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
|
/**
cfile_tools.c
A library to deal transparently with possibly compressed files.
Documentation in the function headers and in cfile_tools.h
Copyright (C) 2004 by Arno Wagner <arno.wagner@acm.org>
Distributed under the Gnu Public License version 2 or the modified
BSD license (see file COPYING)
Support for gzip added by Bernhard Tellenbach <bernhard.tellenbach@gmail.com>
*/
#define _GNU_SOURCE
#define _FILE_OFFSET_BITS 64
//#ifndef DONT_HAVE_BZ2
//#include <bzlib.h>
//#endif
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include <assert.h>
#include "cfile_tools.h"
// Concrete formats. remember to adjust CFR_NUM_FORMATS if changed!
// Note: 0, 1 are special entries.
const char * cfr_formats[CFR_NUM_FORMATS] = {
"not open", // 0
"uncompressed", // 1
"bzip2", // 2
"gzip", // 3
};
const char * cfr_extensions[CFR_NUM_FORMATS] = {
"", // 0
"", // 1
".bz2", // 2
".gz" // 3
};
// Prototypes of non API functions (don't use these from outside this file)
const char * _cfr_compressor_strerror(int format, int err);
const char * _bz2_strerror(int err);
// API Functions
CFRFILE *cfr_open(const char *path) {
/*******************************/
// Analog to 'fopen'. Error in result has to be tested using
// 'cfr_error' on the result!
// Note: The user needs to free the reurn value!
// Opens a possibly compressed file for reading.
// File type is determined by file name ending
int format, ext_len, name_len;
CFRFILE * retval = NULL;
// determine file format
name_len = strlen(path);
format = 2; // skip specials 0, 1
// Do action dependent on file format
retval = (CFRFILE *) calloc(1,sizeof(CFRFILE));
if(retval == NULL)
return (NULL);
retval->eof = 0;
retval->error1 = 0;
retval->error2 = 0;
if((path == NULL) || (strcmp(path, "-") == 0)) {
/* dump from stdin */
gzFile f;
while (format < CFR_NUM_FORMATS) {
if (strcmp(cfr_extensions[format], ".gz") == 0)
break;
format ++;
}
f = gzdopen(0, "r");
if(f == NULL) {
free(retval);
return (NULL);
}
retval->data2 = f;
retval->format = format;
return (retval);
}
while (format < CFR_NUM_FORMATS) {
ext_len = strlen(cfr_extensions[format]);
if (strncmp(cfr_extensions[format],
path+(name_len-ext_len),
ext_len) == 0
) break;
format ++;
}
if (format >= CFR_NUM_FORMATS)
format = 1; // uncompressed
retval->format = format;
switch (format) {
case 1: // uncompressed
{
FILE * in;
in = fopen(path,"r");
if (in == NULL) {
free(retval);
return(NULL);
}
retval->data1 = in;
return(retval);
}
break;
case 2: // bzip2
{
int bzerror;
BZFILE * bzin;
FILE * in;
retval->bz2_stream_end = 0;
// get file
in = fopen(path,"r");
if (in == NULL) {
free(retval);
return(NULL);
}
retval->data1 = in;
// bzip2ify file
bzin = BZ2_bzReadOpen( &bzerror, in, 0, 0, NULL, 0);
if (bzerror != BZ_OK) {
errno = bzerror;
BZ2_bzReadClose( &bzerror, bzin);
fclose(in);
free(retval);
return(NULL);
}
retval->data2 = bzin;
return(retval);
}
break;
case 3: // gzip
{
gzFile f;
// get file
f = gzopen(path, "r");
if(f == NULL) {
free(retval);
return (NULL);
}
retval->data2 = f;
return (retval);
}
break;
default: // this is an internal error, no diag yet.
fprintf(stderr,"illegal format '%d' in cfr_open!\n", format);
exit(1);
}
return NULL;
}
int cfr_close(CFRFILE *stream) {
/**************************/
// Analog to 'fclose'.
