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
|
const std = @import("std");
const assert = std.debug.assert;
const Allocator = std.mem.Allocator;
const ArrayList = std.ArrayList;
const Scanner = @import("./scanner.zig").Scanner;
const Token = @import("./scanner.zig").Token;
const AllocWhen = @import("./scanner.zig").AllocWhen;
const default_max_value_len = @import("./scanner.zig").default_max_value_len;
const isNumberFormattedLikeAnInteger = @import("./scanner.zig").isNumberFormattedLikeAnInteger;
pub const ParseOptions = struct {
/// Behaviour when a duplicate field is encountered.
duplicate_field_behavior: enum {
use_first,
@"error",
use_last,
} = .@"error",
/// If false, finding an unknown field returns an error.
ignore_unknown_fields: bool = false,
/// Passed to json.Scanner.nextAllocMax() or json.Reader.nextAllocMax().
/// The default for parseFromSlice() or parseFromTokenSource() with a *json.Scanner input
/// is the length of the input slice, which means error.ValueTooLong will never be returned.
/// The default for parseFromTokenSource() with a *json.Reader is default_max_value_len.
max_value_len: ?usize = null,
};
/// Parses the json document from s and returns the result.
/// The provided allocator is used both for temporary allocations during parsing the document,
/// and also to allocate any pointer values in the return type.
/// If T contains any pointers, free the memory with `std.json.parseFree`.
/// Note that `error.BufferUnderrun` is not actually possible to return from this function.
pub fn parseFromSlice(comptime T: type, allocator: Allocator, s: []const u8, options: ParseOptions) ParseError(T, Scanner)!T {
var scanner = Scanner.initCompleteInput(allocator, s);
defer scanner.deinit();
return parseFromTokenSource(T, allocator, &scanner, options);
}
/// `scanner_or_reader` must be either a `*std.json.Scanner` with complete input or a `*std.json.Reader`.
/// allocator is used to allocate the data of T if necessary,
/// such as if T is `*u32` or `[]u32`.
/// If T contains any pointers, free the memory with `std.json.parseFree`.
/// If T contains no pointers, the allocator may sometimes be used for temporary allocations,
/// but no call to `std.json.parseFree` will be necessary;
/// all temporary allocations will be freed before this function returns.
/// Note that `error.BufferUnderrun` is not actually possible to return from this function.
pub fn parseFromTokenSource(comptime T: type, allocator: Allocator, scanner_or_reader: anytype, options: ParseOptions) ParseError(T, @TypeOf(scanner_or_reader.*))!T {
if (@TypeOf(scanner_or_reader.*) == Scanner) {
assert(scanner_or_reader.is_end_of_input);
}
var resolved_options = options;
if (resolved_options.max_value_len == null) {
if (@TypeOf(scanner_or_reader.*) == Scanner) {
resolved_options.max_value_len = scanner_or_reader.input.len;
} else {
resolved_options.max_value_len = default_max_value_len;
}
}
const r = try parseInternal(T, allocator, scanner_or_reader, resolved_options);
errdefer parseFree(T, allocator, r);
assert(.end_of_document == try scanner_or_reader.next());
return r;
}
/// The error set that will be returned from parsing T from *Source.
/// Note that this may contain error.BufferUnderrun, but that error will never actually be returned.
pub fn ParseError(comptime T: type, comptime Source: type) type {
// `inferred_types` is used to avoid infinite recursion for recursive type definitions.
const inferred_types = [_]type{};
// A few of these will either always be present or present enough of the time that
// omitting them is more confusing than always including them.
