sqlite/src/json.c at master · codegear/sqlite

History

5467 lines (5271 loc) · 164 KB

Raw

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

692

693

694

695

696

697

698

699

700

701

702

703

704

705

706

707

708

709

710

711

712

713

714

715

716

717

718

719

720

721

722

723

724

725

726

727

728

729

730

731

732

733

734

735

736

737

738

739

740

741

742

743

744

745

746

747

748

749

750

751

752

753

754

755

756

757

758

759

760

761

762

763

764

765

766

767

768

769

770

771

772

773

774

775

776

777

778

779

780

781

782

783

784

785

786

787

788

789

790

791

792

793

794

795

796

797

798

799

800

801

802

803

804

805

806

807

808

809

810

811

812

813

814

815

816

817

818

819

820

821

822

823

824

825

826

827

828

829

830

831

832

833

834

835

836

837

838

839

840

841

842

843

844

845

846

847

848

849

850

851

852

853

854

855

856

857

858

859

860

861

862

863

864

865

866

867

868

869

870

871

872

873

874

875

876

877

878

879

880

881

882

883

884

885

886

887

888

889

890

891

892

893

894

895

896

897

898

899

900

901

902

903

904

905

906

907

908

909

910

911

912

913

914

915

916

917

918

919

920

921

922

923

924

925

926

927

928

929

930

931

932

933

934

935

936

937

938

939

940

941

942

943

944

945

946

947

948

949

950

951

952

953

954

955

956

957

958

959

960

961

962

963

964

965

966

967

968

969

970

971

972

973

974

975

976

977

978

979

980

981

982

983

984

985

986

987

988

989

990

991

992

993

994

995

996

997

998

999

1000

** 2015-08-12

** The author disclaims copyright to this source code. In place of

** a legal notice, here is a blessing:

** May you do good and not evil.

** May you find forgiveness for yourself and forgive others.

** May you share freely, never taking more than you give.

******************************************************************************

** SQLite JSON functions.

** This file began as an extension in ext/misc/json1.c in 2015. That

** extension proved so useful that it has now been moved into the core.

** The original design stored all JSON as pure text, canonical RFC-8259.

** Support for JSON-5 extensions was added with version 3.42.0 (2023-05-16).

** All generated JSON text still conforms strictly to RFC-8259, but text

** with JSON-5 extensions is accepted as input.

** Beginning with version 3.45.0 (circa 2024-01-01), these routines also

** accept BLOB values that have JSON encoded using a binary representation

** called "JSONB". The name JSONB comes from PostgreSQL, however the on-disk

** format SQLite JSONB is completely different and incompatible with

** PostgreSQL JSONB.

** Decoding and interpreting JSONB is still O(N) where N is the size of

** the input, the same as text JSON. However, the constant of proportionality

** for JSONB is much smaller due to faster parsing. The size of each

** element in JSONB is encoded in its header, so there is no need to search

** for delimiters using persnickety syntax rules. JSONB seems to be about

** 3x faster than text JSON as a result. JSONB is also tends to be slightly

** smaller than text JSON, by 5% or 10%, but there are corner cases where

** JSONB can be slightly larger. So you are not far mistaken to say that

** a JSONB blob is the same size as the equivalent RFC-8259 text.

** THE JSONB ENCODING:

** Every JSON element is encoded in JSONB as a header and a payload.

** The header is between 1 and 9 bytes in size. The payload is zero

** or more bytes.

** The lower 4 bits of the first byte of the header determines the

** element type:

** 0: NULL

** 1: TRUE

** 2: FALSE

** 3: INT -- RFC-8259 integer literal

** 4: INT5 -- JSON5 integer literal

** 5: FLOAT -- RFC-8259 floating point literal

** 6: FLOAT5 -- JSON5 floating point literal

** 7: TEXT -- Text literal acceptable to both SQL and JSON

** 8: TEXTJ -- Text containing RFC-8259 escapes

** 9: TEXT5 -- Text containing JSON5 and/or RFC-8259 escapes

** 10: TEXTRAW -- Text containing unescaped syntax characters

** 11: ARRAY

** 12: OBJECT

** The other three possible values (13-15) are reserved for future

** enhancements.

** The upper 4 bits of the first byte determine the size of the header

** and sometimes also the size of the payload. If X is the first byte

** of the element and if X>>4 is between 0 and 11, then the payload

** will be that many bytes in size and the header is exactly one byte

** in size. Other four values for X>>4 (12-15) indicate that the header

** is more than one byte in size and that the payload size is determined

** by the remainder of the header, interpreted as a unsigned big-endian

** integer.

