git-annex.git
3 years agofinalise changelog for 10.20230126-2 upload
Sean Whitton [Fri, 24 Feb 2023 20:37:02 +0000 (13:37 -0700)]
finalise changelog for 10.20230126-2 upload

3 years agoAdd m68k to list of archs for webapp build-deps
Sean Whitton [Thu, 23 Feb 2023 18:06:22 +0000 (11:06 -0700)]
Add m68k to list of archs for webapp build-deps

3 years agofinalise changelog for 10.20230126-1 upload
Sean Whitton [Mon, 13 Feb 2023 18:07:16 +0000 (11:07 -0700)]
finalise changelog for 10.20230126-1 upload

3 years agoCommit Debian 3.0 (quilt) metadata
Sean Whitton [Mon, 13 Feb 2023 17:23:38 +0000 (10:23 -0700)]
Commit Debian 3.0 (quilt) metadata

[dgit (10.5) quilt-fixup]

3 years agochangelog
Sean Whitton [Mon, 13 Feb 2023 17:23:24 +0000 (10:23 -0700)]
changelog

3 years agoMerge tag '10.20230126'
Sean Whitton [Mon, 13 Feb 2023 17:16:46 +0000 (10:16 -0700)]
Merge tag '10.20230126'

tagging package git-annex version 10.20230126

3 years agoadd upper bounds on base version
Joey Hess [Thu, 26 Jan 2023 19:33:52 +0000 (15:33 -0400)]
add upper bounds on base version

hackage now rejects packages without this

3 years agoadd upper bounds on Cabal version
Joey Hess [Thu, 26 Jan 2023 19:31:29 +0000 (15:31 -0400)]
add upper bounds on Cabal version

hackage now rejects packages without this. My bet is any version of
cabal is going to work, I'm using the public API. Annoying.

3 years agoadd news item for git-annex 10.20230126
Joey Hess [Thu, 26 Jan 2023 19:27:44 +0000 (15:27 -0400)]
add news item for git-annex 10.20230126

3 years agoreleasing package git-annex version 10.20230126
Joey Hess [Thu, 26 Jan 2023 19:27:32 +0000 (15:27 -0400)]
releasing package git-annex version 10.20230126

3 years agonew linux autobuilder
Joey Hess [Thu, 26 Jan 2023 19:25:54 +0000 (15:25 -0400)]
new linux autobuilder

3 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Thu, 26 Jan 2023 18:54:02 +0000 (14:54 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

3 years agoremove libghc-pcre-light-dev build dep
Joey Hess [Thu, 26 Jan 2023 18:53:19 +0000 (14:53 -0400)]
remove libghc-pcre-light-dev build dep

Unused build dep.

Also updated some other deps based on debian's fork of this file.

3 years agoAdded a comment: thanks
aca [Wed, 25 Jan 2023 14:01:08 +0000 (14:01 +0000)]
Added a comment: thanks

3 years agoAdded a comment: Thank you!!
nobodyinperson [Tue, 24 Jan 2023 21:15:50 +0000 (21:15 +0000)]
Added a comment: Thank you!!

3 years agoimprove wording
Joey Hess [Tue, 24 Jan 2023 18:11:32 +0000 (14:11 -0400)]
improve wording

3 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Tue, 24 Jan 2023 18:08:40 +0000 (14:08 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

3 years agouse transfer stages for copy --from
Joey Hess [Tue, 24 Jan 2023 18:03:26 +0000 (14:03 -0400)]
use transfer stages for copy --from

See commit e04a9314392037e7b54a098f9709fc6040327411 for an explanation
of why move uses transfer stages for --from, but command stages for
--to. At the point of that commit, copy was actually already using
command stages for everything, so the commit was incorrect about
improving copy --to.

But, the same reasoning about --from applies to copy as to move; when
verification is not done incrementally, download and verification are
the main two stages. The cleanup stage for copy is even less work than
for move (it doesn't drop from the remote).

