cpython/Python/fileutils.c at 3.3 · python/cpython · GitHub

Repository navigation

History

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

#include "Python.h"

#include "osdefs.h"

#include <locale.h>

#ifdef MS_WINDOWS

# include <windows.h>

#endif

#ifdef HAVE_LANGINFO_H

#include <langinfo.h>

#endif

#ifdef __APPLE__

extern wchar_t* _Py_DecodeUTF8_surrogateescape(const char *s, Py_ssize_t size);

#endif

PyObject *

_Py_device_encoding(int fd)

{

#if defined(MS_WINDOWS) || defined(MS_WIN64)

UINT cp;

#endif

if (!_PyVerify_fd(fd) || !isatty(fd)) {

Py_RETURN_NONE;

}

#if defined(MS_WINDOWS) || defined(MS_WIN64)

if (fd == 0)

cp = GetConsoleCP();

else if (fd == 1 || fd == 2)

cp = GetConsoleOutputCP();

else

cp = 0;

/* GetConsoleCP() and GetConsoleOutputCP() return 0 if the application

has no console */

if (cp != 0)

return PyUnicode_FromFormat("cp%u", (unsigned int)cp);

#elif defined(CODESET)

{

char *codeset = nl_langinfo(CODESET);

if (codeset != NULL && codeset[0] != 0)

return PyUnicode_FromString(codeset);

}

#endif

Py_RETURN_NONE;

}

#if !defined(__APPLE__) && !defined(MS_WINDOWS)

extern int _Py_normalize_encoding(const char *, char *, size_t);

/* Workaround FreeBSD and OpenIndiana locale encoding issue with the C locale.

On these operating systems, nl_langinfo(CODESET) announces an alias of the

ASCII encoding, whereas mbstowcs() and wcstombs() functions use the

ISO-8859-1 encoding. The problem is that os.fsencode() and os.fsdecode() use

locale.getpreferredencoding() codec. For example, if command line arguments

are decoded by mbstowcs() and encoded back by os.fsencode(), we get a

UnicodeEncodeError instead of retrieving the original byte string.

The workaround is enabled if setlocale(LC_CTYPE, NULL) returns "C",

nl_langinfo(CODESET) announces "ascii" (or an alias to ASCII), and at least

one byte in range 0x80-0xff can be decoded from the locale encoding. The

workaround is also enabled on error, for example if getting the locale

failed.

Values of locale_is_ascii:

1: the workaround is used: _Py_wchar2char() uses

encode_ascii_surrogateescape() and _Py_char2wchar() uses

decode_ascii_surrogateescape()

0: the workaround is not used: _Py_wchar2char() uses wcstombs() and

_Py_char2wchar() uses mbstowcs()

-1: unknown, need to call check_force_ascii() to get the value

static int force_ascii = -1;

static int

check_force_ascii(void)

{

char *loc;

#if defined(HAVE_LANGINFO_H) && defined(CODESET)

char *codeset, **alias;

char encoding[100];

int is_ascii;

unsigned int i;

char* ascii_aliases[] = {

"ascii",

"646",

"ansi-x3.4-1968",

"ansi-x3-4-1968",

"ansi-x3.4-1986",

"cp367",

"csascii",

"ibm367",

"iso646-us",

"iso-646.irv-1991",

"iso-ir-6",

"us",

"us-ascii",

NULL

};

#endif

loc = setlocale(LC_CTYPE, NULL);

if (loc == NULL)

goto error;

if (strcmp(loc, "C") != 0) {

/* the LC_CTYPE locale is different than C */

return 0;

}

#if defined(HAVE_LANGINFO_H) && defined(CODESET)

codeset = nl_langinfo(CODESET);

if (!codeset || codeset[0] == '\0') {

/* CODESET is not set or empty */

goto error;

}

if (!_Py_normalize_encoding(codeset, encoding, sizeof(encoding)))

goto error;

is_ascii = 0;

for (alias=ascii_aliases; *alias != NULL; alias++) {

if (strcmp(encoding, *alias) == 0) {

is_ascii = 1;

break;

}

if (!is_ascii) {

/* nl_langinfo(CODESET) is not "ascii" or an alias of ASCII */

return 0;

