wormtable/_wormtablemodule.c at master · wormtable/wormtable

History

4454 lines (4185 loc) · 127 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

** This file is part of wormtable.

** Wormtable is free software: you can redistribute it and/or modify

** it under the terms of the GNU Lesser General Public License as published by

** the Free Software Foundation, either version 3 of the License, or

** (at your option) any later version.

** Wormtable 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 Lesser General Public License for more details.

** You should have received a copy of the GNU Lesser General Public License

** along with wormtable. If not, see <http://www.gnu.org/licenses/>.

#include <Python.h>

#include <structmember.h>

#include <db.h>

#include "halffloat.h"

#ifdef _WIN32

#ifdef __MINGW32__

#define fseeko fseeko64

#define ftello ftello64

#else

#define fseeko _fseeki64

#define ftello _ftelli64

#endif

/* Dealing with Windows and MSVC... */

/* the declspec allows compiling of a dll in Visual Studio

and, if necessary, simply renaming the file to .pyd for it

to work as a setuptools-compiled module. We first define it

as an empty preprocessor directive so compilation doesn't break

on non-MSVC build systems. */

#define DLLEXPORT

#ifdef _MSC_VER

#define DLLEXPORT __declspec(dllexport)

#endif

#if PY_MAJOR_VERSION >= 3

#define IS_PY3K

#endif

#define WT_READ 0

#define WT_WRITE 1

#define WT_UINT 0

#define WT_INT 1

#define WT_FLOAT 2

#define WT_CHAR 3

#define WT_VAR_1 0

#define WT_VAR_2 (-1)

#define WT_VAR_1_MAX_ELEMENTS 254

#define WT_VAR_2_MAX_ELEMENTS 65534

#define MAX_ROW_SIZE 65536

#define WT_MISSING_VALUE 1

#define OFFSET_LEN_RECORD_SIZE 10

/* This is the default defined by the linux fopen man pages. */

#define WT_DB_FILE_PERMS 0666

#define MODULE_DOC \

"Low level Berkeley DB interface for wormtable"

static PyObject *WormtableError;

typedef struct Column_t {

PyObject_HEAD

PyObject *name;

PyObject *description;

PyObject *min_element;

PyObject *max_element;

int position;

int element_type;

int element_size;

int num_elements;

int fixed_region_offset;

void **input_elements; /* pointer to each elements in input format */

void *element_buffer; /* parsed input elements in native CPU format */

int num_buffered_elements;

int (*string_to_native)(struct Column_t*, char *);

int (*python_to_native)(struct Column_t*, PyObject *);

int (*verify_elements)(struct Column_t*);

int (*truncate_elements)(struct Column_t*, double);

int (*pack_elements)(struct Column_t*, void *);

int (*unpack_elements)(struct Column_t*, void *);

PyObject *(*native_to_python)(struct Column_t *, int);

} Column;

* We're using a mixture of fixed and non-fixed size integer types because

* of the conflicting needs of being exact about the sizes and interfacing

* with Python. Ideally, all of the types would be fixed size for simplicity

typedef struct {

PyObject_HEAD

DB *db;

PyObject *db_filename;

FILE *data_file;

PyObject *data_filename;

Column **columns;

unsigned long long cache_size;

unsigned int fixed_region_size;

unsigned int num_columns;

void *row_buffer;

uint32_t row_buffer_size; /* max size */

uint32_t current_row_size; /* current size */

unsigned long long num_rows;

/* row stats */

unsigned long long total_row_size;

unsigned int min_row_size;

unsigned int max_row_size;

} Table;

typedef struct {

PyObject_HEAD

Table *table;

DB *db;

PyObject *db_filename;

unsigned long long cache_size;

uint32_t *columns;

uint32_t num_columns;

void *key_buffer;

uint32_t key_buffer_size;

double *bin_widths;

} Index;

typedef struct {

PyObject_HEAD

Index *index;

DBC *cursor;

int completed;

uint32_t *read_columns;

uint32_t num_read_columns;

void *min_key;

uint32_t min_key_size;

void *max_key;

uint32_t max_key_size;

