Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/wkoszek/mini_gzip
mini_gzip - embeddable, minimal, in-memory GZIP API
https://github.com/wkoszek/mini_gzip
c compression decompression embeddable gzip mini-gzip zlib
Last synced: 9 days ago
JSON representation
mini_gzip - embeddable, minimal, in-memory GZIP API
- Host: GitHub
- URL: https://github.com/wkoszek/mini_gzip
- Owner: wkoszek
- License: bsd-2-clause
- Created: 2013-07-15T06:21:14.000Z (over 11 years ago)
- Default Branch: master
- Last Pushed: 2023-06-09T11:32:24.000Z (over 1 year ago)
- Last Synced: 2024-07-31T15:02:56.040Z (3 months ago)
- Topics: c, compression, decompression, embeddable, gzip, mini-gzip, zlib
- Language: C
- Homepage:
- Size: 46.9 KB
- Stars: 109
- Watchers: 12
- Forks: 26
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
mini_gzip - embeddable, minimal, in-memory GZIP API
===================================================
[](https://travis-ci.org/wkoszek/mini_gzip)
Embeddable, minimal GZIP functionality, with API designed to work from
memory. Only supports decompression for now.
I wrote it when I needed a small piece of code to give me decompression in
an embedded system I was working with, and I found out that most of the
GZIP-related programs use POSIX `FILE` API, which made the examples not work
for my case.
The `mini_gzip` is based on `miniz` library
(https://code.google.com/p/miniz/), which provides an API for operating on
data compressed according to deflate algorithm. Added is a layer which
provides a container for the GZIP files and let me do some verification.
# To build
Everything should be fairly self-contained, and built with:
make
# How to use
The API operates on `struct mini_gzip` structures, which are containers for
the memory with GZIPed data. You must call `mini_gz_init()` on the structure
1st for the rest of the API to work correctly. Next, you must call
`mini_gz_start(&gz, mem_in, size)`, where `gz` is the initialized container
and with GZIPed data pointed by `mem_in` of the size of `size`. To unpack,
you call `mini_gz_unpack(&gz, mem_out, memsize)`, where the first argument
is the initialized container, the second `mem_out` is a pointer to the
output memory, and `memsize` is its lenght. It's important to have enough
bytes in the output buffer for decompressed data.
# To test
Simple test is provided:
make test
To test by yourself, you can do:
~~~shell
wk:/w/repos/mini_gzip> ls -la miniz.o
-rw-r--r-- 1 wk staff 114348 20 paź 09:46 miniz.o
wk:/w/repos/mini_gzip> md5 miniz.o
MD5 (miniz.o) = e6199aade2020b6040fa160baee47d68
wk:/w/repos/mini_gzip> gzip miniz.o
wk:/w/repos/mini_gzip> ls -la miniz.o.gz
-rw-r--r-- 1 wk staff 44965 20 paź 09:46 miniz.o.gz
wk:/w/repos/mini_gzip> ./mini_gzip miniz.o.gz miniz.o
flag_c: 0 is_gzipped: 1
in_fn: miniz.o.gz out_fn: miniz.o level 6
--- testing decompression --
out_len = 114348
ret = 114348
wk:/w/repos/mini_gzip> md5 miniz.o
MD5 (miniz.o) = e6199aade2020b6040fa160baee47d68
~~~
# References
After publishing `mini_gzip` on HackerNews, byuu (http://byuu.org/) mentioned he has much leaner implementation of GZIP:
- https://gitlab.com/higan/higan/blob/master/nall/decode/gzip.hpp
- https://gitlab.com/higan/higan/blob/master/nall/decode/inflate.hpp
# Author
- Wojciech Adam Koszek, [[email protected]](mailto:[email protected])
- [http://www.koszek.com](http://www.koszek.com)