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

This commit is contained in:
Joey Hess 2012-09-20 17:40:08 -04:00
commit 7456e9f68c
2 changed files with 33 additions and 0 deletions

View file

@ -0,0 +1,7 @@
[[!comment format=mdwn
username="http://joeyh.name/"
subject="comment 13"
date="2012-09-20T16:21:11Z"
content="""
@Jörn, two days ago I added support in the webapp for multiple independent repositories. So you can create independent repos using it, and switch between them from a menu. Under the hood there are multiple git-annex assistant daemons running, but this is an implementation detail; it's not like these use any more memory or other resources than would a single daemon that managed multiple repositories.
"""]]

View file

@ -0,0 +1,26 @@
[[!comment format=mdwn
username="https://www.google.com/accounts/o8/id?id=AItOawlc1og3PqIGudOMkFNrCCNg66vB7s-jLpc"
nickname="Paul"
subject="can addurl use hashing once the file is downloaded?"
date="2012-09-20T21:01:30Z"
content="""
There are resources that I want to add to my annex that are currently available
via a URL, but it seems like if I add these using `git-annex addurl`, they get
symlinked to file in the annex/objects directory that starts with `URL-...`,
instead of the more typical `SHA256-...`, and this does not change even after
the files are downloaded.
My concern is that I really want to ensure that these files don't change, which
is the appeal of content-addressable symlinking of normal files (as opposed to
URL addressable ones).
Would there be a way to automate the injection of hash-based symlinking for
files that are added via addurl? Sometimes I add a bunch of files via ``addurl
--fast``, and after I've download them via ``get``, it would be nice to have
those files have the same level of data integrity as when I download them using
something outside of git-annex, add them to the annex, and do an ``addurl
--file`` afterward.
Thanks for all of your hard work!
"""]]