} IndexRowIterator;

typedef struct {

PyObject_HEAD

Table *table;

DBC *cursor;

int completed;

uint32_t *read_columns;

uint32_t num_read_columns;

void *min_key;

uint32_t min_key_size;

void *max_key;

uint32_t max_key_size;

} TableRowIterator;

typedef struct {

PyObject_HEAD

Index *index;

DBC *cursor;

} IndexKeyIterator;

static void

handle_bdb_error(int err)

{

PyErr_SetString(WormtableError, db_strerror(err));

}

static void

handle_io_error(void)

{

PyErr_SetFromErrno(WormtableError);

}

#ifndef WORDS_BIGENDIAN

* Copies n bytes of source into destination, swapping the order of the

* bytes.

static void

byteswap_copy(void* dest, void *source, size_t n)

{

size_t j = 0;

unsigned char *dest_c = (unsigned char *) dest;

unsigned char *source_c = (unsigned char *) source;

for (j = 0; j < n; j++) {

dest_c[j] = source_c[n - j - 1];

}

#endif

* Floating point packing and unpacking. Floating point values are stored as a

* slight perturbation to the IEEE standards so that they sort correctly. This

* is done by a simple rule: if the number is positive, flip the sign bit; if

* it is negative, flip all bits. This procedure is reversed when we unpack

* to recover the original values exactly.

static void

pack_half(double value, void *dest)

{

union { uint16_t value; int16_t bits; } conv;

int16_t bits;

conv.value = npy_double_to_half(value);

bits = conv.bits;

bits ^= (bits < 0) ? 0xffff: 0x8000;

#ifdef WORDS_BIGENDIAN

memcpy(dest, &bits, 2);

#else

byteswap_copy(dest, &bits, 2);

#endif

}

static double

unpack_half(void *src)

{

int16_t bits;

double v;

#ifdef WORDS_BIGENDIAN

memcpy(&bits, src, 2);

#else

byteswap_copy(&bits, src, 2);

#endif

bits ^= (bits < 0) ? 0x8000: 0xffff;

v = npy_half_to_double(bits);

return v;

}

static void

pack_float(double value, void *dest)

{

int32_t bits;

union { float value; int32_t bits; } conv;

conv.value = (float) value;

bits = conv.bits;

bits ^= (bits < 0) ? 0xffffffffL: 0x80000000L;

#ifdef WORDS_BIGENDIAN

memcpy(dest, &bits, sizeof(float));

#else

byteswap_copy(dest, &bits, sizeof(float));

#endif

}

static double

unpack_float(void *src)

{

union { float value; int32_t bits; } conv;

int32_t bits;

#ifdef WORDS_BIGENDIAN

memcpy(&bits, src, sizeof(float));

#else

byteswap_copy(&bits, src, sizeof(float));

#endif

bits ^= (bits < 0) ? 0x80000000L: 0xffffffffL;

conv.bits = bits;

return (double) conv.value;

}

static void

pack_double(double value, void *dest)

{

union { double value; int64_t bits; } conv;

int64_t bits;

conv.value = value;

bits = conv.bits;

bits ^= (bits < 0) ? 0xffffffffffffffffLL: 0x8000000000000000LL;

#ifdef WORDS_BIGENDIAN

memcpy(dest, &bits, sizeof(double));

#else

byteswap_copy(dest, &bits, sizeof(double));

#endif

}

static double

unpack_double(void *src)

{

union { double value; int64_t bits; } conv;

int64_t bits;

#ifdef WORDS_BIGENDIAN

memcpy(&bits, src, sizeof(double));

#else

byteswap_copy(&bits, src, sizeof(double));

#endif

bits ^= (bits < 0) ? 0x8000000000000000LL: 0xffffffffffffffffLL;

conv.bits = bits;

return conv.value;

}

/* Integer packing and unpacking.