}

for (i=0x80; i<0xff; i++) {

unsigned char ch;

wchar_t wch;

size_t res;

ch = (unsigned char)i;

res = mbstowcs(&wch, (char*)&ch, 1);

if (res != (size_t)-1) {

/* decoding a non-ASCII character from the locale encoding succeed:

the locale encoding is not ASCII, force ASCII */

return 1;

}

/* None of the bytes in the range 0x80-0xff can be decoded from the locale

encoding: the locale encoding is really ASCII */

return 0;

#else

/* nl_langinfo(CODESET) is not available: always force ASCII */

return 1;

#endif

error:

/* if an error occured, force the ASCII encoding */

return 1;

}

static char*

encode_ascii_surrogateescape(const wchar_t *text, size_t *error_pos)

{

char *result = NULL, *out;

size_t len, i;

wchar_t ch;

if (error_pos != NULL)

*error_pos = (size_t)-1;

len = wcslen(text);

result = PyMem_Malloc(len + 1); /* +1 for NUL byte */

if (result == NULL)

return NULL;

out = result;

for (i=0; i<len; i++) {

ch = text[i];

if (ch <= 0x7f) {

/* ASCII character */

*out++ = (char)ch;

}

else if (0xdc80 <= ch && ch <= 0xdcff) {

/* UTF-8b surrogate */

*out++ = (char)(ch - 0xdc00);

}

else {

if (error_pos != NULL)

*error_pos = i;

PyMem_Free(result);

return NULL;

}

*out = '\0';

return result;

}

#endif /* !defined(__APPLE__) && !defined(MS_WINDOWS) */

#if !defined(__APPLE__) && (!defined(MS_WINDOWS) || !defined(HAVE_MBRTOWC))

static wchar_t*

decode_ascii_surrogateescape(const char *arg, size_t *size)

{

wchar_t *res;

unsigned char *in;

wchar_t *out;

size_t argsize = strlen(arg) + 1;

if (argsize > PY_SSIZE_T_MAX/sizeof(wchar_t))

return NULL;

res = PyMem_Malloc(argsize*sizeof(wchar_t));

if (!res)

return NULL;

in = (unsigned char*)arg;

out = res;

while(*in)

if(*in < 128)

*out++ = *in++;

else

*out++ = 0xdc00 + *in++;

*out = 0;

if (size != NULL)

*size = out - res;

return res;

}

#endif

/* Decode a byte string from the locale encoding with the

surrogateescape error handler (undecodable bytes are decoded as characters

in range U+DC80..U+DCFF). If a byte sequence can be decoded as a surrogate

character, escape the bytes using the surrogateescape error handler instead

of decoding them.

Use _Py_wchar2char() to encode the character string back to a byte string.

Return a pointer to a newly allocated wide character string (use

PyMem_Free() to free the memory) and write the number of written wide

characters excluding the null character into *size if size is not NULL, or

NULL on error (decoding or memory allocation error). If size is not NULL,

*size is set to (size_t)-1 on memory error and (size_t)-2 on decoding

error.

Conversion errors should never happen, unless there is a bug in the C

library. */

wchar_t*

_Py_char2wchar(const char* arg, size_t *size)

{

#ifdef __APPLE__

wchar_t *wstr;

wstr = _Py_DecodeUTF8_surrogateescape(arg, strlen(arg));

if (size != NULL) {

if (wstr != NULL)

*size = wcslen(wstr);

else

*size = (size_t)-1;

}

return wstr;

#else

wchar_t *res;

size_t argsize;

size_t count;

unsigned char *in;

wchar_t *out;

#ifdef HAVE_MBRTOWC

mbstate_t mbs;

#endif

#ifndef MS_WINDOWS

if (force_ascii == -1)

force_ascii = check_force_ascii();

if (force_ascii) {

/* force ASCII encoding to workaround mbstowcs() issue */

res = decode_ascii_surrogateescape(arg, size);

if (res == NULL)

goto oom;

return res;

}

#endif

#ifdef HAVE_BROKEN_MBSTOWCS

/* Some platforms have a broken implementation of

* mbstowcs which does not count the characters that

* would result from conversion. Use an upper bound.

argsize = strlen(arg);

#else

argsize = mbstowcs(NULL, arg, 0);