** Value of X>>4 Size integer Total header size

** ------------- -------------------- -----------------

** 12 1 byte (0-255) 2

** 13 2 byte (0-65535) 3

** 14 4 byte (0-4294967295) 5

** 15 8 byte (0-1.8e19) 9

** The payload size need not be expressed in its minimal form. For example,

** if the payload size is 10, the size can be expressed in any of 5 different

** ways: (1) (X>>4)==10, (2) (X>>4)==12 following by on 0x0a byte,

** (3) (X>>4)==13 followed by 0x00 and 0x0a, (4) (X>>4)==14 followed by

** 0x00 0x00 0x00 0x0a, or (5) (X>>4)==15 followed by 7 bytes of 0x00 and

** a single byte of 0x0a. The shorter forms are preferred, of course, but

** sometimes when generating JSONB, the payload size is not known in advance

** and it is convenient to reserve sufficient header space to cover the

** largest possible payload size and then come back later and patch up

** the size when it becomes known, resulting in a non-minimal encoding.

** The value (X>>4)==15 is not actually used in the current implementation

** (as SQLite is currently unable handle BLOBs larger than about 2GB)

** but is included in the design to allow for future enhancements.

** The payload follows the header. NULL, TRUE, and FALSE have no payload and

** their payload size must always be zero. The payload for INT, INT5,

** FLOAT, FLOAT5, TEXT, TEXTJ, TEXT5, and TEXTROW is text. Note that the

** "..." or '...' delimiters are omitted from the various text encodings.

** The payload for ARRAY and OBJECT is a list of additional elements that

** are the content for the array or object. The payload for an OBJECT

** must be an even number of elements. The first element of each pair is

** the label and must be of type TEXT, TEXTJ, TEXT5, or TEXTRAW.

** A valid JSONB blob consists of a single element, as described above.

** Usually this will be an ARRAY or OBJECT element which has many more

** elements as its content. But the overall blob is just a single element.

** Input validation for JSONB blobs simply checks that the element type

** code is between 0 and 12 and that the total size of the element

** (header plus payload) is the same as the size of the BLOB. If those

** checks are true, the BLOB is assumed to be JSONB and processing continues.

** Errors are only raised if some other miscoding is discovered during

** processing.

** Additional information can be found in the doc/jsonb.md file of the

** canonical SQLite source tree.

#ifndef SQLITE_OMIT_JSON

#include "sqliteInt.h"

/* JSONB element types

#define JSONB_NULL 0 /* "null" */

#define JSONB_TRUE 1 /* "true" */

#define JSONB_FALSE 2 /* "false" */

#define JSONB_INT 3 /* integer acceptable to JSON and SQL */

#define JSONB_INT5 4 /* integer in 0x000 notation */

#define JSONB_FLOAT 5 /* float acceptable to JSON and SQL */

#define JSONB_FLOAT5 6 /* float with JSON5 extensions */

#define JSONB_TEXT 7 /* Text compatible with both JSON and SQL */

#define JSONB_TEXTJ 8 /* Text with JSON escapes */

#define JSONB_TEXT5 9 /* Text with JSON-5 escape */

#define JSONB_TEXTRAW 10 /* SQL text that needs escaping for JSON */

#define JSONB_ARRAY 11 /* An array */

#define JSONB_OBJECT 12 /* An object */

/* Human-readable names for the JSONB values. The index for each

** string must correspond to the JSONB_* integer above.

static const char * const jsonbType[] = {

"null", "true", "false", "integer", "integer",

"real", "real", "text", "text", "text",

"text", "array", "object", "", "", "", ""

};

** Growing our own isspace() routine this way is twice as fast as

** the library isspace() function, resulting in a 7% overall performance

** increase for the text-JSON parser. (Ubuntu14.10 gcc 4.8.4 x64 with -Os).

static const char jsonIsSpace[] = {

0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0,

0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,

1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,

0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,

};

#define jsonIsspace(x) (jsonIsSpace[(unsigned char)x])

** The set of all space characters recognized by jsonIsspace().

** Useful as the second argument to strspn().

static const char jsonSpaces[] = "\011\012\015\040";

** Characters that are special to JSON. Control characters,

** '"' and '\\' and '\''. Actually, '\'' is not special to

** canonical JSON, but it is special in JSON-5, so we include

** it in the set of special characters.

static const char jsonIsOk[256] = {

0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,

1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1,

1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,

1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1,

1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,

1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1

};

/* Objects */

typedef struct JsonCache JsonCache;

typedef struct JsonString JsonString;

typedef struct JsonParse JsonParse;

** Magic number used for the JSON parse cache in sqlite3_get_auxdata()

#define JSON_CACHE_ID (-429938) /* Cache entry */

#define JSON_CACHE_SIZE 4 /* Max number of cache entries */

** jsonUnescapeOneChar() returns this invalid code point if it encounters

** a syntax error.

#define JSON_INVALID_CHAR 0x99999

/* A cache mapping JSON text into JSONB blobs.

