git-annex.git
10 months agoupdate
Joey Hess [Wed, 24 Sep 2025 17:04:51 +0000 (13:04 -0400)]
update

10 months agocomments
Joey Hess [Wed, 24 Sep 2025 16:54:20 +0000 (12:54 -0400)]
comments

10 months agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Wed, 24 Sep 2025 16:21:32 +0000 (12:21 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

10 months agopreparing for a release
Joey Hess [Wed, 24 Sep 2025 16:06:45 +0000 (12:06 -0400)]
preparing for a release

10 months agoreorder
Joey Hess [Wed, 24 Sep 2025 16:05:45 +0000 (12:05 -0400)]
reorder

10 months agofixed
Joey Hess [Wed, 24 Sep 2025 16:03:18 +0000 (12:03 -0400)]
fixed

10 months agoAdded a comment: the X prefix conflicts with the eXternal backend namespace
Arnie97 [Wed, 24 Sep 2025 12:05:05 +0000 (12:05 +0000)]
Added a comment: the X prefix conflicts with the eXternal backend namespace

10 months ago(no commit message)
jnkl [Tue, 23 Sep 2025 18:50:19 +0000 (18:50 +0000)]

10 months agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Tue, 23 Sep 2025 17:57:09 +0000 (13:57 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

10 months agoclean up imports needed by old versions of ghc
Joey Hess [Tue, 23 Sep 2025 17:55:13 +0000 (13:55 -0400)]
clean up imports needed by old versions of ghc

Now that ghc 9.0.2 is the oldest supported version.

Eg cruft from
https://web.archive.org/web/20190424185034/https://prime.haskell.org/wiki/Libraries/Proposals/SemigroupMonoid

Sponsored-by: Jack Hill
10 months ago(no commit message)
jnkl [Tue, 23 Sep 2025 17:48:41 +0000 (17:48 +0000)]

10 months agoavoid unused import on Windows
Joey Hess [Tue, 23 Sep 2025 17:11:32 +0000 (13:11 -0400)]
avoid unused import on Windows

10 months agofix build warnings
Joey Hess [Tue, 23 Sep 2025 17:11:04 +0000 (13:11 -0400)]
fix build warnings

10 months agofix windows build
Joey Hess [Tue, 23 Sep 2025 17:02:24 +0000 (13:02 -0400)]
fix windows build

10 months agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Tue, 23 Sep 2025 16:54:33 +0000 (12:54 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

10 months agoinvalidate recorded content identifier tree when export changes
Joey Hess [Tue, 23 Sep 2025 16:52:55 +0000 (12:52 -0400)]
invalidate recorded content identifier tree when export changes

Fix bug that made changes to a special remote sometimes be missed when
importing a tree from it. The diff import would miss when a change was
exported, then manually undone on the special remote (eg deleting a newly
exported file). A full import is needed to catch such changes.

After upgrading, any such missed changes will be included in the next
tree imported from a special remote. This happens because the previously
recorded content identifier tree does not have export information included,
so it is treated as invalid, and a full import is done.

Fixes reversion introduced in version 10.20230626, commit
40017089f268391f79226592850b58855cdbf808

Unfortunately, this does mean that after each export, the next import will
be a full import. Which can take significantly longer than the diff import
does, when there are a lot of files in the tree.

It would be better if exporting also update the content identifier tree.
However, I don't know if that can be done inexpensively. It would be future
optimisation work, in any case.

(That could only be done for an export that is run in the same
repository as the import. When an export is run in a different repository,
the export.log gets updated, and that propagates to the repository where
import is later run. At that point, a full import is done.)

Sponsored-by: Luke T. Shumaker
10 months agoremove accidentially added debug prints
Joey Hess [Tue, 23 Sep 2025 15:34:16 +0000 (11:34 -0400)]
remove accidentially added debug prints

10 months ago(no commit message)
yarikoptic [Tue, 23 Sep 2025 14:26:07 +0000 (14:26 +0000)]

10 months agoWindows still FTBFS
yarikoptic [Tue, 23 Sep 2025 14:24:30 +0000 (14:24 +0000)]
Windows still FTBFS

10 months agothought
Joey Hess [Tue, 23 Sep 2025 03:55:22 +0000 (23:55 -0400)]
thought

10 months agoworse
Joey Hess [Tue, 23 Sep 2025 03:48:46 +0000 (23:48 -0400)]
worse