Sponsored-by: Dartmouth College's DANDI project
3 years agoimprove concurrency of move/copy --from --to
Joey Hess [Tue, 24 Jan 2023 17:45:01 +0000 (13:45 -0400)]
improve concurrency of move/copy --from --to

Use separate stages for download and upload. In the common case where
it downloads the file from one remote and then uploads to the other,
those are by far the most expensive operations, and there's a decent
chance the two remotes bottleneck on different resources.

Suppose it's being run with -J2 and a bunch of 10 mb files. Two threads
will be started both downloading from the src remote. They will probably
finish at the same time. Then two threads will be started uploading to
the dst remote. They will probably take the same time as well. Before
this change, it would alternate back and forth, bottlenecking on src and dst.
With this change, as soon as the two threads start uploading to dst, two
more threads are able to start, downloading from src. So bandwidth to
both remotes is saturated more often.

Other commands that use transferStages only send in one direction at a
time. So the worker threads for the other direction will sit idle, and
there will be no change in their behavior.

Sponsored-by: Dartmouth College's DANDI project
3 years agoAdded a comment
jpds [Tue, 24 Jan 2023 15:55:36 +0000 (15:55 +0000)]
Added a comment

3 years agoAdded a comment
nobodyinperson [Mon, 23 Jan 2023 22:55:11 +0000 (22:55 +0000)]
Added a comment

3 years agoAdded a comment
nobodyinperson [Mon, 23 Jan 2023 22:38:28 +0000 (22:38 +0000)]
Added a comment

3 years agoupdate
Joey Hess [Mon, 23 Jan 2023 22:08:55 +0000 (18:08 -0400)]
update

3 years agoMerge branch 'fromto'
Joey Hess [Mon, 23 Jan 2023 22:01:57 +0000 (18:01 -0400)]
Merge branch 'fromto'

3 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Mon, 23 Jan 2023 22:01:12 +0000 (18:01 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

3 years agodone
Joey Hess [Mon, 23 Jan 2023 21:57:15 +0000 (17:57 -0400)]
done

3 years agofix behavior of copy --from --to
Joey Hess [Mon, 23 Jan 2023 21:53:57 +0000 (17:53 -0400)]
fix behavior of copy --from --to

Sponsored-by: Dartmouth College's DANDI project
3 years agoMerge branch 'master' into fromto
Joey Hess [Mon, 23 Jan 2023 21:44:44 +0000 (17:44 -0400)]
Merge branch 'master' into fromto

3 years agofinishing up move --from --to
Joey Hess [Mon, 23 Jan 2023 21:07:21 +0000 (17:07 -0400)]
finishing up move --from --to

Lock the local content for drop after getting it from src, to prevent another
process from using the local content as a copy and dropping it from src,
which would prevent dropping the local content after sending it to dest.

Support resuming an interrupted move that downloaded the content from
src, leaving the local content populated. In this case, the location log
has not been updated to say the content is present locally, so we can
assume that it's resuming and go ahead and drop the local content after
sending it to dest.

Note that if a `git-annex get` is being ran at the same time as a
`git-annex move --from --to`, it may get a file just before the move
processes it. So the location log has not been updated yet, and the move
thinks it's resuming. Resulting in local copy being dropped after it's
sent to the dest. This race is something we'll just have to live with,
it seems.

I also gave up on the idea of checking if the location log had been updated
by a `git-annex get` that is ran at the same time. That wouldn't work, because
the location log is precached in the seek stage, so reading it again after
sending the content to dest would not notice changes made to it, unless the cache
were invalidated, which would slow it down a lot. That idea anyway was subject
to races where it would not detect the concurrent `git-annex get`.

So concurrent `git-annex get` will have results that may be surprising.
To make that less surprising, updated the documentation of this feature to
be explicit that it downloads content to the local repository
temporarily.

Sponsored-by: Dartmouth College's DANDI project
3 years agofully working move --from --to (not release quality)
Joey Hess [Mon, 23 Jan 2023 19:36:06 +0000 (15:36 -0400)]
fully working move --from --to (not release quality)

When the destination already has a copy, it behaves the same as
drop --from really, but display it as a move and implement it
reusing the factored out code from fromPerform.

(Note that willDropMakeItWorse never returns DropAllowed in that
situation, because it's told that dest has a copy. So numcopies is
always checked.)

