summaryrefslogtreecommitdiff
path: root/db/json.cpp
blob: b89ff32c43d458ce74ddb3b8af4608c03df2915a (plain)
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
// json.cpp

/**
*    Copyright (C) 2008 10gen Inc.
*
*    This program is free software: you can redistribute it and/or  modify
*    it under the terms of the GNU Affero General Public License, version 3,
*    as published by the Free Software Foundation.
*
*    This program is distributed in the hope that it will be useful,
*    but WITHOUT ANY WARRANTY; without even the implied warranty of
*    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
*    GNU Affero General Public License for more details.
*
*    You should have received a copy of the GNU Affero General Public License
*    along with this program.  If not, see <http://www.gnu.org/licenses/>.
*/

#include "pch.h"

#define BOOST_SPIRIT_THREADSAFE
#if BOOST_VERSION >= 103800
#define BOOST_SPIRIT_USE_OLD_NAMESPACE
#include <boost/spirit/include/classic_core.hpp>
#include <boost/spirit/include/classic_loops.hpp>
#include <boost/spirit/include/classic_lists.hpp>
#else
#include <boost/spirit/core.hpp>
#include <boost/spirit/utility/loops.hpp>
#include <boost/spirit/utility/lists.hpp>
#endif
#undef assert
#define assert MONGO_assert

#include "json.h"
#include "../bson/util/builder.h"
#include "../util/base64.h"
#include "../util/hex.h"


using namespace boost::spirit;

namespace mongo {

    struct ObjectBuilder : boost::noncopyable {
        ~ObjectBuilder() {
            unsigned i = builders.size();
            if ( i ) {
                i--;
                for ( ; i>=1; i-- ) {
                    if ( builders[i] ) {
                        builders[i]->done();
                    }
                }
            }
        }
        BSONObjBuilder *back() {
            return builders.back().get();
        }
        // Storage for field names of elements within builders.back().
        const char *fieldName() {
            return fieldNames.back().c_str();
        }
        bool empty() const {
            return builders.size() == 0;
        }
        void init() {
            boost::shared_ptr< BSONObjBuilder > b( new BSONObjBuilder() );
            builders.push_back( b );
            fieldNames.push_back( "" );
            indexes.push_back( 0 );
        }
        void pushObject( const char *fieldName ) {
            boost::shared_ptr< BSONObjBuilder > b( new BSONObjBuilder( builders.back()->subobjStart( fieldName ) ) );
            builders.push_back( b );
            fieldNames.push_back( "" );
            indexes.push_back( 0 );
        }
        void pushArray( const char *fieldName ) {
            boost::shared_ptr< BSONObjBuilder > b( new BSONObjBuilder( builders.back()->subarrayStart( fieldName ) ) );
            builders.push_back( b );
            fieldNames.push_back( "" );
            indexes.push_back( 0 );
        }
        BSONObj pop() {
            BSONObj ret;
            if ( back()->owned() )
                ret = back()->obj();
            else
                ret = back()->done();
            builders.pop_back();
            fieldNames.pop_back();
            indexes.pop_back();
            return ret;
        }
        void nameFromIndex() {
            fieldNames.back() = BSONObjBuilder::numStr( indexes.back() );
        }
        string popString() {
            string ret = ss.str();
            ss.str( "" );
            return ret;
        }
        // Cannot use auto_ptr because its copy constructor takes a non const reference.
        vector< boost::shared_ptr< BSONObjBuilder > > builders;
        vector< string > fieldNames;
        vector< int > indexes;
        stringstream ss;
        string ns;
        OID oid;
        string binData;
        BinDataType binDataType;
        string regex;
        string regexOptions;
        Date_t date;
    };

    struct objectStart {
        objectStart( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char &c ) const {
            if ( b.empty() )
                b.init();
            else
                b.pushObject( b.fieldName() );
        }
        ObjectBuilder &b;
    };

    struct arrayStart {
        arrayStart( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char &c ) const {
            b.pushArray( b.fieldName() );
            b.nameFromIndex();
        }
        ObjectBuilder &b;
    };

