diff options
author | Devtools Arcadia <arcadia-devtools@yandex-team.ru> | 2022-02-07 18:08:42 +0300 |
---|---|---|
committer | Devtools Arcadia <arcadia-devtools@mous.vla.yp-c.yandex.net> | 2022-02-07 18:08:42 +0300 |
commit | 1110808a9d39d4b808aef724c861a2e1a38d2a69 (patch) | |
tree | e26c9fed0de5d9873cce7e00bc214573dc2195b7 /library/cpp/blockcodecs/README.md | |
download | ydb-1110808a9d39d4b808aef724c861a2e1a38d2a69.tar.gz |
intermediate changes
ref:cde9a383711a11544ce7e107a78147fb96cc4029
Diffstat (limited to 'library/cpp/blockcodecs/README.md')
-rw-r--r-- | library/cpp/blockcodecs/README.md | 23 |
1 files changed, 23 insertions, 0 deletions
diff --git a/library/cpp/blockcodecs/README.md b/library/cpp/blockcodecs/README.md new file mode 100644 index 0000000000..417917a475 --- /dev/null +++ b/library/cpp/blockcodecs/README.md @@ -0,0 +1,23 @@ +This is a simple library for block data compression (this means data is compressed/uncompressed +by whole blocks in memory). It's a lite-version of the `library/cpp/codecs`. Lite here means that it +provide only well-known compression algorithms, without the possibility of learning. + +There are two possible ways to work with it. + +Codec by name +============= +Use `NBlockCodec::Codec` to obtain the codec by name. The codec can be asked to compress +or decompress something and in various ways. + +To get a full list of codecs there is a function `NBlockCodecs::ListAllCodecs()`. + +Streaming +========= +Use `stream.h` to obtain simple streams over block codecs (buffer data, compress them by blocks, +write to the resulting stream). + +Using codec plugins +=================== +If you don't want your code to bloat from unused codecs, you can use the small version of the +library: `library/cpp/blockcodecs/core`. In that case, you need to manually set `PEERDIR()`s to +needed codecs (i.e. `PEERDIR(library/cpp/blockcodecs/codecs/lzma)`). |