** Each cache entry is a JsonParse object with the following restrictions:

** * The bReadOnly flag must be set

** * The aBlob[] array must be owned by the JsonParse object. In other

** words, nBlobAlloc must be non-zero.

** * eEdit and delta must be zero.

** * zJson must be an RCStr. In other words bJsonIsRCStr must be true.

struct JsonCache {

sqlite3 *db; /* Database connection */

int nUsed; /* Number of active entries in the cache */

JsonParse *a[JSON_CACHE_SIZE]; /* One line for each cache entry */

};

/* An instance of this object represents a JSON string

** under construction. Really, this is a generic string accumulator

** that can be and is used to create strings other than JSON.

** If the generated string is longer than will fit into the zSpace[] buffer,

** then it will be an RCStr string. This aids with caching of large

** JSON strings.

struct JsonString {

sqlite3_context *pCtx; /* Function context - put error messages here */

char *zBuf; /* Append JSON content here */

u64 nAlloc; /* Bytes of storage available in zBuf[] */

u64 nUsed; /* Bytes of zBuf[] currently used */

u8 bStatic; /* True if zBuf is static space */

u8 eErr; /* True if an error has been encountered */

char zSpace[100]; /* Initial static space */

};

/* Allowed values for JsonString.eErr */

#define JSTRING_OOM 0x01 /* Out of memory */

#define JSTRING_MALFORMED 0x02 /* Malformed JSONB */

#define JSTRING_ERR 0x04 /* Error already sent to sqlite3_result */

/* The "subtype" set for text JSON values passed through using

** sqlite3_result_subtype() and sqlite3_value_subtype().

#define JSON_SUBTYPE 74 /* Ascii for "J" */

** Bit values for the flags passed into various SQL function implementations

** via the sqlite3_user_data() value.

#define JSON_JSON 0x01 /* Result is always JSON */

#define JSON_SQL 0x02 /* Result is always SQL */

#define JSON_ABPATH 0x03 /* Allow abbreviated JSON path specs */

#define JSON_ISSET 0x04 /* json_set(), not json_insert() */

#define JSON_BLOB 0x08 /* Use the BLOB output format */

/* A parsed JSON value. Lifecycle:

** 1. JSON comes in and is parsed into a JSONB value in aBlob. The

** original text is stored in zJson. This step is skipped if the

** input is JSONB instead of text JSON.

** 2. The aBlob[] array is searched using the JSON path notation, if needed.

** 3. Zero or more changes are made to aBlob[] (via json_remove() or

** json_replace() or json_patch() or similar).

** 4. New JSON text is generated from the aBlob[] for output. This step

** is skipped if the function is one of the jsonb_* functions that

** returns JSONB instead of text JSON.

struct JsonParse {

u8 *aBlob; /* JSONB representation of JSON value */

u32 nBlob; /* Bytes of aBlob[] actually used */

u32 nBlobAlloc; /* Bytes allocated to aBlob[]. 0 if aBlob is external */

char *zJson; /* Json text used for parsing */

sqlite3 *db; /* The database connection to which this object belongs */

int nJson; /* Length of the zJson string in bytes */

u32 nJPRef; /* Number of references to this object */

u32 iErr; /* Error location in zJson[] */

u16 iDepth; /* Nesting depth */

u8 nErr; /* Number of errors seen */

u8 oom; /* Set to true if out of memory */

u8 bJsonIsRCStr; /* True if zJson is an RCStr */

u8 hasNonstd; /* True if input uses non-standard features like JSON5 */

u8 bReadOnly; /* Do not modify. */

/* Search and edit information. See jsonLookupStep() */

u8 eEdit; /* Edit operation to apply */

int delta; /* Size change due to the edit */

u32 nIns; /* Number of bytes to insert */

u32 iLabel; /* Location of label if search landed on an object value */

u8 *aIns; /* Content to be inserted */

};