    struct arrayNext {
        arrayNext( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char &c ) const {
            ++b.indexes.back();
            b.nameFromIndex();
        }
        ObjectBuilder &b;
    };

    struct ch {
        ch( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char c ) const {
            b.ss << c;
        }
        ObjectBuilder &b;
    };

    struct chE {
        chE( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char c ) const {
            char o = '\0';
            switch ( c ) {
            case '\"':
                o = '\"';
                break;
            case '\'':
                o = '\'';
                break;
            case '\\':
                o = '\\';
                break;
            case '/':
                o = '/';
                break;
            case 'b':
                o = '\b';
                break;
            case 'f':
                o = '\f';
                break;
            case 'n':
                o = '\n';
                break;
            case 'r':
                o = '\r';
                break;
            case 't':
                o = '\t';
                break;
            case 'v':
                o = '\v';
                break;
            default:
                assert( false );
            }
            b.ss << o;
        }
        ObjectBuilder &b;
    };

    struct chU {
        chU( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            unsigned char first = fromHex( start );
            unsigned char second = fromHex( start + 2 );
            if ( first == 0 && second < 0x80 )
                b.ss << second;
            else if ( first < 0x08 ) {
                b.ss << char( 0xc0 | ( ( first << 2 ) | ( second >> 6 ) ) );
                b.ss << char( 0x80 | ( ~0xc0 & second ) );
            }
            else {
                b.ss << char( 0xe0 | ( first >> 4 ) );
                b.ss << char( 0x80 | ( ~0xc0 & ( ( first << 2 ) | ( second >> 6 ) ) ) );
                b.ss << char( 0x80 | ( ~0xc0 & second ) );
            }
        }
        ObjectBuilder &b;
    };

    struct chClear {
        chClear( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char c ) const {
            b.popString();
        }
        ObjectBuilder &b;
    };

    struct fieldNameEnd {
        fieldNameEnd( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            string name = b.popString();
            massert( 10338 ,  "Invalid use of reserved field name",
                     name != "$oid" &&
                     name != "$binary" &&
                     name != "$type" &&
                     name != "$date" &&
                     name != "$regex" &&
                     name != "$options" );
            b.fieldNames.back() = name;
        }
        ObjectBuilder &b;
    };

    struct unquotedFieldNameEnd {
        unquotedFieldNameEnd( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            string name( start, end );
            b.fieldNames.back() = name;
        }
        ObjectBuilder &b;
    };

    struct stringEnd {
        stringEnd( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.back()->append( b.fieldName(), b.popString() );
        }
        ObjectBuilder &b;
    };

    struct numberValue {
        numberValue( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            // We re-parse the numeric string here because spirit parsing of strings
            // to doubles produces different results from strtod in some cases and
            // we want to use strtod to ensure consistency with other string to
            // double conversions in our code.
            b.back()->append( b.fieldName(), strtod( start, 0 ) );
        }
        ObjectBuilder &b;
    };

    struct intValue {
        intValue( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( long long num ) const {
            if (num >= numeric_limits<int>::min() && num <= numeric_limits<int>::max())
                b.back()->append( b.fieldName(), (int)num );
            else
                b.back()->append( b.fieldName(), num );
        }
        ObjectBuilder &b;
    };

    struct subobjectEnd {
        subobjectEnd( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.pop();
        }
        ObjectBuilder &b;
    };

    struct arrayEnd {
        arrayEnd( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.pop();
        }
        ObjectBuilder &b;
    };

    struct trueValue {
        trueValue( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.back()->appendBool( b.fieldName(), true );
        }
        ObjectBuilder &b;
    };

    struct falseValue {
        falseValue( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.back()->appendBool( b.fieldName(), false );
        }
        ObjectBuilder &b;
    };

    struct nullValue {
        nullValue( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.back()->appendNull( b.fieldName() );
        }
        ObjectBuilder &b;
    };

    struct undefinedValue {
        undefinedValue( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.back()->appendUndefined( b.fieldName() );
        }
        ObjectBuilder &b;
    };
    
    struct dbrefNS {
        dbrefNS( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.ns = b.popString();
        }
        ObjectBuilder &b;
    };

