summary refs log tree commit diff stats
path: root/lib/pure/parseutils.nim
blob: a4032f8f36c9f0b66d0af69c4769839fabd748d0 (plain) (blame)
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
#
#
#            Nim's Runtime Library
#        (c) Copyright 2012 Andreas Rumpf
#
#    See the file "copying.txt", included in this
#    distribution, for details about the copyright.
#

## This module contains helpers for parsing tokens, numbers, integers, floats,
## identifiers, etc.
##
## To unpack raw bytes look at the `streams <streams.html>`_ module.
##
##
## .. code-block::
##    import parseutils
##
##    let logs = @["2019-01-10: OK_", "2019-01-11: FAIL_", "2019-01: aaaa"]
##
##    for log in logs:
##      var res: string
##      if parseUntil(log, res, ':') == 10: # YYYY-MM-DD == 10
##        echo res & " - " & captureBetween(log, ' ', '_')
##        # => 2019-01-10 - OK
##
##
## .. code-block::
##    import parseutils
##    from strutils import Digits, parseInt
##
##    let userInput1 = "2019 school start"
##    let userInput2 = "3 years back"
##
##    let startYear = input1[0..skipWhile(input1, Digits)-1] # 2019
##    let yearsBack = input2[0..skipWhile(input2, Digits)-1] # 3
##
##    echo "Examination is in " & $(parseInt(startYear) + parseInt(yearsBack))
##
##
## **See also:**
## * `strutils module<strutils.html>`_ for combined and identical parsing proc's
## * `json module<json.html>`_ for a JSON parser
## * `parsecfg module<parsecfg.html>`_ for a configuration file parser
## * `parsecsv module<parsecsv.html>`_ for a simple CSV (comma separated value) parser
## * `parseopt module<parseopt.html>`_ for a command line parser
## * `parsexml module<parsexml.html>`_ for a XML / HTML parser
## * `other parsers<lib.html#pure-libraries-parsers>`_ for other parsers


{.deadCodeElim: on.}  # dce option deprecated

{.push debugger:off .} # the user does not want to trace a part
                       # of the standard library!

include "system/inclrtl"

const
  Whitespace = {' ', '\t', '\v', '\r', '\l', '\f'}
  IdentChars = {'a'..'z', 'A'..'Z', '0'..'9', '_'}
  IdentStartChars = {'a'..'z', 'A'..'Z', '_'}
    ## copied from strutils

proc toLower(c: char): char {.inline.} =
  result = if c in {'A'..'Z'}: chr(ord(c)-ord('A')+ord('a')) else: c

proc parseBin*[T: SomeInteger](s: string, number: var T, start = 0, maxLen = 0): int
  {.noSideEffect.} =
  ## Parses a binary number and stores its value in ``number``.
  ##
  ## Returns the number of the parsed characters or 0 in case of an error.
  ## If error, the value of ``number`` is not changed.
  ##
  ## If ``maxLen == 0``, the parsing continues until the first non-bin character
  ## or to the end of the string. Otherwise, no more than ``maxLen`` characters
  ## are parsed starting from the ``start`` position.
  ## 
  ## It does not check for overflow. If the value represented by the string is
  ## too big to fit into ``number``, only the value of last fitting characters
  ## will be stored in ``number`` without producing an error.
  runnableExamples:
    var num: int
    doAssert parseBin("0100_1110_0110_1001_1110_1101", num) == 29
    doAssert num == 5138925
    doAssert parseBin("3", num) == 0
    var num8: int8
    doAssert parseBin("0b_0100_1110_0110_1001_1110_1101", num8) == 32
    doAssert num8 == 0b1110_1101'i8
    doAssert parseBin("0b_0100_1110_0110_1001_1110_1101", num8, 3, 9) == 9
    doAssert num8 == 0b0100_1110'i8
    var num8u: uint8
    doAssert parseBin("0b_0100_1110_0110_1001_1110_1101", num8u) == 32
    doAssert num8u == 237
    var num64: int64
    doAssert parseBin("0100111001101001111011010100111001101001", num64) == 40
    doAssert num64 == 336784608873
  var i = start
  var output = T(0)
  var foundDigit = false
  let last = min(s.len, if maxLen == 0: s.len else: i + maxLen)
  if i + 1 < last and s[i] == '0' and (s[i+1] in {'b', 'B'}): inc(i, 2)
  while i < last:
    case s[i]
    of '_': discard
    of '0'..'1':
      output = output shl 1 or T(ord(s[i]) - ord('0'))
      foundDigit = true
    else: break
    inc(i)
  if foundDigit:
    number = output
    result = i - start