* TODO document the format.

static void

pack_uint(uint64_t value, void *dest, uint8_t size)

{

char *src;

uint64_t u = value;

/* increment before storing */

u += 1;

src = (char *) &u;

#ifdef WORDS_BIGENDIAN

memcpy(dest, src + (8 - size), size);

#else

byteswap_copy(dest, src, size);

#endif

}

static uint64_t

unpack_uint(void *src, uint8_t size)

{

uint64_t dest = 0;

char *v = (char *) &dest;

#ifdef WORDS_BIGENDIAN

memcpy(v + 8 - size, src, size);

#else

byteswap_copy(v, src, size);

#endif

/* decrement and return */

dest -= 1;

return dest;

}

static void

pack_int(int64_t value, void *dest, uint8_t size)

{

char *src;

int64_t u = value;

const int64_t m = 1LL << (size * 8 - 1);

/* flip the sign bit */

u ^= m;

src = (char *) &u;

#ifdef WORDS_BIGENDIAN

memcpy(dest, src + (8 - size), size);

#else

byteswap_copy(dest, src, size);

#endif

}

static int64_t

unpack_int(void *src, uint8_t size)

{

int64_t dest = 0;

char *v = (char *) &dest;

const int64_t m = 1LL << (size * 8 - 1);

#ifdef WORDS_BIGENDIAN

memcpy(v + 8 - size, src, size);

#else

byteswap_copy(v, src, size);

#endif

/* flip the sign bit */

dest ^= m;

/* sign extend and return */

dest = (dest ^ m) - m;

return dest;

}

* Returns the missing value for a k byte integer.

static int64_t

missing_int(uint32_t k)

{

int64_t v = (-1) * (1ll << (8 * k - 1));

return v;

}

* Returns the minimum value for a k byte integer.

static int64_t

min_int(uint32_t k)

{

int64_t v = (-1) * (1ll << (8 * k - 1)) + 1;

return v;

}

* Returns the maximum value for a k byte integer.

static int64_t

max_int(uint32_t k)

{

int64_t v = (1ll << (8 * k - 1)) - 1;

return v;

}

* Returns the missing value for a k byte unsigned integer.

static uint64_t

missing_uint(uint32_t k)

{

return (uint64_t) -1ll;

}

* Returns the maximum value for a k byte unsigned integer.

static uint64_t

max_uint(uint32_t k)

{

uint64_t v = (uint64_t) -1ll;

if (k < 8) {

v = (1ll << (8 * k)) - 1;

}

return v - 1;

}

* Returns the minumum value for a k byte unsigned integer.

static uint64_t

min_uint(uint32_t k)

{

uint64_t v = 0ll;

return v;

}

* Returns the missing value for a k byte float. This

* is returned as an unsigned integer since we cannot

* compare NaN values as doubles.

static uint64_t

missing_float(uint32_t k)

{

uint64_t zero = 0uLL;

union { double value; uint64_t bits; } conv;

conv.bits = 0;

switch (k) {

case 2:

conv.value = unpack_half(&zero);

break;

case 4:

conv.value = unpack_float(&zero);

break;

case 8:

conv.value = unpack_double(&zero);

break;

}

return conv.bits;

}

/*==========================================================

* Column object

*==========================================================

* Returns true if this Column is variable length.

static int

Column_is_variable(Column *self)

{

return self->num_elements == WT_VAR_1 || self->num_elements == WT_VAR_2;

}

* Returns the maximum number of elements that can be stored in this Column.

static uint32_t

Column_get_max_num_elements(Column *self)

{

uint32_t ret = self->num_elements;

if (Column_is_variable(self)) {

if (self->num_elements == WT_VAR_1) {

ret = WT_VAR_1_MAX_ELEMENTS;

} else {

ret = WT_VAR_2_MAX_ELEMENTS;

}

return ret;

}

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

* Native values to Python conversion.

