You can wget
your way out of the GitHub repo to get a tar file (archive):
wget --no-check-certificate https://github.com/User/repo/archive/master.tar.gz
# better, if the certificate authorities are present:
wget https://github.com/User/repo/archive/master.tar.gz
will get you a file named 'master' from the user 'User''s repo 'repo'.
The updated V3 API url is:
https://api.github.com/repos/User/repo/:archive_format/:ref
#
# two possibilities for fomat:
https://api.github.com/repos/User/repo/tarball/master
https://api.github.com/repos/User/repo/zipball/master
# from github example:
$curl -L https://api.github.com/repos/octokit/octokit.rb/tarball > octokit.tar.gz
You can then tar xpvf master
, getting the full archive. It will create a directory following the naming convention described in the question you mentioned.
No git binary is needed to get an archive from GitHub, thanks to their download service "Nodeload".
ligemer
proposed in an edit the following example:
Edit 2016-08-25 - Shell Example With Wget, Variables, and Untar:
#!/bin/bash -ex
# arguments:
# token = $1
# organization = $2
# repo name = $3
# branch = $4
wget --header="Authorization: token ${1}" --header="Accept:application/vnd.github.v3.raw" -O - https://api.github.com/repos/${2}/${3}/tarball/${4} | tar xz
Call via:
$ scriptName.sh token my-organization site.com master
The command above will download and extract the Github folder to the same directory as the script.
Diogo Quintela suggests in the comments:
The following example allow the download, extract and cut the top level directory
curl -L https://api.github.com/repos/octokit/octokit.rb/tarball | tar xz --strip=1
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…