#endif

if (argsize != (size_t)-1) {

if (argsize == PY_SSIZE_T_MAX)

goto oom;

argsize += 1;

if (argsize > PY_SSIZE_T_MAX/sizeof(wchar_t))

goto oom;

res = (wchar_t *)PyMem_Malloc(argsize*sizeof(wchar_t));

if (!res)

goto oom;

count = mbstowcs(res, arg, argsize);

if (count != (size_t)-1) {

wchar_t *tmp;

/* Only use the result if it contains no

surrogate characters. */

for (tmp = res; *tmp != 0 &&

(*tmp < 0xd800 || *tmp > 0xdfff); tmp++)

;

if (*tmp == 0) {

if (size != NULL)

*size = count;

return res;

}

PyMem_Free(res);

}

/* Conversion failed. Fall back to escaping with surrogateescape. */

#ifdef HAVE_MBRTOWC

/* Try conversion with mbrtwoc (C99), and escape non-decodable bytes. */

/* Overallocate; as multi-byte characters are in the argument, the

actual output could use less memory. */

argsize = strlen(arg) + 1;

if (argsize > PY_SSIZE_T_MAX/sizeof(wchar_t))

goto oom;

res = (wchar_t*)PyMem_Malloc(argsize*sizeof(wchar_t));

if (!res)

goto oom;

in = (unsigned char*)arg;

out = res;

memset(&mbs, 0, sizeof mbs);

while (argsize) {

size_t converted = mbrtowc(out, (char*)in, argsize, &mbs);

if (converted == 0)

/* Reached end of string; null char stored. */

break;

if (converted == (size_t)-2) {

/* Incomplete character. This should never happen,

since we provide everything that we have -

unless there is a bug in the C library, or I

misunderstood how mbrtowc works. */

PyMem_Free(res);

if (size != NULL)

*size = (size_t)-2;

return NULL;

}

if (converted == (size_t)-1) {

/* Conversion error. Escape as UTF-8b, and start over

in the initial shift state. */

*out++ = 0xdc00 + *in++;

argsize--;

memset(&mbs, 0, sizeof mbs);

continue;

}

if (*out >= 0xd800 && *out <= 0xdfff) {

/* Surrogate character. Escape the original

byte sequence with surrogateescape. */

argsize -= converted;

while (converted--)

*out++ = 0xdc00 + *in++;

continue;

}

/* successfully converted some bytes */

in += converted;

argsize -= converted;

out++;

}

if (size != NULL)

*size = out - res;

#else /* HAVE_MBRTOWC */

/* Cannot use C locale for escaping; manually escape as if charset

is ASCII (i.e. escape all bytes > 128. This will still roundtrip

correctly in the locale's charset, which must be an ASCII superset. */

res = decode_ascii_surrogateescape(arg, size);

if (res == NULL)

goto oom;

#endif /* HAVE_MBRTOWC */

return res;

oom:

if (size != NULL)

*size = (size_t)-1;

return NULL;

#endif /* __APPLE__ */

}

/* Encode a (wide) character string to the locale encoding with the

surrogateescape error handler (characters in range U+DC80..U+DCFF are

converted to bytes 0x80..0xFF).

This function is the reverse of _Py_char2wchar().

Return a pointer to a newly allocated byte string (use PyMem_Free() to free

the memory), or NULL on encoding or memory allocation error.

If error_pos is not NULL: *error_pos is the index of the invalid character

on encoding error, or (size_t)-1 otherwise. */

char*

_Py_wchar2char(const wchar_t *text, size_t *error_pos)

{

#ifdef __APPLE__

Py_ssize_t len;

PyObject *unicode, *bytes = NULL;

char *cpath;

unicode = PyUnicode_FromWideChar(text, wcslen(text));

if (unicode == NULL)

return NULL;

bytes = _PyUnicode_AsUTF8String(unicode, "surrogateescape");

Py_DECREF(unicode);

if (bytes == NULL) {

PyErr_Clear();

if (error_pos != NULL)

*error_pos = (size_t)-1;

return NULL;

}

len = PyBytes_GET_SIZE(bytes);

cpath = PyMem_Malloc(len+1);

if (cpath == NULL) {

PyErr_Clear();

Py_DECREF(bytes);

if (error_pos != NULL)