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

static PyObject *

Column_native_to_python_uint(Column *self, int index)

{

PyObject *ret = NULL;

uint64_t *elements = (uint64_t *) self->element_buffer;

uint64_t missing_value = missing_uint(self->element_size);

if (elements[index] == missing_value) {

Py_INCREF(Py_None);

ret = Py_None;

} else {

ret = PyLong_FromUnsignedLongLong((unsigned long long) elements[index]);

if (ret == NULL) {

PyErr_NoMemory();

}

return ret;

}

static PyObject *

Column_native_to_python_int(Column *self, int index)

{

PyObject *ret = NULL;

int64_t *elements = (int64_t *) self->element_buffer;

int64_t missing_value = missing_int(self->element_size);

if (elements[index] == missing_value) {

Py_INCREF(Py_None);

ret = Py_None;

} else {

ret = PyLong_FromLongLong((long long) elements[index]);

if (ret == NULL) {

PyErr_NoMemory();

}

return ret;

}

static PyObject *

Column_native_to_python_float(Column *self, int index)

{

PyObject *ret = NULL;

union { double value; uint64_t bits; } conv;

double *elements = (double *) self->element_buffer;

uint64_t missing_bits = missing_float(self->element_size);

conv.value = elements[index];

if (conv.bits == missing_bits) {

Py_INCREF(Py_None);

ret = Py_None;

} else {

ret = PyFloat_FromDouble(elements[index]);

if (ret == NULL) {

PyErr_NoMemory();

}

return ret;

}

static PyObject *

Column_native_to_python_char(Column *self, int index)

{

PyObject *ret = NULL;

int j = self->num_buffered_elements;

char *str = (char *) self->element_buffer;

ret = PyBytes_FromStringAndSize(str, j);

if (ret == NULL) {

PyErr_NoMemory();

}

return ret;

}

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

* Unpacking from a row to the element buffer.

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

* Unpack values starting at the specified pointer, and return the

* number of missing_values encountered.

static int

Column_unpack_elements_uint(Column *self, void *source)

{

int j;

int ret = 0;

char *v = (char *) source;

uint64_t *elements = (uint64_t *) self->element_buffer;

uint64_t missing_value = missing_uint(self->element_size);

int size = self->element_size;

for (j = 0; j < self->num_buffered_elements; j++) {

elements[j] = unpack_uint(v + j * size, size);

if (elements[j] == missing_value) {

ret += 1;

}

return ret;

}

static int

Column_unpack_elements_int(Column *self, void *source)

{

int j;

int ret = 0;

char *v = (char *) source;

int64_t *elements = (int64_t *) self->element_buffer;

int64_t missing_value = missing_int(self->element_size);

int size = self->element_size;

for (j = 0; j < self->num_buffered_elements; j++) {

elements[j] = unpack_int(v + j * size, size);

if (elements[j] == missing_value) {

ret += 1;

}

return ret;

}

static int

Column_unpack_elements_float_2(Column *self, void *source)

{

int j;

int ret = 0;

char *v = (char *) source;

double *elements = (double *) self->element_buffer;

union { double value; uint64_t bits; } conv;

uint64_t missing_bits = missing_float(self->element_size);

for (j = 0; j < self->num_buffered_elements; j++) {

elements[j] = unpack_half(v);

v += self->element_size;

conv.value = elements[j];

if (conv.bits == missing_bits) {

ret += 1;

}

return ret;

}

static int

Column_unpack_elements_float_4(Column *self, void *source)

{

int j;

int ret = 0;

char *v = (char *) source;

double *elements = (double *) self->element_buffer;

union { double value; uint64_t bits; } conv;

uint64_t missing_bits = missing_float(self->element_size);

for (j = 0; j < self->num_buffered_elements; j++) {

elements[j] = unpack_float(v);

v += self->element_size;

conv.value = elements[j];

if (conv.bits == missing_bits) {

ret += 1;

}

return ret;

}

static int

Column_unpack_elements_float_8(Column *self, void *source)

{

int j;

int ret = 0;

char *v = (char *) source;

double *elements = (double *) self->element_buffer;

union { double value; uint64_t bits; } conv;

uint64_t missing_bits = missing_float(self->element_size);

for (j = 0; j < self->num_buffered_elements; j++) {

elements[j] = unpack_double(v);

v += self->element_size;

conv.value = elements[j];

if (conv.bits == missing_bits) {

ret += 1;

}

return ret;

}

static int

Column_unpack_elements_char(Column *self, void *source)

{

int ret = 0;

int j;

char *v = (char *) source;

memcpy(self->element_buffer, source, self->num_buffered_elements);

/* scan for zero bytes to get missing values */

for (j = 0; j < self->num_buffered_elements; j++) {

if (v[j] == 0) {

ret += 1;

}

return ret;

}

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

* Packing native values from the element_buffer to a row.

