git-annex.git
5 years agoAdded a comment: checkout borg remote
andrew [Sat, 9 Jan 2021 15:25:05 +0000 (15:25 +0000)]
Added a comment: checkout borg remote

5 years agoreport on fresh test fails
yarikoptic [Fri, 8 Jan 2021 15:37:46 +0000 (15:37 +0000)]
report on fresh test fails

5 years agomerged fix from kyle
Joey Hess [Thu, 7 Jan 2021 17:47:29 +0000 (13:47 -0400)]
merged fix from kyle

5 years agoavoid using wildcard near bug kyle fixed
Joey Hess [Thu, 7 Jan 2021 17:44:23 +0000 (13:44 -0400)]
avoid using wildcard near bug kyle fixed

5 years agoadjustTree: Consider submodule deletions
Kyle Meyer [Thu, 7 Jan 2021 00:16:30 +0000 (19:16 -0500)]
adjustTree: Consider submodule deletions

In addition to regular file deletions, the removefiles argument passed
to adjustTree may contain removed submodules.  When making the new
tree, filter these out in the same way that is done for regular files
so that the deletion is propagated.

5 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Thu, 7 Jan 2021 17:38:46 +0000 (13:38 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

5 years agoBehavior change: --trust-glacier option no longer overrides trust
Joey Hess [Thu, 7 Jan 2021 14:37:43 +0000 (10:37 -0400)]
Behavior change: --trust-glacier option no longer overrides trust

Since that can lead to data loss, which should never be enabled by an
option other than --force.

This commit was sponsored by Jake Vosloo on Patreon.

5 years agoBehavior change: --trust option no longer overrides trust
Joey Hess [Thu, 7 Jan 2021 14:12:37 +0000 (10:12 -0400)]
Behavior change: --trust option no longer overrides trust

Since that can lead to data loss, which should never be enabled by an
option other than --force.

I suppose that using --trust was in some situation, safer than --force,
because it doesn't entirely disable checking for data loss, but only
disables checking involving data that is on the specified repository.
But it seems better to be able to say that data loss only happens with
--force.

This commit was sponsored by Graham Spencer on Patreon.

5 years agoBehavior change: git-annex trust now needs --force
Joey Hess [Thu, 7 Jan 2021 13:59:52 +0000 (09:59 -0400)]
Behavior change: git-annex trust now needs --force

Since unconsidered use of trusted repositories can lead to data loss.

Trusted has always been this way, but it used to be acceptable for
git-annex to be set up so that data could be lost without using --force,
and most or all other ways that can happen have already been eliminated.

This commit was sponsored by Mark Reidenbach on Patreon.

5 years ago(no commit message)
kyle [Thu, 7 Jan 2021 00:25:38 +0000 (00:25 +0000)]

5 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Wed, 6 Jan 2021 18:27:03 +0000 (14:27 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

5 years agowording
Joey Hess [Wed, 6 Jan 2021 18:26:39 +0000 (14:26 -0400)]
wording

5 years agoimprove wording
Joey Hess [Wed, 6 Jan 2021 18:24:17 +0000 (14:24 -0400)]
improve wording

5 years agoMerge branch 'master' into requirednumcopies
Joey Hess [Wed, 6 Jan 2021 18:24:09 +0000 (14:24 -0400)]
Merge branch 'master' into requirednumcopies

5 years agomincopies
Joey Hess [Wed, 6 Jan 2021 18:11:08 +0000 (14:11 -0400)]
mincopies

This is conceptually very simple, just making a 1 that was hard coded be
exposed as a config option. The hard part was plumbing all that, and
dealing with complexities like reading it from git attributes at the
same time that numcopies is read.

Behavior change: When numcopies is set to 0, git-annex used to drop
content without requiring any copies. Now to get that (highly unsafe)
behavior, mincopies also needs to be set to 0. It seemed better to
remove that edge case, than complicate mincopies by ignoring it when
numcopies is 0.

This commit was sponsored by Denis Dzyubenko on Patreon.

5 years ago(no commit message)
AlbertZeyer [Wed, 6 Jan 2021 15:24:28 +0000 (15:24 +0000)]

5 years agoAdded a comment
AlbertZeyer [Wed, 6 Jan 2021 14:59:01 +0000 (14:59 +0000)]
Added a comment

5 years agoAdded a comment
AlbertZeyer [Wed, 6 Jan 2021 11:15:43 +0000 (11:15 +0000)]
Added a comment

5 years agobranch
Joey Hess [Tue, 5 Jan 2021 18:28:54 +0000 (14:28 -0400)]
branch

5 years agodocs for requirednumcopies
Joey Hess [Tue, 5 Jan 2021 18:22:27 +0000 (14:22 -0400)]
docs for requirednumcopies

Not implemented yet.

