aboutsummaryrefslogtreecommitdiff
path: root/doc/forum/Is_it_possible_to_make_git-sync_not_nullify_symlinks__63__.mdwn
blob: 9c41acb707762471530e7b4b55663ad314d36647 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Hey,

I've found that git annex works great as a way to publish websites to a web server.  I can edit my website on my computer, `git annex sync` my working directory to the VPS, and then `git annex get files-I-want-to-publish`.  This works great.  I can maintain my normal working directory structure on the VPS and I don't have to worry about people seeing files I DIDN'T want to publish, since the dead symlinks just show up as 404s.

There's one small problem.  

Say:

 1) I've already published a file using `git annex get file-to-update`

 2) I update that file on my computer

 3) I do `git annex sync`

 4) I do `git annex get file-to-update`

Between steps 3 and 4, file-to-update goes from being an accessible web resource to being a dead symlink.  It's not really a problem for me, as hardly anyone visits my site. But it would be nice if I could make `sync` leave the old symlink to the old file until I `get`ed the new one.

Is this possible?

PS: For those who might follow in my footsteps, remember that you probably don't want people reading the contents of your .git dir, so make a re-write rule for this!

Timothy