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

static int

Column_pack_elements_uint(Column *self, void *dest)

{

int j;

int ret = -1;

char *v = (char *) dest;

uint64_t *elements = (uint64_t *) self->element_buffer;

for (j = 0; j < self->num_buffered_elements; j++) {

pack_uint(elements[j], v, self->element_size);

v += self->element_size;

}

ret = 0;

return ret;

}

static int

Column_pack_elements_int(Column *self, void *dest)

{

int j;

int ret = -1;

char *v = (char *) dest;

int64_t *elements = (int64_t *) self->element_buffer;

for (j = 0; j < self->num_buffered_elements; j++) {

pack_int(elements[j], v, self->element_size);

v += self->element_size;

}

ret = 0;

return ret;

}

static int

Column_pack_elements_float_2(Column *self, void *dest)

{

int j;

int ret = -1;

char *v = (char *) dest;

double *elements = (double *) self->element_buffer;

for (j = 0; j < self->num_buffered_elements; j++) {

pack_half(elements[j], v);

v += self->element_size;

}

ret = 0;

return ret;

}

static int

Column_pack_elements_float_4(Column *self, void *dest)

{

int j;

int ret = -1;

char *v = (char *) dest;

double *elements = (double *) self->element_buffer;

for (j = 0; j < self->num_buffered_elements; j++) {

pack_float(elements[j], v);

v += self->element_size;

}

ret = 0;

return ret;

}

static int

Column_pack_elements_float_8(Column *self, void *dest)

{

int j;

int ret = -1;

char *v = (char *) dest;

double *elements = (double *) self->element_buffer;

for (j = 0; j < self->num_buffered_elements; j++) {

pack_double(elements[j], v);

v += self->element_size;

}

ret = 0;

return ret;

}

static int

Column_pack_elements_char(Column *self, void *dest)

{

int ret = -1;

memcpy(dest, self->element_buffer, self->num_buffered_elements);

ret = 0;

return ret;

}

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

* Verify elements in the buffer.

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

/* This doesn't really have any useful function any more, as we were forced

* to put the range checking functionality into the native_to_python_int

* function. This is was because there was no way to exclude the user

* from using the missing_value as an input parameter, and this would

* have led to very annoying bugs. This doesn't do any harm here though,

* so let's leave it in for now.

static int

Column_verify_elements_uint(Column *self)

{

int j;

int ret = -1;

uint64_t *elements = (uint64_t *) self->element_buffer;

uint64_t min_value = min_uint(self->element_size);

uint64_t max_value = max_uint(self->element_size);

uint64_t missing_value = missing_uint(self->element_size);

for (j = 0; j < self->num_buffered_elements; j++) {

if (elements[j] != missing_value &&

(elements[j] < min_value || elements[j] > max_value)) {

PyErr_Format(PyExc_OverflowError,

"Values for column '%s' must be between %lld and %lld",

PyBytes_AsString(self->name), (long long) min_value,

(long long) max_value);

goto out;

}

ret = 0;

out:

return ret;

}

static int

Column_verify_elements_int(Column *self)

{

int j;

int ret = -1;

int64_t *elements = (int64_t *) self->element_buffer;

int64_t min_value = min_int(self->element_size);

int64_t max_value = max_int(self->element_size);

int64_t missing_value = missing_int(self->element_size);

for (j = 0; j < self->num_buffered_elements; j++) {

if (elements[j] != missing_value &&

(elements[j] < min_value || elements[j] > max_value)) {

PyErr_Format(PyExc_OverflowError,

"Values for column '%s' must be between %lld and %lld",

PyBytes_AsString(self->name), (long long) min_value,

(long long) max_value);

goto out;

}

ret = 0;

out:

return ret;

}

static int

Column_verify_elements_float(Column *self)

{

return 0;

}

static int

Column_verify_elements_char(Column *self)

{

return 0;

}

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

* Truncate elements in the buffer.