proc parseOct*[T: SomeInteger](s: string, number: var T, start = 0, maxLen = 0): int
  {.noSideEffect.} =
  ## Parses an octal number and stores its value in ``number``.
  ##
  ## Returns the number of the parsed characters or 0 in case of an error.
  ## If error, the value of ``number`` is not changed.
  ##
  ## If ``maxLen == 0``, the parsing continues until the first non-oct character
  ## or to the end of the string. Otherwise, no more than ``maxLen`` characters
  ## are parsed starting from the ``start`` position.
  ## 
  ## It does not check for overflow. If the value represented by the string is
  ## too big to fit into ``number``, only the value of last fitting characters
  ## will be stored in ``number`` without producing an error.
  runnableExamples:
    var num: int
    doAssert parseOct("0o23464755", num) == 10
    doAssert num == 5138925
    doAssert parseOct("8", num) == 0
    var num8: int8
    doAssert parseOct("0o_1464_755", num8) == 11
    doAssert num8 == -19
    doAssert parseOct("0o_1464_755", num8, 3, 3) == 3
    doAssert num8 == 102
    var num8u: uint8
    doAssert parseOct("1464755", num8u) == 7
    doAssert num8u == 237
    var num64: int64
    doAssert parseOct("2346475523464755", num64) == 16
    doAssert num64 == 86216859871725
  var i = start
  var output = T(0)
  var foundDigit = false
  let last = min(s.len, if maxLen == 0: s.len else: i + maxLen)
  if i + 1 < last and s[i] == '0' and (s[i+1] in {'o', 'O'}): inc(i, 2)
  while i < last:
    case s[i]
    of '_': discard
    of '0'..'7':
      output = output shl 3 or T(ord(s[i]) - ord('0'))
      foundDigit = true
    else: break
    inc(i)
  if foundDigit:
    number = output
    result = i - start

proc parseHex*[T: SomeInteger](s: string, number: var T, start = 0, maxLen = 0): int
  {.noSideEffect.} =
  ## Parses a hexadecimal number and stores its value in ``number``.
  ##
  ## Returns the number of the parsed characters or 0 in case of an error.
  ## If error, the value of ``number`` is not changed.
  ##
  ## If ``maxLen == 0``, the parsing continues until the first non-hex character
  ## or to the end of the string. Otherwise, no more than ``maxLen`` characters
  ## are parsed starting from the ``start`` position.
  ## 
  ## It does not check for overflow. If the value represented by the string is
  ## too big to fit into ``number``, only the value of last fitting characters
  ## will be stored in ``number`` without producing an error.
  runnableExamples:
    var num: int
    doAssert parseHex("4E_69_ED", num) == 8
    doAssert num == 5138925
    doAssert parseHex("X", num) == 0
    doAssert parseHex("#ABC", num) == 4
    var num8: int8
    doAssert parseHex("0x_4E_69_ED", num8) == 11
    doAssert num8 == 0xED'i8
    doAssert parseHex("0x_4E_69_ED", num8, 3, 2) == 2
    doAssert num8 == 0x4E'i8
    var num8u: uint8
    doAssert parseHex("0x_4E_69_ED", num8u) == 11
    doAssert num8u == 237
    var num64: int64
    doAssert parseHex("4E69ED4E69ED", num64) == 12
    doAssert num64 == 86216859871725
  var i = start
  var output = T(0)
  var foundDigit = false
  let last = min(s.len, if maxLen == 0: s.len else: i + maxLen)
  if i + 1 < last and s[i] == '0' and (s[i+1] in {'x', 'X'}): inc(i, 2)
  elif i < last and s[i] == '#': inc(i)
  while i < last:
    case s[i]
    of '_': discard
    of '0'..'9':
      output = output shl 4 or T(ord(s[i]) - ord('0'))
      foundDigit = true
    of 'a'..'f':
      output = output shl 4 or T(ord(s[i]) - ord('a') + 10)
      foundDigit = true
    of 'A'..'F':
      output = output shl 4 or T(ord(s[i]) - ord('A') + 10)
      foundDigit = true
    else: break
    inc(i)
  if foundDigit:
    number = output
    result = i - start

