Merge branch 'master' of ssh://git-annex.branchable.com

This commit is contained in:
Joey Hess 2017-10-11 11:13:43 -04:00
commit adcc73d91a
No known key found for this signature in database
GPG key ID: DB12DB0FF05F8F38
4 changed files with 73 additions and 0 deletions

View file

@ -0,0 +1,22 @@
[[!comment format=mdwn
username="yarikoptic"
avatar="http://cdn.libravatar.org/avatar/f11e9c84cb18d26a1748c33b48c924b4"
subject="comment 3"
date="2017-10-10T17:40:08Z"
content="""
Found a following comment in the code
[[!format haskell \"\"\"
{- Normally, blocks writing to an annexed file, and modifies file
- permissions to allow reading it.
-
- When core.sharedRepository is set, the write bits are not removed from
- the file, but instead the appropriate group write bits are set. This is
- necessary to let other users in the group lock the file. But, in a
- shared repository, the current user may not be able to change a file
- owned by another user, so failure to set this mode is ignored.
-}
\"\"\"]]
So may be it is a \"Feature\" although killing the whole premise of data safety while using git-annex.
In my case, shared permissions are primarily to make files/repositories readable by others, so may be I should have not used 'shared' mode anyways, since reading does not need the shared setting
"""]]

View file

@ -0,0 +1,8 @@
[[!comment format=mdwn
username="yarikoptic"
avatar="http://cdn.libravatar.org/avatar/f11e9c84cb18d26a1748c33b48c924b4"
subject="comment 4"
date="2017-10-10T17:56:47Z"
content="""
our comments crossed in the hyperspace... ;) yeah -- I would have expected a separate lock file to be used for such cases. Now data loss is really a possibility (almost made it happen since opened the file and wrote it without unlocking, good that I caught it and was able to modify back exactly the way it was). Interactions among multiple versions of annex on the same repo is imho a lesser possibility (would require two different versions of annex being available)
"""]]