693285bc87
increases performance when extracting a single entry from a large uncompressed archive, especially on slow devices such as USB hard drives. Requires a number of changes: * New archive_read_open2() supports a 'skip' client function * Old archive_read_open() is implemented as a wrapper now, to continue supporting the old API/ABI. * _read_open_fd and _read_open_file sprout new 'skip' functions. * compression layer gets a new 'skip' operation. * compression_none passes skip requests through to client. * compression_{gzip,bzip2,compress} simply ignore skip requests. Thanks to: Benjamin Lutz, who designed and implemented the whole thing. I'm just committing it. ;-) TODO: Need to update the documentation a little bit.
149 lines
4.3 KiB
C
149 lines
4.3 KiB
C
/*-
|
|
* Copyright (c) 2003-2004 Tim Kientzle
|
|
* All rights reserved.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer
|
|
* in this position and unchanged.
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
|
|
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
|
|
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
|
|
* IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
|
|
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
|
|
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
|
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
|
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
|
|
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
*/
|
|
|
|
#include "archive_platform.h"
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
#include <sys/stat.h>
|
|
#include <errno.h>
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include <unistd.h>
|
|
|
|
#include "archive.h"
|
|
#include "archive_private.h"
|
|
|
|
struct read_fd_data {
|
|
int fd;
|
|
size_t block_size;
|
|
void *buffer;
|
|
};
|
|
|
|
static int file_close(struct archive *, void *);
|
|
static int file_open(struct archive *, void *);
|
|
static ssize_t file_read(struct archive *, void *, const void **buff);
|
|
static ssize_t file_skip(struct archive *, void *, size_t request);
|
|
|
|
int
|
|
archive_read_open_fd(struct archive *a, int fd, size_t block_size)
|
|
{
|
|
struct read_fd_data *mine;
|
|
|
|
mine = malloc(sizeof(*mine));
|
|
if (mine == NULL) {
|
|
archive_set_error(a, ENOMEM, "No memory");
|
|
return (ARCHIVE_FATAL);
|
|
}
|
|
mine->block_size = block_size;
|
|
mine->buffer = malloc(mine->block_size);
|
|
if (mine->buffer == NULL) {
|
|
archive_set_error(a, ENOMEM, "No memory");
|
|
free(mine);
|
|
return (ARCHIVE_FATAL);
|
|
}
|
|
mine->fd = fd;
|
|
return (archive_read_open2(a, mine, file_open, file_read, file_skip, file_close));
|
|
}
|
|
|
|
static int
|
|
file_open(struct archive *a, void *client_data)
|
|
{
|
|
struct read_fd_data *mine = client_data;
|
|
struct stat st;
|
|
|
|
if (fstat(mine->fd, &st) != 0) {
|
|
archive_set_error(a, errno, "Can't stat fd %d", mine->fd);
|
|
return (ARCHIVE_FATAL);
|
|
}
|
|
|
|
a->skip_file_dev = st.st_dev;
|
|
a->skip_file_ino = st.st_ino;
|
|
return (ARCHIVE_OK);
|
|
}
|
|
|
|
static ssize_t
|
|
file_read(struct archive *a, void *client_data, const void **buff)
|
|
{
|
|
struct read_fd_data *mine = client_data;
|
|
ssize_t bytes_read;
|
|
|
|
*buff = mine->buffer;
|
|
bytes_read = read(mine->fd, mine->buffer, mine->block_size);
|
|
if (bytes_read < 0) {
|
|
archive_set_error(a, errno, "Error reading fd %d", mine->fd);
|
|
}
|
|
return (bytes_read);
|
|
}
|
|
|
|
static ssize_t
|
|
file_skip(struct archive *a, void *client_data, size_t request)
|
|
{
|
|
struct read_fd_data *mine = client_data;
|
|
off_t old_offset, new_offset;
|
|
|
|
/* Reduce request to the next smallest multiple of block_size */
|
|
request = (request / mine->block_size) * mine->block_size;
|
|
/*
|
|
* Hurray for lazy evaluation: if the first lseek fails, the second
|
|
* one will not be executed.
|
|
*/
|
|
if (((old_offset = lseek(mine->fd, 0, SEEK_CUR)) < 0) ||
|
|
((new_offset = lseek(mine->fd, request, SEEK_CUR)) < 0))
|
|
{
|
|
if (errno == ESPIPE)
|
|
{
|
|
/*
|
|
* Failure to lseek() can be caused by the file
|
|
* descriptor pointing to a pipe, socket or FIFO.
|
|
* Return 0 here, so the compression layer will use
|
|
* read()s instead to advance the file descriptor.
|
|
* It's slower of course, but works as well.
|
|
*/
|
|
return (0);
|
|
}
|
|
/*
|
|
* There's been an error other than ESPIPE. This is most
|
|
* likely caused by a programmer error (too large request)
|
|
* or a corrupted archive file.
|
|
*/
|
|
archive_set_error(a, errno, "Error seeking");
|
|
return (-1);
|
|
}
|
|
return (new_offset - old_offset);
|
|
}
|
|
|
|
static int
|
|
file_close(struct archive *a, void *client_data)
|
|
{
|
|
struct read_fd_data *mine = client_data;
|
|
|
|
(void)a; /* UNUSED */
|
|
if (mine->buffer != NULL)
|
|
free(mine->buffer);
|
|
free(mine);
|
|
return (ARCHIVE_OK);
|
|
}
|