/* Allowed values for JsonParse.eEdit */

#define JEDIT_DEL 1 /* Delete if exists */

#define JEDIT_REPL 2 /* Overwrite if exists */

#define JEDIT_INS 3 /* Insert if not exists */

#define JEDIT_SET 4 /* Insert or overwrite */

** Maximum nesting depth of JSON for this implementation.

** This limit is needed to avoid a stack overflow in the recursive

** descent parser. A depth of 1000 is far deeper than any sane JSON

** should go. Historical note: This limit was 2000 prior to version 3.42.0

#ifndef SQLITE_JSON_MAX_DEPTH

# define JSON_MAX_DEPTH 1000

#else

# define JSON_MAX_DEPTH SQLITE_JSON_MAX_DEPTH

#endif

** Allowed values for the flgs argument to jsonParseFuncArg();

#define JSON_EDITABLE 0x01 /* Generate a writable JsonParse object */

#define JSON_KEEPERROR 0x02 /* Return non-NULL even if there is an error */

/**************************************************************************

** Forward references

**************************************************************************/

static void jsonReturnStringAsBlob(JsonString*);

static int jsonFuncArgMightBeBinary(sqlite3_value *pJson);

static u32 jsonTranslateBlobToText(const JsonParse*,u32,JsonString*);

static void jsonReturnParse(sqlite3_context*,JsonParse*);

static JsonParse *jsonParseFuncArg(sqlite3_context*,sqlite3_value*,u32);

static void jsonParseFree(JsonParse*);

static u32 jsonbPayloadSize(const JsonParse*, u32, u32*);

static u32 jsonUnescapeOneChar(const char*, u32, u32*);

/**************************************************************************

** Utility routines for dealing with JsonCache objects

**************************************************************************/

** Free a JsonCache object.

static void jsonCacheDelete(JsonCache *p){

int i;

for(i=0; i<p->nUsed; i++){

jsonParseFree(p->a[i]);

}

sqlite3DbFree(p->db, p);

}

static void jsonCacheDeleteGeneric(void *p){

jsonCacheDelete((JsonCache*)p);

}

** Insert a new entry into the cache. If the cache is full, expel

** the least recently used entry. Return SQLITE_OK on success or a

** result code otherwise.

** Cache entries are stored in age order, oldest first.

static int jsonCacheInsert(

sqlite3_context *ctx, /* The SQL statement context holding the cache */

JsonParse *pParse /* The parse object to be added to the cache */

){

JsonCache *p;

assert( pParse->zJson!=0 );

assert( pParse->bJsonIsRCStr );

assert( pParse->delta==0 );

p = sqlite3_get_auxdata(ctx, JSON_CACHE_ID);

if( p==0 ){

sqlite3 *db = sqlite3_context_db_handle(ctx);

p = sqlite3DbMallocZero(db, sizeof(*p));

if( p==0 ) return SQLITE_NOMEM;

p->db = db;

sqlite3_set_auxdata(ctx, JSON_CACHE_ID, p, jsonCacheDeleteGeneric);

p = sqlite3_get_auxdata(ctx, JSON_CACHE_ID);

if( p==0 ) return SQLITE_NOMEM;

}

if( p->nUsed >= JSON_CACHE_SIZE ){

jsonParseFree(p->a[0]);

memmove(p->a, &p->a[1], (JSON_CACHE_SIZE-1)*sizeof(p->a[0]));

p->nUsed = JSON_CACHE_SIZE-1;

}

assert( pParse->nBlobAlloc>0 );

pParse->eEdit = 0;

pParse->nJPRef++;

pParse->bReadOnly = 1;

p->a[p->nUsed] = pParse;

p->nUsed++;

return SQLITE_OK;

}

** Search for a cached translation the json text supplied by pArg. Return

** the JsonParse object if found. Return NULL if not found.

** When a match if found, the matching entry is moved to become the

** most-recently used entry if it isn't so already.

** The JsonParse object returned still belongs to the Cache and might

** be deleted at any moment. If the caller wants the JsonParse to

** linger, it needs to increment the nPJRef reference counter.

static JsonParse *jsonCacheSearch(

sqlite3_context *ctx, /* The SQL statement context holding the cache */

sqlite3_value *pArg /* Function argument containing SQL text */

){

JsonCache *p;

int i;

const char *zJson;

int nJson;

if( sqlite3_value_type(pArg)!=SQLITE_TEXT ){

return 0;

}

zJson = (const char*)sqlite3_value_text(pArg);

if( zJson==0 ) return 0;

nJson = sqlite3_value_bytes(pArg);

p = sqlite3_get_auxdata(ctx, JSON_CACHE_ID);

if( p==0 ){

return 0;