// NOTE s must be 24 characters.
    OID stringToOid( const char *s ) {
        OID oid;
        char *oidP = (char *)( &oid );
        for ( int i = 0; i < 12; ++i )
            oidP[ i ] = fromHex( s + ( i * 2 ) );
        return oid;
    }

    struct oidValue {
        oidValue( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.oid = stringToOid( start );
        }
        ObjectBuilder &b;
    };

    struct dbrefEnd {
        dbrefEnd( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.back()->appendDBRef( b.fieldName(), b.ns, b.oid );
        }
        ObjectBuilder &b;
    };

    struct oidEnd {
        oidEnd( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.back()->appendOID( b.fieldName(), &b.oid );
        }
        ObjectBuilder &b;
    };

    struct binDataBinary {
        binDataBinary( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            massert( 10339 ,  "Badly formatted bindata", ( end - start ) % 4 == 0 );
            string encoded( start, end );
            b.binData = base64::decode( encoded );
        }
        ObjectBuilder &b;
    };

    struct binDataType {
        binDataType( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.binDataType = BinDataType( fromHex( start ) );
        }
        ObjectBuilder &b;
    };

    struct binDataEnd {
        binDataEnd( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.back()->appendBinData( b.fieldName(), b.binData.length(),
                                     b.binDataType, b.binData.data() );
        }
        ObjectBuilder &b;
    };

    struct dateValue {
        dateValue( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( Date_t v ) const {
            b.date = v;
        }
        ObjectBuilder &b;
    };

    struct dateEnd {
        dateEnd( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.back()->appendDate( b.fieldName(), b.date );
        }
        ObjectBuilder &b;
    };

    struct regexValue {
        regexValue( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.regex = b.popString();
        }
        ObjectBuilder &b;
    };

    struct regexOptions {
        regexOptions( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.regexOptions = string( start, end );
        }
        ObjectBuilder &b;
    };

    struct regexEnd {
        regexEnd( ObjectBuilder &_b ) : b( _b ) {}
        void operator() ( const char *start, const char *end ) const {
            b.back()->appendRegex( b.fieldName(), b.regex, b.regexOptions );
        }
        ObjectBuilder &b;
    };

// One gotcha with this parsing library is probably best ilustrated with an
// example.  Say we have a production like this:
// z = ( ch_p( 'a' )[ foo ] >> ch_p( 'b' ) ) | ( ch_p( 'a' )[ foo ] >> ch_p( 'c' ) );
// On input "ac", action foo() will be called twice -- once as the parser tries
// to match "ab", again as the parser successfully matches "ac".  Sometimes
// the grammar can be modified to eliminate these situations.  Here, for example:
// z = ch_p( 'a' )[ foo ] >> ( ch_p( 'b' ) | ch_p( 'c' ) );
// However, this is not always possible.  In my implementation I've tried to
// stick to the following pattern: store fields fed to action callbacks
// temporarily as ObjectBuilder members, then append to a BSONObjBuilder once
// the parser has completely matched a nonterminal and won't backtrack.  It's
// worth noting here that this parser follows a short-circuit convention.  So,
// in the original z example on line 3, if the input was "ab", foo() would only
// be called once.
    struct JsonGrammar : public grammar< JsonGrammar > {
    public:
        JsonGrammar( ObjectBuilder &_b ) : b( _b ) {}