return error{UnexpectedToken} || Source.NextError || Source.PeekError ||
ParseInternalErrorImpl(T, Source, &inferred_types);
}
fn ParseInternalErrorImpl(comptime T: type, comptime Source: type, comptime inferred_types: []const type) type {
for (inferred_types) |ty| {
if (T == ty) return error{};
}
switch (@typeInfo(T)) {
.Bool => return error{},
.Float, .ComptimeFloat => return Source.AllocError || std.fmt.ParseFloatError,
.Int, .ComptimeInt => {
return Source.AllocError || error{ InvalidNumber, Overflow } ||
std.fmt.ParseIntError || std.fmt.ParseFloatError;
},
.Optional => |optional_info| return ParseInternalErrorImpl(optional_info.child, Source, inferred_types ++ [_]type{T}),
.Enum => return Source.AllocError || error{InvalidEnumTag},
.Union => |unionInfo| {
if (unionInfo.tag_type) |_| {
var errors = Source.AllocError || error{UnknownField};
for (unionInfo.fields) |u_field| {
errors = errors || ParseInternalErrorImpl(u_field.type, Source, inferred_types ++ [_]type{T});
}
return errors;
} else {
@compileError("Unable to parse into untagged union '" ++ @typeName(T) ++ "'");
}
},
.Struct => |structInfo| {
var errors = Scanner.AllocError || error{
DuplicateField,
UnknownField,
MissingField,
};
for (structInfo.fields) |field| {
errors = errors || ParseInternalErrorImpl(field.type, Source, inferred_types ++ [_]type{T});
}
return errors;
},
.Array => |arrayInfo| {
return error{LengthMismatch} ||
ParseInternalErrorImpl(arrayInfo.child, Source, inferred_types ++ [_]type{T});
},
.Vector => |vecInfo| {
return error{LengthMismatch} ||
ParseInternalErrorImpl(vecInfo.child, Source, inferred_types ++ [_]type{T});
},
.Pointer => |ptrInfo| {
switch (ptrInfo.size) {
.One, .Slice => {
return ParseInternalErrorImpl(ptrInfo.child, Source, inferred_types ++ [_]type{T});
},
else => @compileError("Unable to parse into type '" ++ @typeName(T) ++ "'"),
}
},
else => return error{},
}
unreachable;
}
fn parseInternal(
comptime T: type,
allocator: Allocator,
source: anytype,
options: ParseOptions,
) ParseError(T, @TypeOf(source.*))!T {
switch (@typeInfo(T)) {
.Bool => {
return switch (try source.next()) {
.true => true,
.false => false,
else => error.UnexpectedToken,
};
},
.Float, .ComptimeFloat => {
const token = try source.nextAllocMax(allocator, .alloc_if_needed, options.max_value_len.?);
defer freeAllocated(allocator, token);
const slice = switch (token) {
.number, .string => |slice| slice,
.allocated_number, .allocated_string => |slice| slice,
else => return error.UnexpectedToken,
};
return try std.fmt.parseFloat(T, slice);
},
.Int, .ComptimeInt => {
const token = try source.nextAllocMax(allocator, .alloc_if_needed, options.max_value_len.?);
defer freeAllocated(allocator, token);
const slice = switch (token) {
.number, .string => |slice| slice,
.allocated_number, .allocated_string => |slice| slice,
else => return error.UnexpectedToken,
};
if (isNumberFormattedLikeAnInteger(slice))
return std.fmt.parseInt(T, slice, 10);
// Try to coerce a float to an integer.
const float = try std.fmt.parseFloat(f128, slice);
if (@round(float) != float) return error.InvalidNumber;
if (float > std.math.maxInt(T) or float < std.math.minInt(T)) return error.Overflow;
return @floatToInt(T, float);
},
.Optional => |optionalInfo| {
switch (try source.peekNextTokenType()) {
.null => {
_ = try source.next();
return null;
},
else => {
return try parseInternal(optionalInfo.child, allocator, source, options);
},
}
},
.Enum => |enumInfo| {
const token = try source.nextAllocMax(allocator, .alloc_if_needed, options.max_value_len.?);
defer freeAllocated(allocator, token);
const slice = switch (token) {
.number, .string => |slice| slice,
.allocated_number, .allocated_string => |slice| slice,
else => return error.UnexpectedToken,
};
// Check for a named value.
if (std.meta.stringToEnum(T, slice)) |value| return value;
// Check for a numeric value.