}

for(i=0; i<p->nUsed; i++){

if( p->a[i]->zJson==zJson ) break;

}

if( i>=p->nUsed ){

for(i=0; i<p->nUsed; i++){

if( p->a[i]->nJson!=nJson ) continue;

if( memcmp(p->a[i]->zJson, zJson, nJson)==0 ) break;

}

if( i<p->nUsed ){

if( i<p->nUsed-1 ){

/* Make the matching entry the most recently used entry */

JsonParse *tmp = p->a[i];

memmove(&p->a[i], &p->a[i+1], (p->nUsed-i-1)*sizeof(tmp));

p->a[p->nUsed-1] = tmp;

i = p->nUsed - 1;

}

assert( p->a[i]->delta==0 );

return p->a[i];

}else{

return 0;

}

/**************************************************************************

** Utility routines for dealing with JsonString objects

**************************************************************************/

/* Turn uninitialized bulk memory into a valid JsonString object

** holding a zero-length string.

static void jsonStringZero(JsonString *p){

p->zBuf = p->zSpace;

p->nAlloc = sizeof(p->zSpace);

p->nUsed = 0;

p->bStatic = 1;

}

/* Initialize the JsonString object

static void jsonStringInit(JsonString *p, sqlite3_context *pCtx){

p->pCtx = pCtx;

p->eErr = 0;

jsonStringZero(p);

}

/* Free all allocated memory and reset the JsonString object back to its

** initial state.

static void jsonStringReset(JsonString *p){

if( !p->bStatic ) sqlite3RCStrUnref(p->zBuf);

jsonStringZero(p);

}

/* Report an out-of-memory (OOM) condition

static void jsonStringOom(JsonString *p){

p->eErr |= JSTRING_OOM;

if( p->pCtx ) sqlite3_result_error_nomem(p->pCtx);

jsonStringReset(p);

}

/* Enlarge pJson->zBuf so that it can hold at least N more bytes.

** Return zero on success. Return non-zero on an OOM error

static int jsonStringGrow(JsonString *p, u32 N){

u64 nTotal = N<p->nAlloc ? p->nAlloc*2 : p->nAlloc+N+10;

char *zNew;

if( p->bStatic ){

if( p->eErr ) return 1;

zNew = sqlite3RCStrNew(nTotal);

if( zNew==0 ){

jsonStringOom(p);

return SQLITE_NOMEM;

}

memcpy(zNew, p->zBuf, (size_t)p->nUsed);

p->zBuf = zNew;

p->bStatic = 0;

}else{

p->zBuf = sqlite3RCStrResize(p->zBuf, nTotal);

if( p->zBuf==0 ){

p->eErr |= JSTRING_OOM;

jsonStringZero(p);

return SQLITE_NOMEM;

}

p->nAlloc = nTotal;

return SQLITE_OK;

}

/* Append N bytes from zIn onto the end of the JsonString string.

static SQLITE_NOINLINE void jsonStringExpandAndAppend(

JsonString *p,

const char *zIn,

u32 N

){

assert( N>0 );

if( jsonStringGrow(p,N) ) return;

memcpy(p->zBuf+p->nUsed, zIn, N);

p->nUsed += N;

}

static void jsonAppendRaw(JsonString *p, const char *zIn, u32 N){

if( N==0 ) return;

if( N+p->nUsed >= p->nAlloc ){

jsonStringExpandAndAppend(p,zIn,N);

}else{

memcpy(p->zBuf+p->nUsed, zIn, N);

p->nUsed += N;

}

static void jsonAppendRawNZ(JsonString *p, const char *zIn, u32 N){

assert( N>0 );

if( N+p->nUsed >= p->nAlloc ){

jsonStringExpandAndAppend(p,zIn,N);

}else{

memcpy(p->zBuf+p->nUsed, zIn, N);

p->nUsed += N;

}

/* Append formatted text (not to exceed N bytes) to the JsonString.

static void jsonPrintf(int N, JsonString *p, const char *zFormat, ...){

va_list ap;

if( (p->nUsed + N >= p->nAlloc) && jsonStringGrow(p, N) ) return;

va_start(ap, zFormat);

sqlite3_vsnprintf(N, p->zBuf+p->nUsed, zFormat, ap);

va_end(ap);

p->nUsed += (int)strlen(p->zBuf+p->nUsed);

}

/* Append a single character

static SQLITE_NOINLINE void jsonAppendCharExpand(JsonString *p, char c){

if( jsonStringGrow(p,1) ) return;

p->zBuf[p->nUsed++] = c;