And when only the source and not the local repo or destination have a
copy, do the full copy from source to local, then copy from local to
dest, then drop from local, then drop from source dance.

This is complicated by fromPerform being hardcoded to assume there is a
local copy, but the local copy has already been dropped. That's why
it uses cleanupfromsrc RemoveNever to avoid the code that makes that
assumption, and finishes with a call to dropfromsrc.

And, since the location log has not yet been updated, checking numcopies
was not working, until I added UnVerifiedRemote dest to the list of
things to check.

This is not yet quite mergeable though. There are two things in the
comment above fromToPerform that are not implemented yet: Checking the
location log before dropping the local copy, and locking the temporary
local copy for drop.

Sponsored-by: Dartmouth College's DANDI project
3 years agopush location log updating up to callers of download
Joey Hess [Mon, 23 Jan 2023 17:45:26 +0000 (13:45 -0400)]
push location log updating up to callers of download

Prep for move --to --from, which needs to download from a src repo
without updating the location log for the local repo, before sending the
content on to the dest repo.

Note that caller of download' already update the log themselves.
See previous commit a422a056f2785999220cde55e07b0fbf999edd44
that pushed it up to download from getViaTmpFrom.

(Also removed in passing a debug print + readline that I accidentially
committed last week on this branch.)

Sponsored-by: Dartmouth College's DANDI project
3 years agoimplement move --from --to when there is a local copy already
Joey Hess [Mon, 23 Jan 2023 17:16:14 +0000 (13:16 -0400)]
implement move --from --to when there is a local copy already

This is rather trivial, since it does not need to temporarily get the
local copy.

Added fromPerform' to handle the situation where the local copy
is dropped by another process during the copy to the dest. This avoids
ever re-downloading the local copy before dropping from the src.

Sponsored-by: Dartmouth College's DANDI project
3 years agoupdate
Joey Hess [Mon, 23 Jan 2023 16:45:01 +0000 (12:45 -0400)]
update

3 years agoFix typo in heading that seems to result in git-annex-drop not showing up in `man...
meribold [Sun, 22 Jan 2023 15:25:35 +0000 (15:25 +0000)]
Fix typo in heading that seems to result in git-annex-drop not showing up in `man -k` output

3 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Fri, 20 Jan 2023 15:23:24 +0000 (11:23 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

3 years agocomment
Joey Hess [Fri, 20 Jan 2023 15:23:04 +0000 (11:23 -0400)]
comment

3 years agoremove --fast from man page
Joey Hess [Wed, 18 Jan 2023 19:15:41 +0000 (15:15 -0400)]
remove --fast from man page

git-annex move does not actually behave any differently with --fast than
without it. (git-annex copy does)

(cherry picked from commit f74904ee2c7cec936cac3f2536daa1d426739b80)

3 years agomove/copy: started implementing --from src --to dest
Joey Hess [Fri, 20 Jan 2023 15:10:38 +0000 (11:10 -0400)]
move/copy: started implementing --from src --to dest

This is not in a usable state, but I have a possible plan for how to do
it.

Sponsored-by: Dartmouth College's DANDI project
3 years ago(no commit message)
nobodyinperson [Fri, 20 Jan 2023 10:29:33 +0000 (10:29 +0000)]

3 years agoAdded a comment
jpds [Thu, 19 Jan 2023 16:28:19 +0000 (16:28 +0000)]
Added a comment

3 years agoAdded a comment
jpds [Thu, 19 Jan 2023 15:09:01 +0000 (15:09 +0000)]
Added a comment

3 years agoAdded a comment
jpds [Wed, 18 Jan 2023 22:57:58 +0000 (22:57 +0000)]
Added a comment

3 years ago(no commit message)
jpds [Wed, 18 Jan 2023 22:45:06 +0000 (22:45 +0000)]

3 years agoremove --fast from man page
Joey Hess [Wed, 18 Jan 2023 19:15:41 +0000 (15:15 -0400)]
remove --fast from man page

git-annex move does not actually behave any differently with --fast than
without it. (git-annex copy does)

3 years agomove/copy: option parsing for --from with --to
Joey Hess [Wed, 18 Jan 2023 18:42:39 +0000 (14:42 -0400)]
move/copy: option parsing for --from with --to

Allowing --from and --to as an alternative to --from or --to
is hard to do with optparse-applicative!

