2 Copyright (C) 2009 Stig Brautaset. All rights reserved.
4 Redistribution and use in source and binary forms, with or without
5 modification, are permitted provided that the following conditions are met:
7 * Redistributions of source code must retain the above copyright notice, this
8 list of conditions and the following disclaimer.
10 * Redistributions in binary form must reproduce the above copyright notice,
11 this list of conditions and the following disclaimer in the documentation
12 and/or other materials provided with the distribution.
14 * Neither the name of the author nor the names of its contributors may be used
15 to endorse or promote products derived from this software without specific
16 prior written permission.
18 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
19 AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
20 IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
21 DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE
22 FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
23 DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
24 SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
25 CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
26 OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
27 OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
30 #import "SBJsonParser.h"
32 @interface SBJsonParser ()
34 - (BOOL)scanValue:(NSObject **)o;
36 - (BOOL)scanRestOfArray:(NSMutableArray **)o;
37 - (BOOL)scanRestOfDictionary:(NSMutableDictionary **)o;
38 - (BOOL)scanRestOfNull:(NSNull **)o;
39 - (BOOL)scanRestOfFalse:(NSNumber **)o;
40 - (BOOL)scanRestOfTrue:(NSNumber **)o;
41 - (BOOL)scanRestOfString:(NSMutableString **)o;
43 // Cannot manage without looking at the first digit
44 - (BOOL)scanNumber:(NSNumber **)o;
46 - (BOOL)scanHexQuad:(unichar *)x;
47 - (BOOL)scanUnicodeChar:(unichar *)x;
53 #define skipWhitespace(c) while (isspace(*c)) c++
54 #define skipDigits(c) while (isdigit(*c)) c++
57 @implementation SBJsonParser
59 static char ctrl[0x22];
65 for (int i = 1; i < 0x20; i++)
71 @deprecated This exists in order to provide fragment support in older APIs in one more version.
72 It should be removed in the next major version.
74 - (id)fragmentWithString:(id)repr {
75 [self clearErrorTrace];
78 [self addErrorWithCode:EINPUT description:@"Input was 'nil'"];
83 c = [repr UTF8String];
86 if (![self scanValue:&o]) {
90 // We found some valid JSON. But did it also contain something else?
91 if (![self scanIsAtEnd]) {
92 [self addErrorWithCode:ETRAILGARBAGE description:@"Garbage after JSON"];
96 NSAssert1(o, @"Should have a valid object from %@", repr);
100 - (id)objectWithString:(NSString *)repr {
102 id o = [self fragmentWithString:repr];
106 // Check that the object we've found is a valid JSON container.
107 if (![o isKindOfClass:[NSDictionary class]] && ![o isKindOfClass:[NSArray class]]) {
108 [self addErrorWithCode:EFRAGMENT description:@"Valid fragment, but not JSON"];
116 In contrast to the public methods, it is an error to omit the error parameter here.
118 - (BOOL)scanValue:(NSObject **)o
124 return [self scanRestOfDictionary:(NSMutableDictionary **)o];
127 return [self scanRestOfArray:(NSMutableArray **)o];
130 return [self scanRestOfString:(NSMutableString **)o];
133 return [self scanRestOfFalse:(NSNumber **)o];
136 return [self scanRestOfTrue:(NSNumber **)o];
139 return [self scanRestOfNull:(NSNull **)o];
143 c--; // cannot verify number correctly without the first character
144 return [self scanNumber:(NSNumber **)o];
147 [self addErrorWithCode:EPARSENUM description: @"Leading + disallowed in number"];
151 [self addErrorWithCode:EEOF description:@"Unexpected end of string"];
155 [self addErrorWithCode:EPARSE description: @"Unrecognised leading character"];
160 NSAssert(0, @"Should never get here");
164 - (BOOL)scanRestOfTrue:(NSNumber **)o
166 if (!strncmp(c, "rue", 3)) {
168 *o = [NSNumber numberWithBool:YES];
171 [self addErrorWithCode:EPARSE description:@"Expected 'true'"];
175 - (BOOL)scanRestOfFalse:(NSNumber **)o
177 if (!strncmp(c, "alse", 4)) {
179 *o = [NSNumber numberWithBool:NO];
182 [self addErrorWithCode:EPARSE description: @"Expected 'false'"];
186 - (BOOL)scanRestOfNull:(NSNull **)o {
187 if (!strncmp(c, "ull", 3)) {
192 [self addErrorWithCode:EPARSE description: @"Expected 'null'"];
196 - (BOOL)scanRestOfArray:(NSMutableArray **)o {
197 if (maxDepth && ++depth > maxDepth) {
198 [self addErrorWithCode:EDEPTH description: @"Nested too deep"];
202 *o = [NSMutableArray arrayWithCapacity:8];
208 if (*c == ']' && c++) {
213 if (![self scanValue:&v]) {
214 [self addErrorWithCode:EPARSE description:@"Expected value while parsing array"];
221 if (*c == ',' && c++) {
224 [self addErrorWithCode:ETRAILCOMMA description: @"Trailing comma disallowed in array"];
230 [self addErrorWithCode:EEOF description: @"End of input while parsing array"];
234 - (BOOL)scanRestOfDictionary:(NSMutableDictionary **)o
236 if (maxDepth && ++depth > maxDepth) {
237 [self addErrorWithCode:EDEPTH description: @"Nested too deep"];
241 *o = [NSMutableDictionary dictionaryWithCapacity:7];
247 if (*c == '}' && c++) {
252 if (!(*c == '\"' && c++ && [self scanRestOfString:&k])) {
253 [self addErrorWithCode:EPARSE description: @"Object key string expected"];
259 [self addErrorWithCode:EPARSE description: @"Expected ':' separating key and value"];
264 if (![self scanValue:&v]) {
265 NSString *string = [NSString stringWithFormat:@"Object value expected for key: %@", k];
266 [self addErrorWithCode:EPARSE description: string];
270 [*o setObject:v forKey:k];
273 if (*c == ',' && c++) {
276 [self addErrorWithCode:ETRAILCOMMA description: @"Trailing comma disallowed in object"];
282 [self addErrorWithCode:EEOF description: @"End of input while parsing object"];
286 - (BOOL)scanRestOfString:(NSMutableString **)o
288 *o = [NSMutableString stringWithCapacity:16];
290 // First see if there's a portion we can grab in one go.
