The day was 21/12/11 18:21 when , Matthew Monaco had this to say......:
On 12/21/2011 02:02 AM, Dan Vrátil wrote:
On Wed, 21 Dec 2011 03:37:53 +0000, Simon Stoakley <sausageandeggs@archlinux.us> wrote:
Hi All, I'm uploading a pkg for an mpris2 plugin for Kupfer. I don't want to clone the whole git repo (seems wasteful and I think half the code there is dead) and the only part that I'm interested in is the mpris2 folder. The only way I've found of checking out part of a repo in is to use the sparse checkout trick. The question I have is whether or not it's OK to use this in a PKGBUILD or not, what say ye wise men?! PKGBUILD below for reference (depends on pkgs I haven't uploaded yet if you try building)
Thanks Simon
Hi,
I don't think git allows you to clone just part of a repo (unlike SVN), but what I usually do is to reduce the clone depth: git clone git://url.com/repo --depth 1
The --depth switch will make git to clone only the most recent history, because when compiling, you are hardly interested in previous commits. It usually makes the clone operation significantly faster and saves a lot of bandwidth.
Cheers, Dan
If the PKGBUILD is supposed to be for some specific release of this package, and it's only available through git, then you need to find a mirror that allows for git archive (even if they only allow tarring with no compression). Because with git clone --depth you can only get the latest. However, if it's a traditional -git package then --depth should work fine.
I think I'm just going to grab the code , put it in my own git repo and grab it from there, can't find any sort of license though so I'm going to email the author. Does anyone know what the usual procedure is if the code is unlicensed? --