The obvious approach of (pfrom <|> pto <|> pfromandto) does not work
when pfromandto uses the same option names as pfrom and pto do.
It compiles but the generated parser does not work for all desired
combinations.

Instead, have to parse optionally from and optionally to. When neither
is provided, the parser succeeds, but it's a result that can't be
handled. So, have to giveup after option parsing. There does not seem to
be a way to make an optparse-applicative Parser give up internally
either.

Also, need seek' because I first tried making fto be a where binding,
but that resulted in a hang when git-annex move was run without --from
or --to. I think because startConcurrency was not expecting the stages
value to contain an exception and so ended up blocking.

Sponsored-by: Dartmouth College's DANDI project
3 years agoAdded a comment
yarikoptic [Wed, 18 Jan 2023 17:55:50 +0000 (17:55 +0000)]
Added a comment

3 years agocomment
Joey Hess [Wed, 18 Jan 2023 17:05:47 +0000 (13:05 -0400)]
comment

3 years agofindkeys: New command, very similar to git-annex find but operating on keys
Joey Hess [Tue, 17 Jan 2023 18:42:29 +0000 (14:42 -0400)]
findkeys: New command, very similar to git-annex find but operating on keys

I've long been asked for `git-annex find --all` or something like that,
but pushed back on it because I feel that the command is analagous to
find(1) and so it would be surprising for it to list keys rather than
files. So instead, add a new findkeys subcommand.

Note that the use of withKeyOptions is rather strange because usually
that is used to fall back to --all rather than listing files, but here
it's made to default to --all like behavior and never list files.

A performance thing that could be improved is that withKeyOptions
always reads and caches location logs. But findkeys with no options does
not need them, so it could be made faster. That caching does speed up
options like --in though. This is really just a subset of a more general
performance thing that --all reads location logs sometimes unncessarily.
Anyway, it needs to read the location log in order to checkDead,
and it seems good that findkeys does skip dead keys.

Also, cleaned up comments on git-annex-find man page asking for --all
option.

Sponsored-by: Dartmouth College's DANDI project
3 years agofix misleading helper function name
Joey Hess [Tue, 17 Jan 2023 18:25:38 +0000 (14:25 -0400)]
fix misleading helper function name

noworktreeitems was false for NoWorkTreeItems which was hard to understand.

Sponsored-by: Dartmouth College's DANDI project
3 years agocomment
Joey Hess [Tue, 17 Jan 2023 17:10:28 +0000 (13:10 -0400)]
comment

3 years agocomment
Joey Hess [Tue, 17 Jan 2023 17:08:49 +0000 (13:08 -0400)]
comment

3 years agoAdded a comment: comment 1 response
daven.quinn@d0ed4e0e5e4462d9a74a5d5a8fbd1b17f85db13e [Mon, 16 Jan 2023 21:45:35 +0000 (21:45 +0000)]
Added a comment: comment 1 response

3 years agoAdded a comment
yarikoptic [Mon, 16 Jan 2023 21:43:40 +0000 (21:43 +0000)]
Added a comment

3 years agoAdded a comment
yarikoptic [Mon, 16 Jan 2023 21:32:09 +0000 (21:32 +0000)]
Added a comment

3 years agoAdded a comment
yarikoptic [Mon, 16 Jan 2023 21:27:12 +0000 (21:27 +0000)]
Added a comment

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 20:12:16 +0000 (16:12 -0400)]
comment

3 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Mon, 16 Jan 2023 19:53:14 +0000 (15:53 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 19:52:52 +0000 (15:52 -0400)]
comment

3 years agoinitial report on difficulty moving frozen file
yarikoptic [Mon, 16 Jan 2023 19:36:07 +0000 (19:36 +0000)]
initial report on difficulty moving frozen file

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 19:26:21 +0000 (15:26 -0400)]
comment

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 19:15:21 +0000 (15:15 -0400)]
comment