5 years agothe author of this forum post deleted it, so remove comments
Joey Hess [Tue, 5 Jan 2021 15:23:31 +0000 (11:23 -0400)]
the author of this forum post deleted it, so remove comments

5 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Tue, 5 Jan 2021 15:23:02 +0000 (11:23 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

5 years agoupdate
Joey Hess [Tue, 5 Jan 2021 13:15:28 +0000 (09:15 -0400)]
update

5 years agoremoved
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Tue, 5 Jan 2021 06:06:35 +0000 (06:06 +0000)]
removed

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Tue, 5 Jan 2021 05:55:14 +0000 (05:55 +0000)]
Added a comment

5 years agopromote forum post to bug report
Joey Hess [Mon, 4 Jan 2021 21:01:51 +0000 (17:01 -0400)]
promote forum post to bug report

5 years agocomment
Joey Hess [Mon, 4 Jan 2021 20:48:28 +0000 (16:48 -0400)]
comment

5 years agowrong, wrong, wrong
Joey Hess [Mon, 4 Jan 2021 20:28:36 +0000 (16:28 -0400)]
wrong, wrong, wrong

5 years agofix --time-limit
Joey Hess [Mon, 4 Jan 2021 19:25:28 +0000 (15:25 -0400)]
fix --time-limit

It got broken in several ways by the streaming seeking optimisations
around version 8.20201007.

Moved time limit checking out of the matcher, which was a hack in the
first place. So everywhere that uses Limit.getMatcher needs to check
time limit. Well, almost everywhere. Command.Info uses it, but it does
not make sense to time limit getting info. And Command.MultiCast uses it
just to build up a list of files that then get passed to a command, so
it would never have hit the timeout in a useful way.

This implementation is a little more expensive when at time limit than
necessary, since it continues seeking only to discard everything after the
time limit. I did try making it close the file handles to force a faster
shutdown, but that didn't work and hung. Could certianly be improved
somehow, but seeking is probably not the expensive bit when a time limit
is hit, so this seems acceptable for now.

5 years agocomment
Joey Hess [Mon, 4 Jan 2021 18:32:29 +0000 (14:32 -0400)]
comment

5 years agoclose
Joey Hess [Mon, 4 Jan 2021 17:46:11 +0000 (13:46 -0400)]
close

5 years agocomment
Joey Hess [Mon, 4 Jan 2021 17:43:01 +0000 (13:43 -0400)]
comment

5 years agoadd pointer to annex.largefiles config docs
Joey Hess [Mon, 4 Jan 2021 17:39:48 +0000 (13:39 -0400)]
add pointer to annex.largefiles config docs

5 years agocomment
Joey Hess [Mon, 4 Jan 2021 17:34:32 +0000 (13:34 -0400)]
comment

5 years agoadd: Significantly speed up adding lots of non-large files to git
Joey Hess [Mon, 4 Jan 2021 17:12:28 +0000 (13:12 -0400)]
add: Significantly speed up adding lots of non-large files to git

* add: Significantly speed up adding lots of non-large files to git,
  by disabling the annex smudge filter when running git add.
* add --force-small: Run git add rather than updating the index itself,
  so any other smudge filters than the annex one that may be enabled will
  be used.

5 years agofix bad annex.largefiles example syntax
Joey Hess [Mon, 4 Jan 2021 17:11:05 +0000 (13:11 -0400)]
fix bad annex.largefiles example syntax

5 years agoGit.Queue: allow providing git common options like -c
Joey Hess [Mon, 4 Jan 2021 16:51:55 +0000 (12:51 -0400)]
Git.Queue: allow providing git common options like -c

5 years agoupdate year
Joey Hess [Mon, 4 Jan 2021 16:45:19 +0000 (12:45 -0400)]
update year

5 years agoavoid combining queued commands with different params
Joey Hess [Mon, 4 Jan 2021 16:38:43 +0000 (12:38 -0400)]
avoid combining queued commands with different params

I don't think this affected git-annex currently, but if the same command
was queued twice with different params, one set of params was thrown
away, and the files going with those were run with the other set of
params.

5 years agocomment and todo
Joey Hess [Mon, 4 Jan 2021 16:26:48 +0000 (12:26 -0400)]
comment and todo

5 years agocomment
Joey Hess [Mon, 4 Jan 2021 16:07:01 +0000 (12:07 -0400)]
comment

5 years agocomment
Joey Hess [Mon, 4 Jan 2021 16:05:32 +0000 (12:05 -0400)]
comment

5 years agocomment
Joey Hess [Mon, 4 Jan 2021 16:04:01 +0000 (12:04 -0400)]
comment

5 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Mon, 4 Jan 2021 15:57:29 +0000 (11:57 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

5 years agoAdded a comment
AlbertZeyer [Mon, 4 Jan 2021 12:04:04 +0000 (12:04 +0000)]
Added a comment

