|
1
|
/* zran.h -- example of deflate stream indexing and random access |
|
2
|
* Copyright (C) 2005, 2012, 2018, 2023, 2024, 2025 Mark Adler |
|
3
|
* For conditions of distribution and use, see copyright notice in zlib.h |
|
4
|
* Version 1.7 16 May 2025 Mark Adler */ |
|
5
|
|
|
6
|
#include <stdio.h> |
|
7
|
#include "zlib.h" |
|
8
|
|
|
9
|
// Access point. |
|
10
|
typedef struct point { |
|
11
|
off_t out; // offset in uncompressed data |
|
12
|
off_t in; // offset in compressed file of first full byte |
|
13
|
int bits; // 0, or number of bits (1-7) from byte at in-1 |
|
14
|
unsigned dict; // number of bytes in window to use as a dictionary |
|
15
|
unsigned char *window; // preceding 32K (or less) of uncompressed data |
|
16
|
} point_t; |
|
17
|
|
|
18
|
// Access point list. |
|
19
|
struct deflate_index { |
|
20
|
int have; // number of access points in list |
|
21
|
int mode; // -15 for raw, 15 for zlib, or 31 for gzip |
|
22
|
off_t length; // total length of uncompressed data |
|
23
|
point_t *list; // allocated list of access points |
|
24
|
z_stream strm; // re-usable inflate engine for extraction |
|
25
|
}; |
|
26
|
|
|
27
|
// Make one pass through a zlib, gzip, or raw deflate compressed stream and |
|
28
|
// build an index, with access points about every span bytes of uncompressed |
|
29
|
// output. gzip files with multiple members are fully indexed. span should be |
|
30
|
// chosen to balance the speed of random access against the memory requirements |
|
31
|
// of the list, which is about 32K bytes per access point. The return value is |
|
32
|
// the number of access points on success (>= 1), Z_MEM_ERROR for out of |
|
33
|
// memory, Z_BUF_ERROR for a premature end of input, Z_DATA_ERROR for a format |
|
34
|
// or verification error in the input file, or Z_ERRNO for a file read error. |
|
35
|
// On success, *built points to the resulting index, otherwise it's NULL. |
|
36
|
int deflate_index_build(FILE *in, off_t span, struct deflate_index **built); |
|
37
|
|
|
38
|
// Use the index to read len bytes from offset into buf. Return the number of |
|
39
|
// bytes read or a negative error code. If data is requested past the end of |
|
40
|
// the uncompressed data, then deflate_index_extract() will return a value less |
|
41
|
// than len, indicating how much was actually read into buf. If given a valid |
|
42
|
// index, this function should not return an error unless the file was modified |
|
43
|
// somehow since the index was generated, given that deflate_index_build() had |
|
44
|
// validated all of the input. If nevertheless there is a failure, Z_BUF_ERROR |
|
45
|
// is returned if the compressed data ends prematurely, Z_DATA_ERROR if the |
|
46
|
// deflate compressed data is not valid, Z_MEM_ERROR if out of memory, |
|
47
|
// Z_STREAM_ERROR if the index is not valid, or Z_ERRNO if there is an error |
|
48
|
// reading or seeking on the input file. |
|
49
|
ptrdiff_t deflate_index_extract(FILE *in, struct deflate_index *index, |
|
50
|
off_t offset, unsigned char *buf, size_t len); |
|
51
|
|
|
52
|
// Deallocate an index built by deflate_index_build(). |
|
53
|
void deflate_index_free(struct deflate_index *index); |
|
54
|
|