3 years agoresponse
Joey Hess [Mon, 16 Jan 2023 19:13:00 +0000 (15:13 -0400)]
response

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 19:07:46 +0000 (15:07 -0400)]
comment

3 years agoclose wontfix with submitter agreement
Joey Hess [Mon, 16 Jan 2023 18:40:30 +0000 (14:40 -0400)]
close wontfix with submitter agreement

3 years agoclose as dup
Joey Hess [Mon, 16 Jan 2023 18:37:44 +0000 (14:37 -0400)]
close as dup

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 18:28:46 +0000 (14:28 -0400)]
comment

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 18:16:13 +0000 (14:16 -0400)]
comment

3 years agoclose wontfix
Joey Hess [Mon, 16 Jan 2023 18:08:37 +0000 (14:08 -0400)]
close wontfix

3 years agoclose
Joey Hess [Mon, 16 Jan 2023 18:02:10 +0000 (14:02 -0400)]
close

3 years agoclose
Joey Hess [Mon, 16 Jan 2023 17:57:54 +0000 (13:57 -0400)]
close

3 years agoclose wontfix
Joey Hess [Mon, 16 Jan 2023 17:53:54 +0000 (13:53 -0400)]
close wontfix

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 17:51:19 +0000 (13:51 -0400)]
comment

3 years agoclose this old todo
Joey Hess [Mon, 16 Jan 2023 17:45:24 +0000 (13:45 -0400)]
close this old todo

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 17:36:08 +0000 (13:36 -0400)]
comment

3 years agorespond and wontfix
Joey Hess [Mon, 16 Jan 2023 17:35:14 +0000 (13:35 -0400)]
respond and wontfix

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 17:16:02 +0000 (13:16 -0400)]
comment

3 years agocomment
Joey Hess [Mon, 16 Jan 2023 16:58:32 +0000 (12:58 -0400)]
comment

3 years agoAdded a comment
yarikoptic [Thu, 12 Jan 2023 20:48:48 +0000 (20:48 +0000)]
Added a comment

3 years agoadded an optional cost= configuration to all special remotes
Joey Hess [Thu, 12 Jan 2023 17:42:28 +0000 (13:42 -0400)]
added an optional cost= configuration to all special remotes

Note that when this is specified and an older git-annex is used to
enableremote such a special remote, it will simply ignore the cost= field
and use whatever the default cost is.

In passing, fixed adb to support the remote.name.cost and
remote.name.cost-command configs.

Sponsored-by: Dartmouth College's DANDI project
3 years agoremove a debug print and fix build
Joey Hess [Thu, 12 Jan 2023 17:18:25 +0000 (13:18 -0400)]
remove a debug print and fix build

3 years ago(no commit message)
nobodyinperson [Thu, 12 Jan 2023 10:10:37 +0000 (10:10 +0000)]

3 years agorespect urlinclude/urlexclude of other web special remotes
Joey Hess [Tue, 10 Jan 2023 18:58:53 +0000 (14:58 -0400)]
respect urlinclude/urlexclude of other web special remotes

When a web special remote does not have urlinclude/urlexclude
configured, make it respect the configuration of other web special
remotes and avoid using urls that match the config of another.

Note that the other web special remote does not have to be enabled.
That seems ok, it would have been extra work to check for only ones that
are enabled.

The implementation does mean that the web special remote re-parses
its own config once at startup, as well as re-parsing the configs of any
other web special remotes. This should be a very small slowdown
unless there are lots of web special remotes.

Sponsored-by: Dartmouth College's DANDI project
3 years agocomments
Joey Hess [Tue, 10 Jan 2023 15:52:11 +0000 (11:52 -0400)]
comments

3 years agoAdded a comment: Concern
yarikoptic [Tue, 10 Jan 2023 13:17:30 +0000 (13:17 +0000)]
Added a comment: Concern

3 years agoAdded a comment: {copy,move} with both --to and --from would be great!
nobodyinperson [Tue, 10 Jan 2023 09:38:39 +0000 (09:38 +0000)]
Added a comment: {copy,move} with both --to and --from would be great!

