eqv? c terminator) (finish pending))
(else
(unget-char port c)
(let ((x (read* port tag-rdr default-tag-rdr constructors)))
(when (eof-object? x)
(error (string-append "Unexpected end of file while reading edn "
name)))
(read-container-content port name terminator
(add x pending)
tag-rdr default-tag-rdr constructors
add finish))))))
(define (read-map-content port pending tag-rdr default-tag-rdr constructors)
(let ((c (get-char port)))
(cond
((eof-object? c)
(error (string-append "Unexpected end of file while map")))
((eqv? c #\}) ((map-finish constructors) pending))
(else
(unget-char port c)
(let* ((k (read* port tag-rdr default-tag-rdr constructors))
(_ (when (eof-object? k)
(error "Unexpected end of file while reading edn map")))
(v (read* port tag-rdr default-tag-rdr constructors))
(_ (when (eof-object? v)
(error "Unexpected end of file after edn map key:" k))))
(read-map-content port ((map-add constructors) pending k v)
tag-rdr default-tag-rdr constructors))))))
(define (read-tagged-element port tag-rdr default-tag-rdr constructors)
(let ((tag (read* port tag-rdr default-tag-rdr constructors)))
(cond
((eof-object? tag)
(error "Unexpected end of file in edn tagged element"))
((not (symbol? tag))
(error "Found non-symbol at start of edn tagged element:" tag))
(else
(let ((data (read* port tag-rdr default-tag-rdr constructors)))
(when (eof-object? data)
(error "End of file in edn tagged element data" tag))
(let ((read-elt (or (and tag-rdr (tag-rdr tag))
default-tag-rdr)))
(unless read-elt
(error "Unrecognized edn tagged element type:" tag))
(read-elt tag data)))))))
(define (read-#-remainder port tag-rdr default-tag-rdr constructors)
(let ((c (get-char port)))
(cond
((eof-object? c) (error "Unexpected end of file while reading # form"))
(else
(case c
((#\{)
(read-container-content port "set" #\}
((set-init constructors))
tag-rdr default-tag-rdr constructors
(set-add constructors)
(set-finish constructors)))
;; Match the (1.9+) JVM behavior, even though these are not
;; (yet?) in the spec.
((#\#)
(let ((x (read* port tag-rdr default-tag-rdr constructors)))
(unless (symbol? x)
(error "Value read after ## was not a symbol:" x))
(case x
((Inf) +inf.0)
((-Inf) -inf.0)
((NaN) +nan.0)
(else
(error "Unrecognized ## literal:"
(string-append "##" (symbol->string x)))))))
((#\_)
(read* port tag-rdr default-tag-rdr constructors)
(read* port tag-rdr default-tag-rdr constructors))
(else
(unget-char port c)
(read-tagged-element port tag-rdr default-tag-rdr constructors)))))))
(define* (read* port tag-rdr default-tag-rdr constructors)
(let ((c (get-char port)))
(cond
((eof-object? c) c)
((blank? c) (read* port tag-rdr default-tag-rdr constructors))
(else
(case c
((#\() (read-container-content port "list" #\)
((list-init constructors))
tag-rdr default-tag-rdr constructors
(list-add constructors)
(list-finish constructors)))
((#\[) (read-container-content port "vector" #\]
((vector-init constructors))
tag-rdr default-tag-rdr constructors
(vector-add constructors)
(vector-finish constructors)))
((#\{) (read-map-content port ((map-init constructors))
tag-rdr default-tag-rdr constructors))
((#\") (read-string-remainder port '() #f))
((#\:) (read-keyword-remainder port))
((#\n) (read-symbol-or-literal-remainder port (list c) #(#\l #\i) #nil))
((#\t) (read-symbol-or-literal-remainder port (list c) #(#\e #\u #\r) #t))
((#\f) (read-symbol-or-literal-remainder port (list c) #(#\e #\s #\l #\a) #f))
((#\-) (read-symbol-or-number port (list c) #f))
((#\\) (read-character-remainder port))
((#\.) (read-symbol-or-number port (list c) #t))
((#\+) (read-symbol-or-number port (list c) #f))
((#\#) (read-#-remainder port tag-rdr default-tag-rdr constructors))
(else
(cond
((char-set-contains? edn-digit-set c)
(read-numeric-remainder port (list c) #f))
((eqv? c #\/) (read-symbolic-remainder port (list c) "symbol" #t))
((char-set-contains? symbol-first-set c)
(read-symbolic-remainder port (list c) "symbol" #f))
((char-set-contains? maybe-likely-symbol-letter-set c)
(read-symbolic-remainder port (list c) "symbol" #f))
((char-set-contains? remaining-symbol-letter-set c)
(read-symbolic-remainder port (list c) "symbol" #f))
((char-set-contains? char-set:digit c)
;; Match the JVM reader for now
(error "Number begins with digit other than 0-9" c))
(else
(error "Unrecognized edn construct:" c)))))))))
(define* (read-top port on-eof tag-rdr default-tag-rdr constructors)
(let ((c (lookahead-char port)))
(if (eof-object? c)
(on-eof)
(read* port tag-rdr default-tag-rdr constructors))))
(define* (reader #:key
(on-eof eof-object)
(tag-reader scheme-tagged-element-readers)
(default-tag-reader #f)
(constructors scheme-constructors)
(fixed? #t))
(if fixed?
(case-lambda
(()
(read-top (current-input-port) on-eof
tag-reader default-tag-reader
constructors))
((port)
(read-top port on-eof
tag-reader default-tag-reader
constructors)))
(lambda* (#:optional
(port (current-input-port))
#:key
(on-eof on-eof)
(tag-reader tag-reader)
(default-tag-reader default-tag-reader)
(constructors constructors))
(read-top port on-eof tag-reader default-tag-reader constructors))))
(define (string-reader . args)
(let ((read (apply reader args)))
(lambda (s . opts)
(call-with-input-string s
(lambda (port) (apply read port opts))))))
(define read (reader))
(define read-string (string-reader))