}

static void jsonAppendChar(JsonString *p, char c){

if( p->nUsed>=p->nAlloc ){

jsonAppendCharExpand(p,c);

}else{

p->zBuf[p->nUsed++] = c;

}

/* Remove a single character from the end of the string

static void jsonStringTrimOneChar(JsonString *p){

if( p->eErr==0 ){

assert( p->nUsed>0 );

p->nUsed--;

}

/* Make sure there is a zero terminator on p->zBuf[]

** Return true on success. Return false if an OOM prevents this

** from happening.

static int jsonStringTerminate(JsonString *p){

jsonAppendChar(p, 0);

jsonStringTrimOneChar(p);

return p->eErr==0;

}

/* Append a comma separator to the output buffer, if the previous

** character is not '[' or '{'.

static void jsonAppendSeparator(JsonString *p){

char c;

if( p->nUsed==0 ) return;

c = p->zBuf[p->nUsed-1];

if( c=='[' || c=='{' ) return;

jsonAppendChar(p, ',');

}

/* c is a control character. Append the canonical JSON representation

** of that control character to p.

** This routine assumes that the output buffer has already been enlarged

** sufficiently to hold the worst-case encoding plus a nul terminator.

static void jsonAppendControlChar(JsonString *p, u8 c){

static const char aSpecial[] = {

0, 0, 0, 0, 0, 0, 0, 0, 'b', 't', 'n', 0, 'f', 'r', 0, 0,

0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0

};

assert( sizeof(aSpecial)==32 );

assert( aSpecial['\b']=='b' );

assert( aSpecial['\f']=='f' );

assert( aSpecial['\n']=='n' );

assert( aSpecial['\r']=='r' );

assert( aSpecial['\t']=='t' );

assert( c>=0 && c<sizeof(aSpecial) );

assert( p->nUsed+7 <= p->nAlloc );

if( aSpecial[c] ){

p->zBuf[p->nUsed] = '\\';

p->zBuf[p->nUsed+1] = aSpecial[c];

p->nUsed += 2;

}else{

p->zBuf[p->nUsed] = '\\';

p->zBuf[p->nUsed+1] = 'u';

p->zBuf[p->nUsed+2] = '0';

p->zBuf[p->nUsed+3] = '0';

p->zBuf[p->nUsed+4] = "0123456789abcdef"[c>>4];

p->zBuf[p->nUsed+5] = "0123456789abcdef&quo 33FA t;[c&0xf];

p->nUsed += 6;

}

/* Append the N-byte string in zIn to the end of the JsonString string

** under construction. Enclose the string in double-quotes ("...") and

** escape any double-quotes or backslash characters contained within the

** string.

** This routine is a high-runner. There is a measurable performance

** increase associated with unwinding the jsonIsOk[] loop.

static void jsonAppendString(JsonString *p, const char *zIn, u32 N){

u32 k;

u8 c;

const u8 *z = (const u8*)zIn;

if( z==0 ) return;

if( (N+p->nUsed+2 >= p->nAlloc) && jsonStringGrow(p,N+2)!=0 ) return;

p->zBuf[p->nUsed++] = '"';

while( 1 /*exit-by-break*/ ){

k = 0;

/* The following while() is the 4-way unwound equivalent of

** while( k<N && jsonIsOk[z[k]] ){ k++; }

while( 1 /* Exit by break */ ){

if( k+3>=N ){

while( k<N && jsonIsOk[z[k]] ){ k++; }

break;

}

if( !jsonIsOk[z[k]] ){

break;

}

if( !jsonIsOk[z[k+1]] ){

k += 1;

break;

}

if( !jsonIsOk[z[k+2]] ){

k += 2;

break;

}

if( !jsonIsOk[z[k+3]] ){

k += 3;

break;

}else{

k += 4;

}

if( k>=N ){

if( k>0 ){

memcpy(&p->zBuf[p->nUsed], z, k);

p->nUsed += k;

}

break;

}

if( k>0 ){

memcpy(&p->zBuf[p->nUsed], z, k);

p->nUsed += k;

z += k;

N -= k;

}

c = z[0];

if( c=='"' || c=='\\' ){

if( (p->nUsed+N+3 > p->nAlloc) && jsonStringGrow(p,N+3)!=0 ) return;

p->zBuf[p->nUsed++] = '\\';

p->zBuf[p->nUsed++] = c;

}else if( c=='\'' ){

p->zBuf[p->nUsed++] = c;