proc parseIdent*(s: string, ident: var string, start = 0): int =
  ## Parses an identifier and stores it in ``ident``. Returns
  ## the number of the parsed characters or 0 in case of an error.
  runnableExamples:
    var res: string
    doAssert parseIdent("Hello World", res, 0) == 5
    doAssert res == "Hello"
    doAssert parseIdent("Hello World", res, 1) == 4
    doAssert res == "ello"
    doAssert parseIdent("Hello World", res, 6) == 5
    doAssert res == "World"
  var i = start
  if i < s.len and s[i] in IdentStartChars:
    inc(i)
    while i < s.len and s[i] in IdentChars: inc(i)
    ident = substr(s, start, i-1)
    result = i-start

proc parseIdent*(s: string, start = 0): string =
  ## Parses an identifier and returns it or an empty string in
  ## case of an error.
  runnableExamples:
    doAssert parseIdent("Hello World", 0) == "Hello"
    doAssert parseIdent("Hello World", 1) == "ello"
    doAssert parseIdent("Hello World", 5) == ""
    doAssert parseIdent("Hello World", 6) == "World"
  result = ""
  var i = start
  if i < s.len and s[i] in IdentStartChars:
    inc(i)
    while i < s.len and s[i] in IdentChars: inc(i)
    result = substr(s, start, i-1)

proc skipWhitespace*(s: string, start = 0): int {.inline.} =
  ## Skips the whitespace starting at ``s[start]``. Returns the number of
  ## skipped characters.
  runnableExamples:
    doAssert skipWhitespace("Hello World", 0) == 0
    doAssert skipWhitespace(" Hello World", 0) == 1
    doAssert skipWhitespace("Hello World", 5) == 1
    doAssert skipWhitespace("Hello  World", 5) == 2
  while start+result < s.len and s[start+result] in Whitespace: inc(result)

proc skip*(s, token: string, start = 0): int {.inline.} =
  ## Skips the `token` starting at ``s[start]``. Returns the length of `token`
  ## or 0 if there was no `token` at ``s[start]``.
  runnableExamples:
    doAssert skip("2019-01-22", "2019", 0) == 4
    doAssert skip("2019-01-22", "19", 0) == 0
    doAssert skip("2019-01-22", "19", 2) == 2
    doAssert skip("CAPlow", "CAP", 0) == 3
    doAssert skip("CAPlow", "cap", 0) == 0
  while start+result < s.len and result < token.len and
      s[result+start] == token[result]:
    inc(result)
  if result != token.len: result = 0

proc skipIgnoreCase*(s, token: string, start = 0): int =
  ## Same as `skip` but case is ignored for token matching.
  runnableExamples:
    doAssert skipIgnoreCase("CAPlow", "CAP", 0) == 3
    doAssert skipIgnoreCase("CAPlow", "cap", 0) == 3
  while start+result < s.len and result < token.len and
      toLower(s[result+start]) == toLower(token[result]): inc(result)
  if result != token.len: result = 0