if (!isNumberFormattedLikeAnInteger(slice)) return error.InvalidEnumTag;
const n = std.fmt.parseInt(enumInfo.tag_type, slice, 10) catch return error.InvalidEnumTag;
return try std.meta.intToEnum(T, n);
},
.Union => |unionInfo| {
const UnionTagType = unionInfo.tag_type orelse @compileError("Unable to parse into untagged union '" ++ @typeName(T) ++ "'");
if (.object_begin != try source.next()) return error.UnexpectedToken;
var result: ?T = null;
errdefer {
if (result) |r| {
inline for (unionInfo.fields) |u_field| {
if (r == @field(UnionTagType, u_field.name)) {
parseFree(u_field.type, allocator, @field(r, u_field.name));
}
}
}
}
var name_token: ?Token = try source.nextAllocMax(allocator, .alloc_if_needed, options.max_value_len.?);
errdefer {
if (name_token) |t| {
freeAllocated(allocator, t);
}
}
const field_name = switch (name_token.?) {
.string => |slice| slice,
.allocated_string => |slice| slice,
else => return error.UnexpectedToken,
};
inline for (unionInfo.fields) |u_field| {
if (std.mem.eql(u8, u_field.name, field_name)) {
// Free the name token now in case we're using an allocator that optimizes freeing the last allocated object.
// (Recursing into parseInternal() might trigger more allocations.)
freeAllocated(allocator, name_token.?);
name_token = null;
if (u_field.type == void) {
// void isn't really a json type, but we can support void payload union tags with {} as a value.
if (.object_begin != try source.next()) return error.UnexpectedToken;
if (.object_end != try source.next()) return error.UnexpectedToken;
result = @unionInit(T, u_field.name, {});
} else {
// Recurse.
result = @unionInit(T, u_field.name, try parseInternal(u_field.type, allocator, source, options));
}
break;
}
} else {
// Didn't match anything.
return error.UnknownField;
}
if (.object_end != try source.next()) return error.UnexpectedToken;
return result.?;
},
.Struct => |structInfo| {
if (structInfo.is_tuple) {
if (.array_begin != try source.next()) return error.UnexpectedToken;
var r: T = undefined;
var fields_seen: usize = 0;
errdefer {
inline for (0..structInfo.fields.len) |i| {
if (i < fields_seen) {
parseFree(structInfo.fields[i].type, allocator, r[i]);
}
}
}
inline for (0..structInfo.fields.len) |i| {
r[i] = try parseInternal(structInfo.fields[i].type, allocator, source, options);
fields_seen = i + 1;
}
if (.array_end != try source.next()) return error.UnexpectedToken;
return r;
}
if (.object_begin != try source.next()) return error.UnexpectedToken;
var r: T = undefined;
var fields_seen = [_]bool{false} ** structInfo.fields.len;
errdefer {
inline for (structInfo.fields, 0..) |field, i| {
if (fields_seen[i]) {
parseFree(field.type, allocator, @field(r, field.name));
}
}
}
while (true) {
var name_token: ?Token = try source.nextAllocMax(allocator, .alloc_if_needed, options.max_value_len.?);
errdefer {
if (name_token) |t| {
freeAllocated(allocator, t);
}
}
const field_name = switch (name_token.?) {
.object_end => break, // No more fields.
.string => |slice| slice,
.allocated_string => |slice| slice,
else => return error.UnexpectedToken,
};
inline for (structInfo.fields, 0..) |field, i| {
if (field.is_comptime) @compileError("comptime fields are not supported: " ++ @typeName(T) ++ "." ++ field.name);
if (std.mem.eql(u8, field.name, field_name)) {
// Free the name token now in case we're using an allocator that optimizes freeing the last allocated object.
// (Recursing into parseInternal() might trigger more allocations.)
freeAllocated(allocator, name_token.?);
name_token = null;
if (fields_seen[i]) {
switch (options.duplicate_field_behavior) {
.use_first => {
// Parse and then delete the redundant value.