3 years agoAdded a comment
yarikoptic [Mon, 9 Jan 2023 22:04:11 +0000 (22:04 +0000)]
Added a comment

3 years agoinitial report on copy --from --to
yarikoptic [Mon, 9 Jan 2023 21:51:18 +0000 (21:51 +0000)]
initial report on copy --from --to

3 years agoAdded a comment: could be of help to DANDI
yarikoptic [Mon, 9 Jan 2023 21:36:44 +0000 (21:36 +0000)]
Added a comment: could be of help to DANDI

3 years agoweb: Add urlinclude and urlexclude configuration settings
Joey Hess [Mon, 9 Jan 2023 21:16:53 +0000 (17:16 -0400)]
web: Add urlinclude and urlexclude configuration settings

Sponsored-by: Dartmouth College's DANDI project
3 years agoweb special remote is no longer a singleton
Joey Hess [Mon, 9 Jan 2023 19:49:20 +0000 (15:49 -0400)]
web special remote is no longer a singleton

Allow initremote of additional special remotes with type=web, in addition
to the default web special remote.

When --sameas=web is used, these provide additional names for the web
special remote, and may also have their own additional configuration
(once there is any for the web special remote) and cost.

Sponsored-by: Dartmouth College's DANDI project
3 years agogeneral copyright year bump
Joey Hess [Mon, 9 Jan 2023 19:05:27 +0000 (15:05 -0400)]
general copyright year bump

3 years agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Sat, 7 Jan 2023 17:56:29 +0000 (13:56 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

3 years agoupdate
Joey Hess [Sat, 7 Jan 2023 17:56:23 +0000 (13:56 -0400)]
update

3 years agoupdate
Joey Hess [Sat, 7 Jan 2023 17:56:06 +0000 (13:56 -0400)]
update

3 years agoAdded a comment
sb-beryllium@6e2c477eac63b823bd315ef8aaf5f93173c1f15b [Sat, 7 Jan 2023 13:05:24 +0000 (13:05 +0000)]
Added a comment

3 years agoAdded a comment: use list order for cost
satra [Fri, 6 Jan 2023 19:36:07 +0000 (19:36 +0000)]
Added a comment: use list order for cost

3 years agoinitial todo on prioritization of URLs
yarikoptic [Fri, 6 Jan 2023 17:27:34 +0000 (17:27 +0000)]
initial todo on prioritization of URLs

3 years agoAdded a comment: importtree readonly remote folder?
yarikoptic [Fri, 6 Jan 2023 14:21:36 +0000 (14:21 +0000)]
Added a comment: importtree readonly remote folder?

3 years agoremove old closed bugs and todo items to speed up wiki updates and reduce size
Joey Hess [Thu, 5 Jan 2023 19:09:30 +0000 (15:09 -0400)]
remove old closed bugs and todo items to speed up wiki updates and reduce size

Remove closed bugs and todos that were last edited or commented before 2022.

Except for ones tagged projects/* since projects like datalad want to keep
around records of old deleted bugs longer.

Command line used:

for f in $(grep -l '|done\]\]' -- ./*.mdwn); do if ! grep -q "projects/" "$f"; then d="$(echo "$f" | sed 's/.mdwn$//')"; if [ -z "$(git log --since=01-01-2022 --pretty=oneline -- "$f")" -a -z "$(git log --since=01-01-2022 --pretty=oneline -- "$d")" ]; then git rm -- "./$f" ; git rm -rf "./$d"; fi; fi; done
for f in $(grep -l '\[\[done\]\]' -- ./*.mdwn); do if ! grep -q "projects/" "$f"; then d="$(echo "$f" | sed 's/.mdwn$//')"; if [ -z "$(git log --since=01-01-2022 --pretty=oneline -- "$f")" -a -z "$(git log --since=01-01-2022 --pretty=oneline -- "$d")" ]; then git rm -- "./$f" ; git rm -rf "./$d"; fi; fi; done

3 years agomove old closed bugs to datalad's bugs
Joey Hess [Thu, 5 Jan 2023 18:57:27 +0000 (14:57 -0400)]
move old closed bugs to datalad's bugs

These were not tagged datalad, but they did pay to get them fixed, and
it impacted them as seen in the bug reports..