proc skipUntil*(s: string, until: set[char], start = 0): int {.inline.} =
  ## Skips all characters until one char from the set `until` is found
  ## or the end is reached.
  ## Returns number of characters skipped.
  runnableExamples:
    doAssert skipUntil("Hello World", {'W', 'e'}, 0) == 1
    doAssert skipUntil("Hello World", {'W'}, 0) == 6
    doAssert skipUntil("Hello World", {'W', 'd'}, 0) == 6
  while start+result < s.len and s[result+start] notin until: inc(result)

proc skipUntil*(s: string, until: char, start = 0): int {.inline.} =
  ## Skips all characters until the char `until` is found
  ## or the end is reached.
  ## Returns number of characters skipped.
  runnableExamples:
    doAssert skipUntil("Hello World", 'o', 0) == 4
    doAssert skipUntil("Hello World", 'o', 4) == 0
    doAssert skipUntil("Hello World", 'W', 0) == 6
    doAssert skipUntil("Hello World", 'w', 0) == 11
  while start+result < s.len and s[result+start] != until: inc(result)

proc skipWhile*(s: string, toSkip: set[char], start = 0): int {.inline.} =
  ## Skips all characters while one char from the set `token` is found.
  ## Returns number of characters skipped.
  runnableExamples:
    doAssert skipWhile("Hello World", {'H', 'e'}) == 2
    doAssert skipWhile("Hello World", {'e'}) == 0
    doAssert skipWhile("Hello World", {'W', 'o', 'r'}, 6) == 3
  while start+result < s.len and s[result+start] in toSkip: inc(result)

proc parseUntil*(s: string, token: var string, until: set[char],
                 start = 0): int {.inline.} =
  ## Parses a token and stores it in ``token``. Returns
  ## the number of the parsed characters or 0 in case of an error. A token
  ## consists of the characters notin `until`.
  runnableExamples:
    var myToken: string
    doAssert parseUntil("Hello World", myToken, {'W', 'o', 'r'}) == 4
    doAssert myToken == "Hell"
    doAssert parseUntil("Hello World", myToken, {'W', 'r'}) == 6
    doAssert myToken == "Hello "
    doAssert parseUntil("Hello World", myToken, {'W', 'r'}, 3) == 3
    doAssert myToken == "lo "
  var i = start
  while i < s.len and s[i] notin until: inc(i)
  result = i-start
  token = substr(s, start, i-1)

proc parseUntil*(s: string, token: var string, until: char,
                 start = 0): int {.inline.} =
  ## Parses a token and stores it in ``token``. Returns
  ## the number of the parsed characters or 0 in case of an error. A token
  ## consists of any character that is not the `until` character.
  runnableExamples:
    var myToken: string
    doAssert parseUntil("Hello World", myToken, 'W') == 6
    doAssert myToken == "Hello "
    doAssert parseUntil("Hello World", myToken, 'o') == 4
    doAssert myToken == "Hell"
    doAssert parseUntil("Hello World", myToken, 'o', 2) == 2
    doAssert myToken == "ll"
  var i = start
  while i < s.len and s[i] != until: inc(i)
  result = i-start
  token = substr(s, start, i-1)

proc parseUntil*(s: string, token: var string, until: string,
                 start = 0): int {.inline.} =
  ## Parses a token and stores it in ``token``. Returns
  ## the number of the parsed characters or 0 in case of an error. A token
  ## consists of any character that comes before the `until`  token.
  runnableExamples:
    var myToken: string
    doAssert parseUntil("Hello World", myToken, "Wor") == 6
    doAssert myToken == "Hello "
    doAssert parseUntil("Hello World", myToken, "Wor", 2) == 4
    doAssert myToken == "llo "
  if until.len == 0:
    token.setLen(0)
    return 0
  var i = start
  while i < s.len:
    if s[i] == until[0]:
      var u = 1
      while i+u < s.len and u < until.len and s[i+u] == until[u]:
        inc u
      if u >= until.len: break
    inc(i)
  result = i-start
  token = substr(s, start, i-1)