// We don't want to skip the value, because we want type checking.
const ignored_value = try parseInternal(field.type, allocator, source, options);
parseFree(field.type, allocator, ignored_value);
break;
},
.@"error" => return error.DuplicateField,
.use_last => {
// Delete the stale value. We're about to get a new one.
parseFree(field.type, allocator, @field(r, field.name));
fields_seen[i] = false;
},
}
}
@field(r, field.name) = try parseInternal(field.type, allocator, source, options);
fields_seen[i] = true;
break;
}
} else {
// Didn't match anything.
freeAllocated(allocator, name_token.?);
if (options.ignore_unknown_fields) {
try source.skipValue();
} else {
return error.UnknownField;
}
}
}
inline for (structInfo.fields, 0..) |field, i| {
if (!fields_seen[i]) {
if (field.default_value) |default_ptr| {
const default = @ptrCast(*align(1) const field.type, default_ptr).*;
@field(r, field.name) = default;
} else {
return error.MissingField;
}
}
}
return r;
},
.Array => |arrayInfo| {
switch (try source.peekNextTokenType()) {
.array_begin => {
// Typical array.
return parseInternalArray(T, arrayInfo.child, arrayInfo.len, allocator, source, options);
},
.string => {
if (arrayInfo.child != u8) return error.UnexpectedToken;
// Fixed-length string.
var r: T = undefined;
var i: usize = 0;
while (true) {
switch (try source.next()) {
.string => |slice| {
if (i + slice.len != r.len) return error.LengthMismatch;
@memcpy(r[i..][0..slice.len], slice);
break;
},
.partial_string => |slice| {
if (i + slice.len > r.len) return error.LengthMismatch;
@memcpy(r[i..][0..slice.len], slice);
i += slice.len;
},
.partial_string_escaped_1 => |arr| {
if (i + arr.len > r.len) return error.LengthMismatch;
@memcpy(r[i..][0..arr.len], arr[0..]);
i += arr.len;
},
.partial_string_escaped_2 => |arr| {
if (i + arr.len > r.len) return error.LengthMismatch;
@memcpy(r[i..][0..arr.len], arr[0..]);
i += arr.len;
},
.partial_string_escaped_3 => |arr| {
if (i + arr.len > r.len) return error.LengthMismatch;
@memcpy(r[i..][0..arr.len], arr[0..]);
i += arr.len;
},
.partial_string_escaped_4 => |arr| {
if (i + arr.len > r.len) return error.LengthMismatch;
@memcpy(r[i..][0..arr.len], arr[0..]);
i += arr.len;
},
else => unreachable,
}
}
return r;
},
else => return error.UnexpectedToken,
}
},
.Vector => |vecInfo| {
switch (try source.peekNextTokenType()) {
.array_begin => {
return parseInternalArray(T, vecInfo.child, vecInfo.len, allocator, source, options);
},
else => return error.UnexpectedToken,
}
},
.Pointer => |ptrInfo| {
switch (ptrInfo.size) {
.One => {
const r: *ptrInfo.child = try allocator.create(ptrInfo.child);
errdefer allocator.destroy(r);
r.* = try parseInternal(ptrInfo.child, allocator, source, options);
return r;
},
.Slice => {
switch (try source.peekNextTokenType()) {
.array_begin => {
_ = try source.next();
// Typical array.
var arraylist = ArrayList(ptrInfo.child).init(allocator);
errdefer {
while (arraylist.popOrNull()) |v| {
parseFree(ptrInfo.child, allocator, v);
}
arraylist.deinit();
}
while (true) {
switch (try source.peekNextTokenType()) {
.array_end => {
_ = try source.next();
break;
},
else => {},
}
try arraylist.ensureUnusedCapacity(1);
arraylist.appendAssumeCapacity(try parseInternal(ptrInfo.child, allocator, source, options));
}
if (ptrInfo.sentinel) |some| {
const sentinel_value = @ptrCast(*align(1) const ptrInfo.child, some).*;
return try arraylist.toOwnedSliceSentinel(sentinel_value);
}
return try arraylist.toOwnedSlice();
},
.string => {
if (ptrInfo.child != u8) return error.UnexpectedToken;
// Dynamic length string.
if (ptrInfo.sentinel) |sentinel_ptr| {
// Use our own array list so we can append the sentinel.
