<div class="gmail_quote">2009/5/23 Chani <span dir="ltr"><<a href="mailto:chanika@gmail.com">chanika@gmail.com</a>></span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class="im">the only thing that might make it slow or bad for hte (git)</div>
server is if git makes it really hard to take only the doc folder from the<br>
repo.</blockquote><div><br></div><div>It is not possible to do a partial clone of a repository since Git doesn't work that way. Now, Git does have very good compression and if most of the files there seldom changes, the bandwidth usage shouldn't really be a problem.</div>
<div><br></div><div>If you really need to clone a specific directory, you have two solutions:</div><div>Â Â Â 1. Make the directory a separate repository and have it as a submodule in the parent repository. Cons: Complex and not really easy to use.</div>
<div>Â Â Â 2. Don't use git for that directory; just fetch a tar.gz via HTTP. Cons: Should be obvious. Not really a solution.</div><div><br></div><div>/me goes back to lurking.</div></div><br>-- <br>Magnus Bergmark - magnus DOT bergmark AT gmail DOT com<br>
GPG/PGP: 0x7BE84794DB6AA648<br>Fingerprint: 0E6F D2DB F0EF 534A 2184 52AF 7BE8 4794 DB6A A648<br>