}else{

if( (p->nUsed+N+7 > p->nAlloc) && jsonStringGrow(p,N+7)!=0 ) return;

jsonAppendControlChar(p, c);

}

z++;

N--;

}

p->zBuf[p->nUsed++] = '"';

assert( p->nUsed<p->nAlloc );

}

** Append an sqlite3_value (such as a function parameter) to the JSON

** string under construction in p.

static void jsonAppendSqlValue(

JsonString *p, /* Append to this JSON string */

sqlite3_value *pValue /* Value to append */

){

switch( sqlite3_value_type(pValue) ){

case SQLITE_NULL: {

jsonAppendRawNZ(p, "null", 4);

break;

}

case SQLITE_FLOAT: {

jsonPrintf(100, p, "%!0.15g", sqlite3_value_double(pValue));

break;

}

case SQLITE_INTEGER: {

const char *z = (const char*)sqlite3_value_text(pValue);

u32 n = (u32)sqlite3_value_bytes(pValue);

jsonAppendRaw(p, z, n);

break;

}

case SQLITE_TEXT: {

const char *z = (const char*)sqlite3_value_text(pValue);

u32 n = (u32)sqlite3_value_bytes(pValue);

if( sqlite3_value_subtype(pValue)==JSON_SUBTYPE ){

jsonAppendRaw(p, z, n);

}else{

jsonAppendString(p, z, n);

}

break;

}

default: {

if( jsonFuncArgMightBeBinary(pValue) ){

JsonParse px;

memset(&px, 0, sizeof(px));

px.aBlob = (u8*)sqlite3_value_blob(pValue);

px.nBlob = sqlite3_value_bytes(pValue);

jsonTranslateBlobToText(&px, 0, p);

}else if( p->eErr==0 ){

sqlite3_result_error(p->pCtx, "JSON cannot hold BLOB values", -1);

p->eErr = JSTRING_ERR;

jsonStringReset(p);

}

break;

}

/* Make the text in p (which is probably a generated JSON text string)

** the result of the SQL function.

** The JsonString is reset.

** If pParse and ctx are both non-NULL, then the SQL string in p is

** loaded into the zJson field of the pParse object as a RCStr and the

** pParse is added to the cache.

static void jsonReturnString(

JsonString *p, /* String to return */

JsonParse *pParse, /* JSONB source or NULL */

sqlite3_context *ctx /* Where to cache */

){

assert( (pParse!=0)==(ctx!=0) );

assert( ctx==0 || ctx==p->pCtx );

if( p->eErr==0 ){

int flags = SQLITE_PTR_TO_INT(sqlite3_user_data(p->pCtx));

if( flags & JSON_BLOB ){

jsonReturnStringAsBlob(p);

}else if( p->bStatic ){

sqlite3_result_text64(p->pCtx, p->zBuf, p->nUsed,

SQLITE_TRANSIENT, SQLITE_UTF8);

}else if( jsonStringTerminate(p) ){

if( pParse && pParse->bJsonIsRCStr==0 && pParse->nBlobAlloc>0 ){

int rc;

pParse->zJson = sqlite3RCStrRef(p->zBuf);

pParse->nJson = p->nUsed;

pParse->bJsonIsRCStr = 1;

rc = jsonCacheInsert(ctx, pParse);

if( rc==SQLITE_NOMEM ){

sqlite3_result_error_nomem(ctx);

jsonStringReset(p);

return;

}

sqlite3_result_text64(p->pCtx, sqlite3RCStrRef(p->zBuf), p->nUsed,

sqlite3RCStrUnref,

SQLITE_UTF8);

}else{

sqlite3_result_error_nomem(p->pCtx);

}

}else if( p->eErr & JSTRING_OOM ){

sqlite3_result_error_nomem(p->pCtx);

}else if( p->eErr & JSTRING_MALFORMED ){

sqlite3_result_error(p->pCtx, "malformed JSON", -1);

}

jsonStringReset(p);

}

/**************************************************************************

** Utility routines for dealing with JsonParse objects

**************************************************************************/

** Reclaim all memory allocated by a JsonParse object. But do not

** delete the JsonParse object itself.

static void jsonParseReset(JsonParse *pParse){

assert( pParse->nJPRef<=1 );

if( pParse->bJsonIsRCStr ){

sqlite3RCStrUnref(pParse->zJson);

pParse->zJson = 0;

pParse->nJson = 0;

pParse->bJsonIsRCStr = 0;