var value_list = ArrayList(u8).init(allocator);
errdefer value_list.deinit();
_ = try source.allocNextIntoArrayList(&value_list, .alloc_always);
return try value_list.toOwnedSliceSentinel(@ptrCast(*const u8, sentinel_ptr).*);
}
switch (try source.nextAllocMax(allocator, .alloc_always, options.max_value_len.?)) {
.allocated_string => |slice| return slice,
else => unreachable,
}
},
else => return error.UnexpectedToken,
}
},
else => @compileError("Unable to parse into type '" ++ @typeName(T) ++ "'"),
}
},
else => @compileError("Unable to parse into type '" ++ @typeName(T) ++ "'"),
}
unreachable;
}
fn parseInternalArray(
comptime T: type,
comptime Child: type,
comptime len: comptime_int,
allocator: Allocator,
source: anytype,
options: ParseOptions,
) !T {
assert(.array_begin == try source.next());
var r: T = undefined;
var i: usize = 0;
errdefer {
// Without the len check `r[i]` is not allowed
if (len > 0) while (true) : (i -= 1) {
parseFree(Child, allocator, r[i]);
if (i == 0) break;
};
}
while (i < len) : (i += 1) {
r[i] = try parseInternal(Child, allocator, source, options);
}
if (.array_end != try source.next()) return error.UnexpectedToken;
return r;
}
fn freeAllocated(allocator: Allocator, token: Token) void {
switch (token) {
.allocated_number, .allocated_string => |slice| {
allocator.free(slice);
},
else => {},
}
}
/// Releases resources created by parseFromSlice() or parseFromTokenSource().
pub fn parseFree(comptime T: type, allocator: Allocator, value: T) void {
switch (@typeInfo(T)) {
.Bool, .Float, .ComptimeFloat, .Int, .ComptimeInt, .Enum => {},
.Optional => {
if (value) |v| {
return parseFree(@TypeOf(v), allocator, v);
}
},
.Union => |unionInfo| {
if (unionInfo.tag_type) |UnionTagType| {
inline for (unionInfo.fields) |u_field| {
if (value == @field(UnionTagType, u_field.name)) {
parseFree(u_field.type, allocator, @field(value, u_field.name));
break;
}
}
} else {
unreachable;
}
},
.Struct => |structInfo| {
inline for (structInfo.fields) |field| {
var should_free = true;
if (field.default_value) |default| {
switch (@typeInfo(field.type)) {
// We must not attempt to free pointers to struct default values
.Pointer => |fieldPtrInfo| {
const field_value = @field(value, field.name);
const field_ptr = switch (fieldPtrInfo.size) {
.One => field_value,
.Slice => field_value.ptr,
else => unreachable, // Other pointer types are not parseable
};
const field_addr = @ptrToInt(field_ptr);
const casted_default = @ptrCast(*const field.type, @alignCast(@alignOf(field.type), default)).*;
const default_ptr = switch (fieldPtrInfo.size) {
.One => casted_default,
.Slice => casted_default.ptr,
else => unreachable, // Other pointer types are not parseable
};
const default_addr = @ptrToInt(default_ptr);
if (field_addr == default_addr) {
should_free = false;
}
},
else => {},
}
}
if (should_free) {
parseFree(field.type, allocator, @field(value, field.name));
}
}
},
.Array => |arrayInfo| {
for (value) |v| {
parseFree(arrayInfo.child, allocator, v);
}
},
.Vector => |vecInfo| {
var i: usize = 0;
while (i < vecInfo.len) : (i += 1) {
parseFree(vecInfo.child, allocator, value[i]);
}
},
.Pointer => |ptrInfo| {
switch (ptrInfo.size) {
.One => {
parseFree(ptrInfo.child, allocator, value.*);
allocator.destroy(value);
},
.Slice => {
for (value) |v| {
parseFree(ptrInfo.child, allocator, v);
}
allocator.free(value);
},
else => unreachable,
}
},
else => unreachable,
}
}
test {
_ = @import("./static_test.zig");
}
|