291 // Doing this caused a massive speedup on the long string.
292 size_t len = strcspn(c, ctrl);
295 id t = [[NSString alloc] initWithBytesNoCopy:(char*)c
297 encoding:NSUTF8StringEncoding
310 } else if (*c == '\\') {
318 case 'b': uc = '\b'; break;
319 case 'n': uc = '\n'; break;
320 case 'r': uc = '\r'; break;
321 case 't': uc = '\t'; break;
322 case 'f': uc = '\f'; break;
326 if (![self scanUnicodeChar:&uc]) {
327 [self addErrorWithCode:EUNICODE description: @"Broken unicode character"];
333 [self addErrorWithCode:EESCAPE description: [NSString stringWithFormat:@"Illegal escape sequence '0x%x'", uc]];
337 CFStringAppendCharacters((CFMutableStringRef)*o, &uc, 1);
340 } else if (*c < 0x20) {
341 [self addErrorWithCode:ECTRL description: [NSString stringWithFormat:@"Unescaped control character '0x%x'", *c]];
345 NSLog(@"should not be able to get here");
349 [self addErrorWithCode:EEOF description:@"Unexpected EOF while parsing string"];
353 - (BOOL)scanUnicodeChar:(unichar *)x
357 if (![self scanHexQuad:&hi]) {
358 [self addErrorWithCode:EUNICODE description: @"Missing hex quad"];
362 if (hi >= 0xd800) { // high surrogate char?
363 if (hi < 0xdc00) { // yes - expect a low char
365 if (!(*c == '\\' && ++c && *c == 'u' && ++c && [self scanHexQuad:&lo])) {
366 [self addErrorWithCode:EUNICODE description: @"Missing low character in surrogate pair"];
370 if (lo < 0xdc00 || lo >= 0xdfff) {
371 [self addErrorWithCode:EUNICODE description:@"Invalid low surrogate char"];
375 hi = (hi - 0xd800) * 0x400 + (lo - 0xdc00) + 0x10000;
377 } else if (hi < 0xe000) {
378 [self addErrorWithCode:EUNICODE description:@"Invalid high character in surrogate pair"];
387 - (BOOL)scanHexQuad:(unichar *)x
390 for (int i = 0; i < 4; i++) {
393 int d = (uc >= '0' && uc <= '9')
394 ? uc - '0' : (uc >= 'a' && uc <= 'f')
395 ? (uc - 'a' + 10) : (uc >= 'A' && uc <= 'F')
396 ? (uc - 'A' + 10) : -1;
398 [self addErrorWithCode:EUNICODE description:@"Missing hex digit in quad"];
407 - (BOOL)scanNumber:(NSNumber **)o
411 // The logic to test for validity of the number formatting is relicensed
412 // from JSON::XS with permission from its author Marc Lehmann.
413 // (Available at the CPAN: http://search.cpan.org/dist/JSON-XS/ .)
418 if ('0' == *c && c++) {
420 [self addErrorWithCode:EPARSENUM description: @"Leading 0 disallowed in number"];
424 } else if (!isdigit(*c) && c != ns) {
425 [self addErrorWithCode:EPARSENUM description: @"No digits after initial minus"];
433 if ('.' == *c && c++) {
436 [self addErrorWithCode:EPARSENUM description: @"No digits after decimal point"];
443 if ('e' == *c || 'E' == *c) {
446 if ('-' == *c || '+' == *c)
450 [self addErrorWithCode:EPARSENUM description: @"No digits after exponent"];
456 id str = [[NSString alloc] initWithBytesNoCopy:(char*)ns
458 encoding:NSUTF8StringEncoding
461 if (str && (*o = [NSDecimalNumber decimalNumberWithString:str]))
464 [self addErrorWithCode:EPARSENUM description: @"Failed creating decimal instance"];