5 years agoAdded a comment
AlbertZeyer [Sun, 3 Jan 2021 22:07:02 +0000 (22:07 +0000)]
Added a comment

5 years ago(no commit message)
AlbertZeyer [Sun, 3 Jan 2021 22:02:02 +0000 (22:02 +0000)]

5 years agoAdded a comment
AlbertZeyer [Sun, 3 Jan 2021 21:48:51 +0000 (21:48 +0000)]
Added a comment

5 years agoAdded a comment
AlbertZeyer [Sat, 2 Jan 2021 23:49:14 +0000 (23:49 +0000)]
Added a comment

5 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Sat, 2 Jan 2021 23:33:55 +0000 (19:33 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

5 years agoAdded a comment
AlbertZeyer [Sat, 2 Jan 2021 16:23:05 +0000 (16:23 +0000)]
Added a comment

5 years ago(no commit message)
Lukey [Sat, 2 Jan 2021 15:28:02 +0000 (15:28 +0000)]

5 years agoAdded a comment
Lukey [Sat, 2 Jan 2021 15:22:18 +0000 (15:22 +0000)]
Added a comment

5 years agoAdded a comment
Lukey [Sat, 2 Jan 2021 15:16:04 +0000 (15:16 +0000)]
Added a comment

5 years agoAdded a comment
Lukey [Sat, 2 Jan 2021 15:12:49 +0000 (15:12 +0000)]
Added a comment

5 years agoAdded a comment
Lukey [Sat, 2 Jan 2021 15:05:01 +0000 (15:05 +0000)]
Added a comment

5 years agoAdded a comment
AlbertZeyer [Fri, 1 Jan 2021 22:30:34 +0000 (22:30 +0000)]
Added a comment

5 years agoAdded a comment: Difference to import/add?
AlbertZeyer [Fri, 1 Jan 2021 21:49:29 +0000 (21:49 +0000)]
Added a comment: Difference to import/add?

5 years agoAdded a comment: rename or move files
AlbertZeyer [Fri, 1 Jan 2021 21:33:41 +0000 (21:33 +0000)]
Added a comment: rename or move files

5 years agoAdded a comment: Adding external files
AlbertZeyer [Fri, 1 Jan 2021 21:30:38 +0000 (21:30 +0000)]
Added a comment: Adding external files

5 years agoAdded a comment: annex.largefiles
AlbertZeyer [Fri, 1 Jan 2021 21:25:44 +0000 (21:25 +0000)]
Added a comment: annex.largefiles

5 years ago(no commit message)
Lukey [Fri, 1 Jan 2021 15:43:43 +0000 (15:43 +0000)]

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Fri, 1 Jan 2021 04:08:44 +0000 (04:08 +0000)]
Added a comment

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 22:45:03 +0000 (22:45 +0000)]
Added a comment

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 22:05:19 +0000 (22:05 +0000)]
Added a comment

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 22:00:08 +0000 (22:00 +0000)]
Added a comment

5 years agoAdded a comment
Lukey [Thu, 31 Dec 2020 21:46:33 +0000 (21:46 +0000)]
Added a comment

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 20:51:12 +0000 (20:51 +0000)]
Added a comment

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 20:34:55 +0000 (20:34 +0000)]
Added a comment

5 years agoremoved
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 19:41:18 +0000 (19:41 +0000)]
removed

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 19:40:22 +0000 (19:40 +0000)]
Added a comment

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 19:40:01 +0000 (19:40 +0000)]
Added a comment

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 19:34:56 +0000 (19:34 +0000)]
Added a comment

5 years agoAdded a comment
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 18:55:30 +0000 (18:55 +0000)]
Added a comment

5 years ago(no commit message)
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 18:51:50 +0000 (18:51 +0000)]

5 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Thu, 31 Dec 2020 17:46:25 +0000 (13:46 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

5 years agoAdded a comment
Lukey [Thu, 31 Dec 2020 17:32:47 +0000 (17:32 +0000)]
Added a comment

5 years ago(no commit message)
eric.w@eee65cd362d995ced72640c7cfae388ae93a4234 [Thu, 31 Dec 2020 16:58:50 +0000 (16:58 +0000)]

5 years agoclose import_tree todo
Joey Hess [Wed, 30 Dec 2020 17:40:49 +0000 (13:40 -0400)]
close import_tree todo

Split out two todos for things that were mentioned as still open items
in there. Most of the others were already dealt with. I didn't open a
new todo for the import from readonly S3 bucket because I guess if
someone needs that, they can ask for it.

5 years agodocument what importKey returning Nothing does
Joey Hess [Wed, 30 Dec 2020 17:21:40 +0000 (13:21 -0400)]
document what importKey returning Nothing does

That was added for thirdpartypopulated remotes, but for others it also
has the effect of skipping including the file in the imported tree.

