Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/awetzel/gitex
Elixir implementation of the Git object storage, but with the goal to implement the same semantic with other storage and topics
https://github.com/awetzel/gitex
Last synced: 3 months ago
JSON representation
Elixir implementation of the Git object storage, but with the goal to implement the same semantic with other storage and topics
- Host: GitHub
- URL: https://github.com/awetzel/gitex
- Owner: kbrw
- Created: 2015-03-23T01:33:47.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2021-10-29T15:34:04.000Z (about 3 years ago)
- Last Synced: 2024-04-14T04:10:26.383Z (7 months ago)
- Language: Elixir
- Size: 39.1 KB
- Stars: 66
- Watchers: 5
- Forks: 5
- Open Issues: 5
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome-elixir - gitex - Elixir implementation of the Git object storage, but with the goal to implement the same semantic with other storage and topics. (Version Control)
- fucking-awesome-elixir - gitex - Elixir implementation of the Git object storage, but with the goal to implement the same semantic with other storage and topics. (Version Control)
- freaking_awesome_elixir - Elixir - Elixir implementation of the Git object storage, but with the goal to implement the same semantic with other storage and topics. (Version Control)
README
Gitex [](https://travis-ci.org/kbrw/gitex)
=======
See API documentation at [http://hexdocs.pm/gitex](http://hexdocs.pm/gitex).
- Reference implementation in pure Elixir of the Git object model and storage,
including optimized pack-refs and pack-objects/deltas).
- Protocol over Git codec and backend to customize them and reuse the same
versioning logic in a completely different environment and use case: JSON
into Riak for instance.
TODO:
- test it (only for regression, currently it works on many open source git repo, so it can be considered as tested)
- add a `Gitex.merge` helper to help you construct a commit tree from multiple trees
- add impl `Gitex.Repo` for Pid as a GenServer RPC
- implementation example of previous GenServer maintaining ETS LRU cache of standard git fs objects and deltas
- add some useful alternative implementations, currently only standard object encoding and storage
## Usage example
```elixir
r = Gitex.Git.open #Gitex.Git is the .git fs object storage
Gitex.get("master",r) #get commit
Gitex.get("myannotatedtag",r) #get tag object
Gitex.get("master",r,"/path/to/dir") #get tree object
Gitex.get("master",r,"/path/to/file") #get blob
# get all commits from master to 1st January 2015
Gitex.history("master",r)
|> Enum.take_while(& &1.committer.utc_time > {{2015,1,1},{0,0,0}})
# get the stream of version history of a given file
Gitex.history("master",r)
|> Stream.map(&Gitex.get_hash(&1,r,"/path/to/file"))
|> Stream.dedup
|> Stream.map(&Gitex.object(&1,r))
# commit history stream is powerful, play with it
Gitex.get("master",r) # return commit
|> Gitex.put(r,"/some/path/file1","file1 content") #put new trees and return new root tree hash
|> Gitex.put(r,"/some/other/path/file2","file2 content") ##put new trees and return new root tree hash
|> Gitex.commit(r,"master","some commit message") #save tree in a commit with "master" parent then update "master" and return commit hash
|> Gitex.tag(r,"mytag") # save this commit to a soft tag return commit_tag
|> Gitex.tag(r,"myannotatedtag","my message") # save this commit to a tag object with comment, return tag hash
# Currently "put" is the only helper to construct a new "tree", for merging you have to construct the tree yourself
```
A nice function `Gitex.align_history` allows you to lazily add an index number to your
history stream in order to construct a pretty visualizer very easily (d3.js for instance)
```elixir
Gitex.history(:head,repo) |> Gitex.align_history
```
## The Gitex.Repo protocol
Any repo implementing the `Gitex.Repo` protocol : (basically object codec, ref
setter/resolver, binary get/put) can be managed with the `Gitex` API.
# CONTRIBUTING
Hi, and thank you for wanting to contribute.
Please refer to the centralized informations available at: https://github.com/kbrw#contributing