proc parseWhile*(s: string, token: var string, validChars: set[char],
                 start = 0): int {.inline.} =
  ## Parses a token and stores it in ``token``. Returns
  ## the number of the parsed characters or 0 in case of an error. A token
  ## consists of the characters in `validChars`.
  runnableExamples:
    var myToken: string
    doAssert parseWhile("Hello World", myToken, {'W', 'o', 'r'}, 0) == 0
    doAssert myToken.len() == 0
    doAssert parseWhile("Hello World", myToken, {'W', 'o', 'r'}, 6) == 3
    doAssert myToken == "Wor"
  var i = start
  while i < s.len and s[i] in validChars: inc(i)
  result = i-start
  token = substr(s, start, i-1)

proc captureBetween*(s: string, first: char, second = '\0', start = 0): string =
  ## Finds the first occurrence of ``first``, then returns everything from there
  ## up to ``second`` (if ``second`` is '\0', then ``first`` is used).
  runnableExamples:
    doAssert captureBetween("Hello World", 'e') == "llo World"
    doAssert captureBetween("Hello World", 'e', 'r') == "llo Wo"
    doAssert captureBetween("Hello World", 'l', start = 6) == "d"
  var i = skipUntil(s, first, start)+1+start
  result = ""
  discard s.parseUntil(result, if second == '\0': first else: second, i)

proc integerOutOfRangeError() {.noinline.} =
  raise newException(ValueError, "Parsed integer outside of valid range")

# See #6752
when defined(js):
  {.push overflowChecks: off.}

proc rawParseInt(s: string, b: var BiggestInt, start = 0): int =
  var
    sign: BiggestInt = -1
    i = start
  if i < s.len:
    if s[i] == '+': inc(i)
    elif s[i] == '-':
      inc(i)
      sign = 1
  if i < s.len and s[i] in {'0'..'9'}:
    b = 0
    while i < s.len and s[i] in {'0'..'9'}:
      let c = ord(s[i]) - ord('0')
      if b >= (low(BiggestInt) + c) div 10:
        b = b * 10 - c
      else:
        integerOutOfRangeError()
      inc(i)
      while i < s.len and s[i] == '_': inc(i) # underscores are allowed and ignored
    if sign == -1 and b == low(BiggestInt):
      integerOutOfRangeError()
    else:
      b = b * sign
      result = i - start

when defined(js):
  {.pop.} # overflowChecks: off

proc parseBiggestInt*(s: string, number: var BiggestInt, start = 0): int {.
  rtl, extern: "npuParseBiggestInt", noSideEffect, raises: [ValueError].} =
  ## Parses an integer starting at `start` and stores the value into `number`.
  ## Result is the number of processed chars or 0 if there is no integer.
  ## `ValueError` is raised if the parsed integer is out of the valid range.
  runnableExamples:
    var res: BiggestInt
    doAssert parseBiggestInt("9223372036854775807", res, 0) == 19
    doAssert res == 9223372036854775807
  var res: BiggestInt
  # use 'res' for exception safety (don't write to 'number' in case of an
  # overflow exception):
  result = rawParseInt(s, res, start)
  if result != 0:
    number = res

proc parseInt*(s: string, number: var int, start = 0): int {.
  rtl, extern: "npuParseInt", noSideEffect, raises: [ValueError].} =
  ## Parses an integer starting at `start` and stores the value into `number`.
  ## Result is the number of processed chars or 0 if there is no integer.
  ## `ValueError` is raised if the parsed integer is out of the valid range.
  runnableExamples:
    var res: int
    doAssert parseInt("2019", res, 0) == 4
    doAssert res == 2019
    doAssert parseInt("2019", res, 2) == 2
    doAssert res == 19
  var res: BiggestInt
  result = parseBiggestInt(s, res, start)
  when sizeof(int) <= 4:
    if res < low(int) or res > high(int):
      integerOutOfRangeError()
  if result != 0:
    number = int(res)