5 years agoAdded a comment
Lukey [Tue, 29 Dec 2020 23:30:47 +0000 (23:30 +0000)]
Added a comment

5 years agoadd news item for git-annex 8.20201129
Joey Hess [Tue, 29 Dec 2020 17:52:34 +0000 (13:52 -0400)]
add news item for git-annex 8.20201129

5 years agoreleasing package git-annex version 8.20201129
Joey Hess [Tue, 29 Dec 2020 17:51:40 +0000 (13:51 -0400)]
releasing package git-annex version 8.20201129

5 years agoupdate to lts-16.27
Joey Hess [Tue, 29 Dec 2020 17:51:08 +0000 (13:51 -0400)]
update to lts-16.27

5 years agoupdate
Joey Hess [Tue, 29 Dec 2020 16:52:55 +0000 (12:52 -0400)]
update

5 years agoclose
Joey Hess [Tue, 29 Dec 2020 16:48:21 +0000 (12:48 -0400)]
close

5 years agoimprove tip
Joey Hess [Mon, 28 Dec 2020 23:38:12 +0000 (19:38 -0400)]
improve tip

5 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Mon, 28 Dec 2020 21:05:00 +0000 (17:05 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

5 years agoupdate
Joey Hess [Mon, 28 Dec 2020 21:04:41 +0000 (17:04 -0400)]
update

5 years agotip for probably best use case for borg with git-annex
Joey Hess [Mon, 28 Dec 2020 20:55:04 +0000 (16:55 -0400)]
tip for probably best use case for borg with git-annex

5 years agodon't support dropping from thirdPartyPopulated for now
Joey Hess [Mon, 28 Dec 2020 20:48:38 +0000 (16:48 -0400)]
don't support dropping from thirdPartyPopulated for now

This code I'm reverting works. But it has a problem: The export db and
log and the ContentIdentifier db and log still list the content as being
stored in the remote. So when I ran borg create again and stored the
content in borg again in a new archive, git-annex sync noticed that, but
since it didn't update the tree for the old archives, it then thought
the content that had been removed from them was still in them, and so
git-annex get failed in an ugly way:

Include pattern 'tmp/x/.git/annex/objects/pX/ZJ/SHA256E-s0--e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855/SHA256E-s0--e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855' never matched.
[2020-12-28 16:40:44.878952393] process [933616] done ExitFailure 1

  user error (borg ["extract","/tmp/b::abs4","tmp/x/.git/annex/objects/pX/ZJ/SHA256E-s0--e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855/SHA256E-s0--e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"] exited 1)

It does not seem worth it to update the git tree for the export when dropping
content, that would make drop of many files very expensive in git tree objects
created. So, let's not support this I suppose..

5 years agosupport removing objects from borg
Joey Hess [Mon, 28 Dec 2020 20:36:52 +0000 (16:36 -0400)]
support removing objects from borg

5 years agosupport removal from thirdPartyPopulated
Joey Hess [Mon, 28 Dec 2020 20:36:26 +0000 (16:36 -0400)]
support removal from thirdPartyPopulated

also some other fixes to thirdPartyPopulated

5 years agoborg appendonly config
Joey Hess [Mon, 28 Dec 2020 20:23:38 +0000 (16:23 -0400)]
borg appendonly config

5 years agorevert recent lockContent change
Joey Hess [Mon, 28 Dec 2020 20:05:14 +0000 (16:05 -0400)]
revert recent lockContent change

lockContent should only be done when it's versioned

5 years agoBug report: annex.adjustedbranchrefresh ignored
gb@4a49bb1afcf3d183bba8f07297b0395808768c6c [Mon, 28 Dec 2020 20:03:34 +0000 (20:03 +0000)]
Bug report: annex.adjustedbranchrefresh ignored

5 years agomove untrust forcing from Logs.Trust into Remote
Joey Hess [Mon, 28 Dec 2020 19:08:53 +0000 (15:08 -0400)]
move untrust forcing from Logs.Trust into Remote

No behavior changes here, but this is groundwork for letting remotes
such as borg vary untrust forcing depending on configuration.

5 years agosimplify
Joey Hess [Mon, 28 Dec 2020 18:57:23 +0000 (14:57 -0400)]
simplify

adjustExportImport' is never called with both isexport and isimport False.

5 years agosplit off versionedExport from appendonly
Joey Hess [Mon, 28 Dec 2020 18:37:15 +0000 (14:37 -0400)]
split off versionedExport from appendonly

S3 uses versionedExport, while GitLFS uses appendonly.

This is groundwork for later changes.

5 years ago(no commit message)
AlbertZeyer [Mon, 28 Dec 2020 16:31:09 +0000 (16:31 +0000)]