// FIXME - why is stream->* set, then freed?
if (stream == NULL || stream->closed) {
errno = EBADF;
return -1;
}
int retval = -1;
switch (stream->format) {
case 1: // uncompressed
retval = fclose((FILE *)(stream->data1));
stream->error1 = retval;
break;
case 2: // bzip2
BZ2_bzReadClose( &stream->error2, (BZFILE *)stream->data2);
stream->error1 = retval = fclose((FILE *)(stream->data1));
break;
case 3: // gzip
if(stream->data2!=NULL)
retval = gzclose(stream->data2);
stream->error2 = retval;
break;
default: // internal error
assert("illegal stream->format" && 0);
}
free(stream);
return(retval);
}
size_t cfr_read_n(CFRFILE *stream, void *ptr, size_t bytes) {
/******************************************************************/
// Wrapper, will return either 'bytes' (the number of bytes to read) or 0
return(cfr_read(ptr, bytes, 1, stream)*bytes);
}
size_t cfr_read(void *ptr, size_t size, size_t nmemb, CFRFILE *stream) {
/******************************************************************/
// Analog to 'fread'. Will not return with partial elements, only
// full ones. Hence calling this function with one large element
// size will result in a complete or no read.
size_t retval = 0;
if (stream == NULL) return(0);
// shortcut
if (stream->eof) return(0);
switch (stream->format) {
case 1: // uncompressed
{
FILE * in;
in = (FILE *)(stream->data1);
retval = fread(ptr, size, nmemb, in);
if (retval != nmemb) {
// fprintf(stderr,"short read!!!\n");
stream->eof = feof(in);
stream->error1 = ferror(in);
retval = 0;
}
return (retval);
}
break;
case 2: // bzip2
{
BZFILE * bzin;
int bzerror;
int buffsize;
if (stream->bz2_stream_end == 1) {
// feof-behaviour: Last read did consume last byte but not more
stream->eof = 1;
return(0);
}
bzerror = BZ_OK;
bzin = (BZFILE *) (stream->data2);
buffsize = size * nmemb;
retval = BZ2_bzRead(&bzerror, bzin, ptr, buffsize);
if (bzerror == BZ_STREAM_END ) {
stream->bz2_stream_end = 1;
stream->error2 = bzerror;
if (retval == buffsize) {
// feof-behaviour: no eof yet
} else {
// feof-behaviour: read past end, set eof
stream->eof = 1;
retval = 0;
}
return(retval/size);
}
if (bzerror == BZ_OK) {
// Normal case, no error.
// A short read here is an error, so catch it
if (retval == buffsize) {
return(retval/size);
}
}
// Other error...
stream->error2 = bzerror;
BZ2_bzReadClose( &bzerror, bzin );
if (bzerror != BZ_OK) {
stream->error2 = bzerror;
}
retval = fclose((FILE *)(stream->data1));
stream->error1 = retval;
stream->closed = 1;
return(0);
}
break;
case 3: // gzip
{
gzFile in;
in = (gzFile)(stream->data2);
retval = gzread(in, ptr, size*nmemb);
if (retval != nmemb*size) {
// fprintf(stderr,"short read!!!\n");
stream->eof = gzeof(in);
stream->error2 = errno;
retval = 0;
}
return (retval/size);
}
break;
default: // this is an internal error, no diag yet.
fprintf(stderr,"illegal format '%d' in cfr_read!\n",stream->format);
exit(1);
}
}
ssize_t cfr_getline(char **lineptr, size_t *n, CFRFILE *stream) {
/************************************************************/
// May not be very efficient, since it uses single-char reads
// for formats where there is no native getline in the library.
// For bzip2 the speedup for additional buffering was only 5%
// so I dropped it.
// Returns -1 in case of an error.
char *tmp;
if (stream == NULL) return(-1);
switch (stream->format) {
case 1: // uncompressed
{
if (fgets(*lineptr, *n, (FILE *)(stream->data1)) == NULL) {
stream->error1 = errno;
return -1;
}
return 0;
}
break;
case 2: // bzip2
{
size_t count;
char c;
size_t ret;
//bzin = (BZFILE *) (stream->data2);
// allocate initial buffer if none was passed or size was zero
if (*lineptr == NULL) {
*lineptr = (char *) calloc(120, 1);
if(*lineptr == NULL) {
stream->error1 = errno;
return(-1);
}
*n = 120;
}
if (*n == 0) {
*n = 120;
tmp = (char *) realloc(*lineptr, *n); // to avoid memory-leaks
if(tmp == NULL) {
stream->error1 = errno;
return(-1);
}
*lineptr = tmp;
}
count = 0;
// read until '\n'
do {
ret = cfr_read(&c, 1, 1, stream);
if (ret != 1) {
return(-1);
}
count ++;
if (count >= *n) {
*n = 2 * *n;
tmp = (char *) realloc(*lineptr, *n);
if (tmp == NULL) {
stream->error1 = errno;
return(-1);
}
*lineptr = tmp;
}
(*lineptr)[count-1] = c;
} while (c != '\n');
(*lineptr)[count] = 0;
return(count);
}
break;
case 3: // gzip
{
char * return_ptr = gzgets((gzFile)(stream->data2), *lineptr, *n );
if (return_ptr == Z_NULL) {
stream->error2 = errno;
return(-1);
}
return *n;
}
break;
default: // this is an internal error, no diag yet.
fprintf(stderr,"illegal format '%d' in cfr_getline!\n",stream->format);
exit(1);
return(-1);
}
}
int cfr_eof(CFRFILE *stream) {
// Returns true on end of file/end of compressed data.