proc parseSaturatedNatural*(s: string, b: var int, start = 0): int {.
  raises: [].}=
  ## Parses a natural number into ``b``. This cannot raise an overflow
  ## error. ``high(int)`` is returned for an overflow.
  ## The number of processed character is returned.
  ## This is usually what you really want to use instead of `parseInt`:idx:.
  runnableExamples:
    var res = 0
    discard parseSaturatedNatural("848", res)
    doAssert res == 848
  var i = start
  if i < s.len and s[i] == '+': inc(i)
  if i < s.len and s[i] in {'0'..'9'}:
    b = 0
    while i < s.len and s[i] in {'0'..'9'}:
      let c = ord(s[i]) - ord('0')
      if b <= (high(int) - c) div 10:
        b = b * 10 + c
      else:
        b = high(int)
      inc(i)
      while i < s.len and s[i] == '_': inc(i) # underscores are allowed and ignored
    result = i - start

proc rawParseUInt(s: string, b: var BiggestUInt, start = 0): int =
  var
    res = 0.BiggestUInt
    prev = 0.BiggestUInt
    i = start
  if i < s.len - 1 and s[i] == '-' and s[i + 1] in {'0'..'9'}:
    integerOutOfRangeError()
  if i < s.len and s[i] == '+': inc(i) # Allow
  if i < s.len and s[i] in {'0'..'9'}:
    b = 0
    while i < s.len and s[i] in {'0'..'9'}:
      prev = res
      res = res * 10 + (ord(s[i]) - ord('0')).BiggestUInt
      if prev > res:
        integerOutOfRangeError()
      inc(i)
      while i < s.len and s[i] == '_': inc(i) # underscores are allowed and ignored
    b = res
    result = i - start

proc parseBiggestUInt*(s: string, number: var BiggestUInt, start = 0): int {.
  rtl, extern: "npuParseBiggestUInt", noSideEffect, raises: [ValueError].} =
  ## Parses an unsigned integer starting at `start` and stores the value
  ## into `number`.
  ## `ValueError` is raised if the parsed integer is out of the valid range.
  runnableExamples:
    var res: BiggestUInt
    doAssert parseBiggestUInt("12", res, 0) == 2
    doAssert res == 12
    doAssert parseBiggestUInt("1111111111111111111", res, 0) == 19
    doAssert res == 1111111111111111111'u64
  var res: BiggestUInt
  # use 'res' for exception safety (don't write to 'number' in case of an
  # overflow exception):
  result = rawParseUInt(s, res, start)
  if result != 0:
    number = res

proc parseUInt*(s: string, number: var uint, start = 0): int {.
  rtl, extern: "npuParseUInt", noSideEffect, raises: [ValueError].} =
  ## Parses an unsigned integer starting at `start` and stores the value
  ## into `number`.
  ## `ValueError` is raised if the parsed integer is out of the valid range.
  runnableExamples:
    var res: uint
    doAssert parseUInt("3450", res) == 4
    doAssert res == 3450
    doAssert parseUInt("3450", res, 2) == 2
    doAssert res == 50
  var res: BiggestUInt
  result = parseBiggestUInt(s, res, start)
  when sizeof(BiggestUInt) > sizeof(uint) and sizeof(uint) <= 4:
    if res > 0xFFFF_FFFF'u64:
      integerOutOfRangeError()
  if result != 0:
    number = uint(res)

proc parseBiggestFloat*(s: string, number: var BiggestFloat, start = 0): int {.
  magic: "ParseBiggestFloat", importc: "nimParseBiggestFloat", noSideEffect.}
  ## Parses a float starting at `start` and stores the value into `number`.
  ## Result is the number of processed chars or 0 if a parsing error
  ## occurred.