}

if( pParse->nBlobAlloc ){

sqlite3DbFree(pParse->db, pParse->aBlob);

pParse->aBlob = 0;

pParse->nBlob = 0;

pParse->nBlobAlloc = 0;

}

** Decrement the reference count on the JsonParse object. When the

** count reaches zero, free the object.

static void jsonParseFree(JsonParse *pParse){

if( pParse ){

if( pParse->nJPRef>1 ){

pParse->nJPRef--;

}else{

jsonParseReset(pParse);

sqlite3DbFree(pParse->db, pParse);

}

/**************************************************************************

** Utility routines for the JSON text parser

**************************************************************************/

** Translate a single byte of Hex into an integer.

** This routine only gives a correct answer if h really is a valid hexadecimal

** character: 0..9a..fA..F. But unlike sqlite3HexToInt(), it does not

** assert() if the digit is not hex.

static u8 jsonHexToInt(int h){

#ifdef SQLITE_ASCII

h += 9*(1&(h>>6));

#endif

#ifdef SQLITE_EBCDIC

h += 9*(1&~(h>>4));

#endif

return (u8)(h & 0xf);

}

** Convert a 4-byte hex string into an integer

static u32 jsonHexToInt4(const char *z){

u32 v;

v = (jsonHexToInt(z[0])<<12)

+ (jsonHexToInt(z[1])<<8)

+ (jsonHexToInt(z[2])<<4)

+ jsonHexToInt(z[3]);

return v;

}

** Return true if z[] begins with 2 (or more) hexadecimal digits

static int jsonIs2Hex(const char *z){

return sqlite3Isxdigit(z[0]) && sqlite3Isxdigit(z[1]);

}

** Return true if z[] begins with 4 (or more) hexadecimal digits

static int jsonIs4Hex(const char *z){

return jsonIs2Hex(z) && jsonIs2Hex(&z[2]);

}

** Return the number of bytes of JSON5 whitespace at the beginning of

** the input string z[].

** JSON5 whitespace consists of any of the following characters:

** Unicode UTF-8 Name

** U+0009 09 horizontal tab

** U+000a 0a line feed

** U+000b 0b vertical tab

** U+000c 0c form feed

** U+000d 0d carriage return

** U+0020 20 space

** U+00a0 c2 a0 non-breaking space

** U+1680 e1 9a 80 ogham space mark

** U+2000 e2 80 80 en quad

** U+2001 e2 80 81 em quad

** U+2002 e2 80 82 en space

** U+2003 e2 80 83 em space

** U+2004 e2 80 84 three-per-em space

** U+2005 e2 80 85 four-per-em space

** U+2006 e2 80 86 six-per-em space

** U+2007 e2 80 87 figure space

** U+2008 e2 80 88 punctuation space

** U+2009 e2 80 89 thin space

** U+200a e2 80 8a hair space

** U+2028 e2 80 a8 line separator

** U+2029 e2 80 a9 paragraph separator

** U+202f e2 80 af narrow no-break space (NNBSP)

** U+205f e2 81 9f medium mathematical space (MMSP)

** U+3000 e3 80 80 ideographical space

** U+FEFF ef bb bf byte order mark

** In addition, comments between '/', '*' and '*', '/' and

** from '/', '/' to end-of-line are also considered to be whitespace.

static int json5Whitespace(const char *zIn){

int n = 0;

const u8 *z = (u8*)zIn;

while( 1 /*exit by "goto whitespace_done"*/ ){

switch( z[n] ){

case 0x09:

case 0x0a:

case 0x0b:

case 0x0c:

case 0x0d:

case 0x20: {

n++;

break;

}

case '/': {

if( z[n+1]=='*' && z[n+2]!=0 ){

int j;

for(j=n+3; z[j]!='/' || z[j-1]!='*'; j++){

if( z[j]==0 ) goto whitespace_done;

}

n = j+1;

break;

}else if( z[n+1]=='/' ){

45B9 int j;

char c;

for(j=n+2; (c = z[j])!=0; j++){

if( c=='\n' || c=='\r' ) break;

if( 0xe2==(u8)c && 0x80==(u8)z[j+1]

&& (0xa8==(u8)z[j+2] || 0xa9==(u8)z[j+2])

){

j += 2;

break;

}

n = j;

if( z[n] ) n++;

break;

}

goto whitespace_done;

}

case 0xc2: {

if( z[n+1]==0xa0 ){

n += 2;

break;

}

goto whitespace_done;

}

case 0xe1: {

View remainder of file in raw view

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

FilesExpand file tree

json.c

Latest commit

History

json.c

File metadata and controls