// The end of the compressed data is regarded as end of file
// in this library, embedded or multiple compressed data per
// file is not supported by this library.
//
// Note: The sematics is that cfr_eof is true only after
// the first byte after the end of file was read. Some compressors
// report EOF already when the last availale character has been
// read (far more sensible IMO), but for consistency we follow the
// convention of the standard c library here.
return(stream->eof);
}
int cfr_error(CFRFILE *stream) {
// Returns true on error.
// Errors can be ordinary errors from fopen.fclose/fread
// or can originate from the underlying compression.
// This function just returns 0 when there is no error or
// 1 in case of error.
// To get a more detailed report cfr_strerror will try to
// come up with a description of the whole situation.
// For numeric details, more query functions would need to be
// implemented.
if (stream == NULL) return(1);
return(stream->error1 || stream->error2);
}
#if 0
char * cfr_strerror(CFRFILE *stream) {
// Result is "stream-i/o: <stream-error> <compressor>[: <compressor error>]"
// Do not modify result.
// Result may change on subsequent call to this function.
static char res[120];
int ret;
char * msg, * msg2;
if (stream == NULL) {
snprintf(res, sizeof(res), "%s", "Error: stream is NULL, i.e. not opened");
return(res);
}
ret = asprintf(&msg,
"stream-i/o: %s, %s [%s]",
stream->eof?"EOF":"",
strerror(stream->error1),
cfr_compressor_str(stream));
if (ret == -1)
goto oom;
if (stream->format == 2) {
ret = asprintf(&msg2,
"%s: %s",
msg,
_cfr_compressor_strerror(stream->format, stream->error2));
free(msg);
if (ret == -1)
goto oom;
msg = msg2;
}
if (stream->format == 3) {
ret = asprintf(&msg2,
"%s: %s",
msg,
gzerror((gzFile)(stream->data2), &(stream->error2)));
free(msg);
if (ret == -1)
goto oom;
msg = msg2;
}
snprintf(res, sizeof(res), "%s", msg);
free(msg);
return(res);
oom:
snprintf(res, sizeof(res), "%s", "Error: asprintf: out of memory");
return(res);
}
#endif
const char * cfr_compressor_str(CFRFILE *stream) {
// Returns the name of the compressor used
if ((stream->format < 0) || (stream->format >= CFR_NUM_FORMATS)) {
return("undefined compression type");
} else {
return (cfr_formats[stream->format]);
}
}
// Utility functions for compressor errors.
// * Not part of the API, do not call directly as they may change! *
const char * _cfr_compressor_strerror(int format, int err) {
// Transforms error code to string for all compressors
switch (format) {
case 0:
return("file not open");
break;
case 1:
return("file not compressed");
break;
case 2:
return(_bz2_strerror(err));
break;
case 3:
return NULL;
break;
default:
return("unknowen compressor code");
}
}
const char * _bz2_strerror(int err) {
// Since bzlib does not have strerror, we do it here manually.
// This works for version 1.0 of 21 March 2000 of bzlib.h
switch (err) {
case BZ_OK: return("BZ_OK");
case BZ_RUN_OK: return("BZ_RUN_OK");
case BZ_FLUSH_OK: return("BZ_FLUSH_OK");
case BZ_FINISH_OK: return("BZ_FINISH_OK");
case BZ_STREAM_END: return("BZ_STREAM_END");
case BZ_SEQUENCE_ERROR: return("BZ_SEQUENCE_ERROR");
case BZ_PARAM_ERROR: return("BZ_PARAM_ERROR");
case BZ_MEM_ERROR: return("BZ_MEM_ERROR");
case BZ_DATA_ERROR: return("BZ_DATA_ERROR");
case BZ_DATA_ERROR_MAGIC: return("BZ_DATA_ERROR_MAGIC");
case BZ_IO_ERROR: return("BZ_IO_ERROR");
case BZ_UNEXPECTED_EOF: return("BZ_UNEXPECTED_EOF");
case BZ_OUTBUFF_FULL: return("BZ_OUTBUFF_FULL");
case BZ_CONFIG_ERROR: return("BZ_CONFIG_ERROR");
default: return("unknowen bzip2 error code");
}
}
|