        template < typename ScannerT >
        struct definition {
            definition( JsonGrammar const &self ) {
                object = ch_p( '{' )[ objectStart( self.b ) ] >> !members >> '}';
                members = list_p((fieldName >> ':' >> value) , ',');
                fieldName =
                    str[ fieldNameEnd( self.b ) ] |
                    singleQuoteStr[ fieldNameEnd( self.b ) ] |
                    unquotedFieldName[ unquotedFieldNameEnd( self.b ) ];
                array = ch_p( '[' )[ arrayStart( self.b ) ] >> !elements >> ']';
                elements = list_p(value, ch_p(',')[arrayNext( self.b )]);
                value =
                    str[ stringEnd( self.b ) ] |
                    number[ numberValue( self.b ) ] |
                    integer |
                    array[ arrayEnd( self.b ) ] |
                    lexeme_d[ str_p( "true" ) ][ trueValue( self.b ) ] |
                    lexeme_d[ str_p( "false" ) ][ falseValue( self.b ) ] |
                    lexeme_d[ str_p( "null" ) ][ nullValue( self.b ) ] |
                    lexeme_d[ str_p( "undefined" ) ][ undefinedValue( self.b ) ] |
                    singleQuoteStr[ stringEnd( self.b ) ] |
                    date[ dateEnd( self.b ) ] |
                    oid[ oidEnd( self.b ) ] |
                    bindata[ binDataEnd( self.b ) ] |
                    dbref[ dbrefEnd( self.b ) ] |
                    regex[ regexEnd( self.b ) ] |
                    object[ subobjectEnd( self.b ) ] ;
                // NOTE lexeme_d and rules don't mix well, so we have this mess.
                // NOTE We use range_p rather than cntrl_p, because the latter is locale dependent.
                str = lexeme_d[ ch_p( '"' )[ chClear( self.b ) ] >>
                                *( ( ch_p( '\\' ) >>
                                     (
                                         ch_p( 'b' )[ chE( self.b ) ] |
                                         ch_p( 'f' )[ chE( self.b ) ] |
                                         ch_p( 'n' )[ chE( self.b ) ] |
                                         ch_p( 'r' )[ chE( self.b ) ] |
                                         ch_p( 't' )[ chE( self.b ) ] |
                                         ch_p( 'v' )[ chE( self.b ) ] |
                                         ( ch_p( 'u' ) >> ( repeat_p( 4 )[ xdigit_p ][ chU( self.b ) ] ) ) |
                                         ( ~ch_p('x') & (~range_p('0','9'))[ ch( self.b ) ] ) // hex and octal aren't supported
                                     )
                                   ) |
                                   ( ~range_p( 0x00, 0x1f ) & ~ch_p( '"' ) & ( ~ch_p( '\\' ) )[ ch( self.b ) ] ) ) >> '"' ];

                singleQuoteStr = lexeme_d[ ch_p( '\'' )[ chClear( self.b ) ] >>
                                           *( ( ch_p( '\\' ) >>
                                                (
                                                    ch_p( 'b' )[ chE( self.b ) ] |
                                                    ch_p( 'f' )[ chE( self.b ) ] |
                                                    ch_p( 'n' )[ chE( self.b ) ] |
                                                    ch_p( 'r' )[ chE( self.b ) ] |
                                                    ch_p( 't' )[ chE( self.b ) ] |
                                                    ch_p( 'v' )[ chE( self.b ) ] |
                                                    ( ch_p( 'u' ) >> ( repeat_p( 4 )[ xdigit_p ][ chU( self.b ) ] ) ) |
                                                    ( ~ch_p('x') & (~range_p('0','9'))[ ch( self.b ) ] ) // hex and octal aren't supported
                                                )
                                              ) |
                                              ( ~range_p( 0x00, 0x1f ) & ~ch_p( '\'' ) & ( ~ch_p( '\\' ) )[ ch( self.b ) ] ) ) >> '\'' ];

                // real_p accepts numbers with nonsignificant zero prefixes, which
                // aren't allowed in JSON.  Oh well.
                number = strict_real_p;

                static int_parser<long long, 10,  1, numeric_limits<long long>::digits10 + 1> long_long_p;
                integer = long_long_p[ intValue(self.b) ];

                // We allow a subset of valid js identifier names here.
                unquotedFieldName = lexeme_d[ ( alpha_p | ch_p( '$' ) | ch_p( '_' ) ) >> *( ( alnum_p | ch_p( '$' ) | ch_p( '_'  )) ) ];

                dbref = dbrefS | dbrefT;
                dbrefS = ch_p( '{' ) >> "\"$ref\"" >> ':' >>
                         str[ dbrefNS( self.b ) ] >> ',' >> "\"$id\"" >> ':' >> quotedOid >> '}';
                dbrefT = str_p( "Dbref" ) >> '(' >> str[ dbrefNS( self.b ) ] >> ',' >>
                         quotedOid >> ')';

