User:Tancre/2/hackpact/Git Stalker
< User:Tancre | 2 | hackpact
Git Stalker
From web stalker
I imagine a sort of 'git stalker' (it could be an online service).
- Crawler > ask the link of the git repository and start the session through the Github's API (+ further data on the connection, maybe).
- Overview (Map) > tree -hJ >> overview.json print the data structure in a file. D3.js use the json to produce a visualization of the structure of the directory. Kinda like the map of the links in 'web stalker': central main folder and then the other folders/files distributed around the main folder and so on recursively. (It is relevant the concept of Mind Map and Rhizome)
Github's API
Crawler
Access repository.
- API
<body>
<h1>Git Stalker _ API</h1>
<button id="btnRepos">Repos</button>
<div id="divResult"></div>
<script type="text/javascript">
const btnRepos = document.getElementById('btnRepos')
btnRepos.addEventListener("click", getRepos)
async function getRepos(){
const url = "http://api.github.com/search/repositories?q=stars:>100000"
const response = await fetch(url)
const result = await response.json()
console.log(result)
}
</script>
</body>
- Git clone (request link)
#!/bin/bash git clone 'link' tree -hJ >> overview.json
Some note
It is interesting to find the web stalker on Rhizome and think about this concept at the same time.
Evolution of the data viz: linear > tree > rhizome
D&G - 1. Introduction: Rhizome (from A Thousand Plateaux)
Pasquinelli - The arborescent mind