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

static int

Column_truncate_elements_uint(Column *self, double bin_width)

{

int ret = 0;

unsigned int j;

uint64_t w = (uint64_t) bin_width;

uint64_t *elements = (uint64_t *) self->element_buffer;

uint64_t missing_value = missing_uint(self->element_size);

uint64_t u;

if (bin_width <= 0.0) {

PyErr_Format(PyExc_SystemError, "bin_width for column '%s' must > 0",

PyBytes_AsString(self->name));

goto out;

}

for (j = 0; j < self->num_buffered_elements; j++) {

u = elements[j];

if (u != missing_value) {

elements[j] = u - (u % w);

}

ret = 0;

out:

return ret;

}

static int

Column_truncate_elements_int(Column *self, double bin_width)

{

int ret = 0;

unsigned int j;

int64_t w = (int64_t) bin_width;

int64_t *elements = (int64_t *) self->element_buffer;

int64_t missing_value = missing_int(self->element_size);

int64_t u;

if (bin_width <= 0.0) {

PyErr_Format(PyExc_SystemError, "bin_width for column '%s' must > 0",

PyBytes_AsString(self->name));

goto out;

}

for (j = 0; j < self->num_buffered_elements; j++) {

u = elements[j];

if (u != missing_value) {

elements[j] = u - (u % w);

}

ret = 0;

out:

return ret;

}

static int

Column_truncate_elements_float(Column *self, double bin_width)

{

int ret = -1;

unsigned int j;

union { double value; uint64_t bits; } conv;

uint64_t missing_bits = missing_float(self->element_size);

double *elements = (double *) self->element_buffer;

double u;

if (bin_width <= 0.0) {

PyErr_Format(PyExc_SystemError, "bin_width for column '%s' must > 0",

PyBytes_AsString(self->name));

goto out;

}

for (j = 0; j < self->num_buffered_elements; j++) {

u = elements[j];

conv.value = u;

if (conv.bits != missing_bits) {

elements[j] = u - fmod(u, bin_width);

}

ret = 0;

out:

return ret;

}

static int

Column_truncate_elements_char(Column *self, double bin_width)

{

return 0;

}

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

* Python input element parsing.

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

* Takes a Python sequence and places pointers to the Python

* elements into the input_elements list. Checks for various

* errors in the format of this sequence.

static int

Column_parse_python_sequence(Column *self, PyObject *elements)

{

int ret = -1;

int j, num_elements;

PyObject *seq = NULL;

PyObject *v;

self->num_buffered_elements = 0;

if (self->num_elements == 1) {

self->input_elements[0] = elements;

num_elements = 1;

} else {

seq = PySequence_Fast(elements, "Sequence required");

if (seq == NULL) {

goto out;

}

num_elements = PySequence_Fast_GET_SIZE(seq);

if (Column_is_variable(self)) {

if (num_elements > Column_get_max_num_elements(self)) {

PyErr_Format(PyExc_ValueError,

"too many elements for column '%s'",

PyBytes_AsString(self->name));

goto out;

}

} else {

if (num_elements != self->num_elements) {

PyErr_Format(PyExc_ValueError,

"incorrect number of elements for column '%s'",

PyBytes_AsString(self->name));

goto out;

}

for (j = 0; j < num_elements; j++) {

v = PySequence_Fast_GET_ITEM(seq, j);

self->input_elements[j] = v;

}

self->num_buffered_elements = num_elements;

View remainder of file in raw view

Navigation Menu

Uh oh!

Uh oh!

FilesExpand file tree

_wormtablemodule.c

Latest commit

History

_wormtablemodule.c

File metadata and controls