                oid = oidS | oidT;
                oidS = ch_p( '{' ) >> "\"$oid\"" >> ':' >> quotedOid >> '}';
                oidT = str_p( "ObjectId" ) >> '(' >> quotedOid >> ')';

                quotedOid = lexeme_d[ '"' >> ( repeat_p( 24 )[ xdigit_p ] )[ oidValue( self.b ) ] >> '"' ];

                bindata = ch_p( '{' ) >> "\"$binary\"" >> ':' >>
                          lexeme_d[ '"' >> ( *( range_p( 'A', 'Z' ) | range_p( 'a', 'z' ) | range_p( '0', '9' ) | ch_p( '+' ) | ch_p( '/' ) ) >> *ch_p( '=' ) )[ binDataBinary( self.b ) ] >> '"' ] >> ',' >> "\"$type\"" >> ':' >>
                          lexeme_d[ '"' >> ( repeat_p( 2 )[ xdigit_p ] )[ binDataType( self.b ) ] >> '"' ] >> '}';

                // TODO: this will need to use a signed parser at some point
                date = dateS | dateT;
                dateS = ch_p( '{' ) >> "\"$date\"" >> ':' >> uint_parser< Date_t >()[ dateValue( self.b ) ] >> '}';
                dateT = !str_p("new") >> str_p( "Date" ) >> '(' >> uint_parser< Date_t >()[ dateValue( self.b ) ] >> ')';

                regex = regexS | regexT;
                regexS = ch_p( '{' ) >> "\"$regex\"" >> ':' >> str[ regexValue( self.b ) ] >> ',' >> "\"$options\"" >> ':' >> lexeme_d[ '"' >> ( *( alpha_p ) )[ regexOptions( self.b ) ] >> '"' ] >> '}';
                // FIXME Obviously it would be nice to unify this with str.
                regexT = lexeme_d[ ch_p( '/' )[ chClear( self.b ) ] >>
                                   *( ( ch_p( '\\' ) >>
                                        ( ch_p( '"' )[ chE( self.b ) ] |
                                          ch_p( '\\' )[ chE( self.b ) ] |
                                          ch_p( '/' )[ chE( self.b ) ] |
                                          ch_p( 'b' )[ chE( self.b ) ] |
                                          ch_p( 'f' )[ chE( self.b ) ] |
                                          ch_p( 'n' )[ chE( self.b ) ] |
                                          ch_p( 'r' )[ chE( self.b ) ] |
                                          ch_p( 't' )[ chE( self.b ) ] |
                                          ( ch_p( 'u' ) >> ( repeat_p( 4 )[ xdigit_p ][ chU( self.b ) ] ) ) ) ) |
                                      ( ~range_p( 0x00, 0x1f ) & ~ch_p( '/' ) & ( ~ch_p( '\\' ) )[ ch( self.b ) ] ) ) >> str_p( "/" )[ regexValue( self.b ) ]
                                   >> ( *( ch_p( 'i' ) | ch_p( 'g' ) | ch_p( 'm' ) ) )[ regexOptions( self.b ) ] ];
            }
            rule< ScannerT > object, members, array, elements, value, str, number, integer,
                  dbref, dbrefS, dbrefT, oid, oidS, oidT, bindata, date, dateS, dateT,
                  regex, regexS, regexT, quotedOid, fieldName, unquotedFieldName, singleQuoteStr;
            const rule< ScannerT > &start() const {
                return object;
            }
        };
        ObjectBuilder &b;
    };

    BSONObj fromjson( const char *str , int* len) {
        if ( str[0] == '\0' ) {
            if (len) *len = 0;
            return BSONObj();
        }

        ObjectBuilder b;
        JsonGrammar parser( b );
        parse_info<> result = parse( str, parser, space_p );
        if (len) {
            *len = result.stop - str;
        }
        else if ( !result.full ) {
            int limit = strnlen(result.stop , 10);
            if (limit == -1) limit = 10;
            msgasserted(10340, "Failure parsing JSON string near: " + string( result.stop, limit ));
        }
        BSONObj ret = b.pop();
        assert( b.empty() );
        return ret;
    }

    BSONObj fromjson( const string &str ) {
        return fromjson( str.c_str() );
    }

} // namespace mongo