*error_pos = (size_t)-1;

return NULL;

}

memcpy(cpath, PyBytes_AsString(bytes), len + 1);

Py_DECREF(bytes);

return cpath;

#else /* __APPLE__ */

const size_t len = wcslen(text);

char *result = NULL, *bytes = NULL;

size_t i, size, converted;

wchar_t c, buf[2];

#ifndef MS_WINDOWS

if (force_ascii == -1)

force_ascii = check_force_ascii();

if (force_ascii)

return encode_ascii_surrogateescape(text, error_pos);

#endif

/* The function works in two steps:

1. compute the length of the output buffer in bytes (size)

2. outputs the bytes */

size = 0;

buf[1] = 0;

while (1) {

for (i=0; i < len; i++) {

c = text[i];

if (c >= 0xdc80 && c <= 0xdcff) {

/* UTF-8b surrogate */

if (bytes != NULL) {

*bytes++ = c - 0xdc00;

size--;

}

else

size++;

continue;

}

else {

buf[0] = c;

if (bytes != NULL)

converted = wcstombs(bytes, buf, size);

else

converted = wcstombs(NULL, buf, 0);

if (converted == (size_t)-1) {

if (result != NULL)

PyMem_Free(result);

if (error_pos != NULL)

*error_pos = i;

return NULL;

}

if (bytes != NULL) {

bytes += converted;

size -= converted;

}

else

size += converted;

}

if (result != NULL) {

*bytes = '\0';

break;

}

size += 1; /* nul byte at the end */

result = PyMem_Malloc(size);

if (result == NULL) {

if (error_pos != NULL)

*error_pos = (size_t)-1;

return NULL;

}

bytes = result;

}

return result;

#endif /* __APPLE__ */

}

/* In principle, this should use HAVE__WSTAT, and _wstat

should be detected by autoconf. However, no current

POSIX system provides that function, so testing for

it is pointless.

Not sure whether the MS_WINDOWS guards are necessary:

perhaps for cygwin/mingw builds?

#if defined(HAVE_STAT) && !defined(MS_WINDOWS)

/* Get file status. Encode the path to the locale encoding. */

int

_Py_wstat(const wchar_t* path, struct stat *buf)

{

int err;

char *fname;

fname = _Py_wchar2char(path, NULL);

if (fname == NULL) {

errno = EINVAL;

return -1;

}

err = stat(fname, buf);

PyMem_Free(fname);

return err;

}

#endif

#ifdef HAVE_STAT

/* Call _wstat() on Windows, or encode the path to the filesystem encoding and

call stat() otherwise. Only fill st_mode attribute on Windows.

Return 0 on success, -1 on _wstat() / stat() error, -2 if an exception was

raised. */

int

_Py_stat(PyObject *path, struct stat *statbuf)

{

#ifdef MS_WINDOWS

int err;

struct _stat wstatbuf;

wchar_t *wpath;

wpath = PyUnicode_AsUnicode(path);

if (wpath == NULL)

return -2;

err = _wstat(wpath, &wstatbuf);

if (!err)

statbuf->st_mode = wstatbuf.st_mode;

return err;

#else

int ret;

PyObject *bytes = PyUnicode_EncodeFSDefault(path);

if (bytes == NULL)

return -2;

ret = stat(PyBytes_AS_STRING(bytes), statbuf);

Py_DECREF(bytes);

return ret;

#endif

}

#endif

/* Open a file. Use _wfopen() on Windows, encode the path to the locale

encoding and use fopen() otherwise. */

FILE *

_Py_wfopen(const wchar_t *path, const wchar_t *mode)

{

#ifndef MS_WINDOWS

FILE *f;

char *cpath;

char cmode[10];

size_t r;

r = wcstombs(cmode, mode, 10);

if (r == (size_t)-1 || r >= 10) {

errno = EINVAL;

return NULL;

}

cpath = _Py_wchar2char(path, NULL);

if (cpath == NULL)

return NULL;

f = fopen(cpath, cmode);

PyMem_Free(cpath);

return f;

#else

return _wfopen(path, mode);

#endif

}

/* Call _wfopen() on Windows, or encode the path to the filesystem encoding and

call fopen() otherwise.