proc parseFloat*(s: string, number: var float, start = 0): int {.
  rtl, extern: "npuParseFloat", noSideEffect.} =
  ## Parses a float starting at `start` and stores the value into `number`.
  ## Result is the number of processed chars or 0 if there occurred a parsing
  ## error.
  runnableExamples:
    var res: float
    doAssert parseFloat("32", res, 0) == 2
    doAssert res == 32.0
    doAssert parseFloat("32.57", res, 0) == 5
    doAssert res == 32.57
    doAssert parseFloat("32.57", res, 3) == 2
    doAssert res == 57.00
  var bf: BiggestFloat
  result = parseBiggestFloat(s, bf, start)
  if result != 0:
    number = bf

type
  InterpolatedKind* = enum   ## Describes for `interpolatedFragments`
                             ## which part of the interpolated string is
                             ## yielded; for example in "str$$$var${expr}"
    ikStr,                   ## ``str`` part of the interpolated string
    ikDollar,                ## escaped ``$`` part of the interpolated string
    ikVar,                   ## ``var`` part of the interpolated string
    ikExpr                   ## ``expr`` part of the interpolated string

iterator interpolatedFragments*(s: string): tuple[kind: InterpolatedKind,
  value: string] =
  ## Tokenizes the string `s` into substrings for interpolation purposes.
  ##
  ## Example:
  ##
  ## .. code-block:: nim
  ##   for k, v in interpolatedFragments("  $this is ${an  example}  $$"):
  ##     echo "(", k, ", \"", v, "\")"
  ##
  ## Results in:
  ##
  ## .. code-block:: nim
  ##   (ikString, "  ")
  ##   (ikExpr, "this")
  ##   (ikString, " is ")
  ##   (ikExpr, "an  example")
  ##   (ikString, "  ")
  ##   (ikDollar, "$")
  var i = 0
  var kind: InterpolatedKind
  while true:
    var j = i
    if j < s.len and s[j] == '$':
      if j+1 < s.len and s[j+1] == '{':
        inc j, 2
        var nesting = 0
        block curlies:
          while j < s.len:
            case s[j]
            of '{': inc nesting
            of '}':
              if nesting == 0:
                inc j
                break curlies
              dec nesting
            else: discard
            inc j
          raise newException(ValueError,
            "Expected closing '}': " & substr(s, i, s.high))
        inc i, 2 # skip ${
        kind = ikExpr
      elif j+1 < s.len and s[j+1] in IdentStartChars:
        inc j, 2
        while j < s.len and s[j] in IdentChars: inc(j)
        inc i # skip $
        kind = ikVar
      elif j+1 < s.len and s[j+1] == '$':
        inc j, 2
        inc i # skip $
        kind = ikDollar
      else:
        raise newException(ValueError,
          "Unable to parse a varible name at " & substr(s, i, s.high))
    else:
      while j < s.len and s[j] != '$': inc j
      kind = ikStr
    if j > i:
      # do not copy the trailing } for ikExpr:
      yield (kind, substr(s, i, j-1-ord(kind == ikExpr)))
    else:
      break
    i = j

when isMainModule:
  import sequtils
  let input = "$test{}  $this is ${an{  example}}  "
  let expected = @[(ikVar, "test"), (ikStr, "{}  "), (ikVar, "this"),
                   (ikStr, " is "), (ikExpr, "an{  example}"), (ikStr, "  ")]
  doAssert toSeq(interpolatedFragments(input)) == expected

  var value = 0
  discard parseHex("0x38", value)
  doAssert value == 56

  value = -1
  doAssert(parseSaturatedNatural("848", value) == 3)
  doAssert value == 848

  value = -1
  discard parseSaturatedNatural("84899999999999999999324234243143142342135435342532453", value)
  doAssert value == high(int)

  value = -1
  discard parseSaturatedNatural("9223372036854775808", value)
  doAssert value == high(int)

  value = -1
  discard parseSaturatedNatural("9223372036854775807", value)
  doAssert value == high(int)

  value = -1
  discard parseSaturatedNatural("18446744073709551616", value)
  doAssert value == high(int)

  value = -1
  discard parseSaturatedNatural("18446744073709551615", value)
  doAssert value == high(int)

  value = -1
  doAssert(parseSaturatedNatural("1_000_000", value) == 9)
  doAssert value == 1_000_000

  var i64Value: int64
  discard parseBiggestInt("9223372036854775807", i64Value)
  doAssert i64Value == 9223372036854775807

{.pop.}