Return the new file object on success, or NULL if the file cannot be open or

(if PyErr_Occurred()) on unicode error */

FILE*

_Py_fopen(PyObject *path, const char *mode)

{

#ifdef MS_WINDOWS

wchar_t *wpath;

wchar_t wmode[10];

int usize;

if (!PyUnicode_Check(path)) {

PyErr_Format(PyExc_TypeError,

"str file path expected under Windows, got %R",

Py_TYPE(path));

return NULL;

}

wpath = PyUnicode_AsUnicode(path);

if (wpath == NULL)

return NULL;

usize = MultiByteToWideChar(CP_ACP, 0, mode, -1, wmode, sizeof(wmode));

if (usize == 0)

return NULL;

return _wfopen(wpath, wmode);

#else

FILE *f;

PyObject *bytes;

if (!PyUnicode_FSConverter(path, &bytes))

return NULL;

f = fopen(PyBytes_AS_STRING(bytes), mode);

Py_DECREF(bytes);

return f;

#endif

}

#ifdef HAVE_READLINK

/* Read value of symbolic link. Encode the path to the locale encoding, decode

the result from the locale encoding. Return -1 on error. */

int

_Py_wreadlink(const wchar_t *path, wchar_t *buf, size_t bufsiz)

{

char *cpath;

char cbuf[MAXPATHLEN];

wchar_t *wbuf;

int res;

size_t r1;

cpath = _Py_wchar2char(path, NULL);

if (cpath == NULL) {

errno = EINVAL;

return -1;

}

res = (int)readlink(cpath, cbuf, Py_ARRAY_LENGTH(cbuf));

PyMem_Free(cpath);

if (res == -1)

return -1;

if (res == Py_ARRAY_LENGTH(cbuf)) {

errno = EINVAL;

return -1;

}

cbuf[res] = '\0'; /* buf will be null terminated */

wbuf = _Py_char2wchar(cbuf, &r1);

if (wbuf == NULL) {

errno = EINVAL;

return -1;

}

if (bufsiz <= r1) {

PyMem_Free(wbuf);

errno = EINVAL;

return -1;

}

wcsncpy(buf, wbuf, bufsiz);

PyMem_Free(wbuf);

return (int)r1;

}

#endif

#ifdef HAVE_REALPATH

/* Return the canonicalized absolute pathname. Encode path to the locale

encoding, decode the result from the locale encoding.

Return NULL on error. */

wchar_t*

_Py_wrealpath(const wchar_t *path,

wchar_t *resolved_path, size_t resolved_path_size)

{

char *cpath;

char cresolved_path[MAXPATHLEN];

wchar_t *wresolved_path;

char *res;

size_t r;

cpath = _Py_wchar2char(path, NULL);

if (cpath == NULL) {

errno = EINVAL;

return NULL;

}

res = realpath(cpath, cresolved_path);

PyMem_Free(cpath);

if (res == NULL)

return NULL;

wresolved_path = _Py_char2wchar(cresolved_path, &r);

if (wresolved_path == NULL) {

errno = EINVAL;

return NULL;

}

if (resolved_path_size <= r) {

PyMem_Free(wresolved_path);

errno = EINVAL;

return NULL;

}

wcsncpy(resolved_path, wresolved_path, resolved_path_size);

PyMem_Free(wresolved_path);

return resolved_path;

}

#endif

/* Get the current directory. size is the buffer size in wide characters

including the null character. Decode the path from the locale encoding.

Return NULL on error. */

wchar_t*

_Py_wgetcwd(wchar_t *buf, size_t size)

{

#ifdef MS_WINDOWS

return _wgetcwd(buf, size);

#else

char fname[MAXPATHLEN];

wchar_t *wname;

size_t len;

if (getcwd(fname, Py_ARRAY_LENGTH(fname)) == NULL)

return NULL;

wname = _Py_char2wchar(fname, &len);

if (wname == NULL)

return NULL;

if (size <= len) {

PyMem_Free(wname);

return NULL;

}

wcsncpy(buf, wname, size);

PyMem_Free(wname);

return buf;

#endif

}

cpython/Python/fileutils.c at 3.3 · python/cpython · GitHub

Repository navigation

Expand file tree

Breadcrumbs

fileutils.c

Latest commit

History

Breadcrumbs

fileutils.c

File metadata and controls