10 months agopromote forum post to bug, analysis
Joey Hess [Mon, 22 Sep 2025 23:30:02 +0000 (19:30 -0400)]
promote forum post to bug, analysis

10 months agofix test suite
Joey Hess [Mon, 22 Sep 2025 19:00:26 +0000 (15:00 -0400)]
fix test suite

dfbf76e2ca50d31eca6dee253c5bf3c892301463 broke it.

enableremote with the same type= as initremote had before might be worth
allowing, but I'll wait and see if someone else complains

10 months agoprevent deadlock when reconcileStaged runs restagePointerFiles
Joey Hess [Mon, 22 Sep 2025 18:56:50 +0000 (14:56 -0400)]
prevent deadlock when reconcileStaged runs restagePointerFiles

Fix hang that could occur when using git-annex adjust on a branch with a
number of files greater than annex.queuesize. Or potentially other
commands.

When reconcileStaged is running, the database is being opened. But
restagePointerFiles closes the database, and later writes to it. So it will
deadlock if called by reconcileStaged.

The deadlock occurred when the git queue happened to be full, causing
adding a call to restagePointerFiles to it to flush the queue and
restagePointerFiles to run at the wrong time.

Fixed by making reconcileStaged, when it populates or depopulates a pointer
file, arrange for restagePointerFiles to be run as a cleanup action, rather
than from the git queue.

But, what if restagePointerFiles is already in the git queue before
reconcileStaged is run? If it adds anything else to the git queue, causing
the queue to flush, it would still deadlock. To avoid this hypothetical
situation, added a Annex.inreconcilestaged, and made restagePointerFiles
check it and not do anything.

Note that, I did consider the simpler approach of only running
restagePointerFiles as a cleanup action, rather than from the git queue.
But see commit 6a3bd283b8af53f810982e002e435c0d7c040c59 for why it was made
to use the queue in the first place. I wanted to avoid tying this bug fix
to a behavior change.

Sponsored-by: mycroft
10 months agoupdate
Joey Hess [Mon, 22 Sep 2025 16:07:16 +0000 (12:07 -0400)]
update

10 months agocomment
Joey Hess [Mon, 22 Sep 2025 15:19:13 +0000 (11:19 -0400)]
comment

10 months agocomment
Joey Hess [Mon, 22 Sep 2025 15:15:32 +0000 (11:15 -0400)]
comment

10 months agocomment
Joey Hess [Mon, 22 Sep 2025 15:10:32 +0000 (11:10 -0400)]
comment

10 months agocomment
Joey Hess [Mon, 22 Sep 2025 15:02:24 +0000 (11:02 -0400)]
comment

10 months agoenableremote: Disallow using type= to attempt to change the type of an existing remote
Joey Hess [Mon, 22 Sep 2025 14:54:16 +0000 (10:54 -0400)]
enableremote: Disallow using type= to attempt to change the type of an existing remote

Changing the type out from under an existing special remote exposes the
existing config to something that may interpret it wildly differently. As
seen in the bug report, this can even result in behavior that makes
git-annex say it's buggy. So prevent the user from doing this. --sameas is
the better way.

Sponsored-by: Kevin Mueller
10 months agocomment
Joey Hess [Mon, 22 Sep 2025 14:27:16 +0000 (10:27 -0400)]
comment

10 months agoAdd git-remote-p2p-annex and git-remote-tor-annex to standalone builds
Joey Hess [Mon, 22 Sep 2025 14:20:36 +0000 (10:20 -0400)]
Add git-remote-p2p-annex and git-remote-tor-annex to standalone builds

10 months ago(no commit message)
matrss [Sat, 20 Sep 2025 17:54:22 +0000 (17:54 +0000)]

10 months agofix windows build
Joey Hess [Thu, 18 Sep 2025 15:59:22 +0000 (11:59 -0400)]
fix windows build

10 months agoresponse
Joey Hess [Thu, 18 Sep 2025 15:54:51 +0000 (11:54 -0400)]
response

10 months agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Thu, 18 Sep 2025 15:07:20 +0000 (11:07 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

10 months agoFTBFS on Windows
yarikoptic [Tue, 16 Sep 2025 20:48:22 +0000 (20:48 +0000)]
FTBFS on Windows

10 months agotypo
Joey Hess [Tue, 16 Sep 2025 20:38:54 +0000 (16:38 -0400)]
typo

10 months agoAdded a comment
yarikoptic [Tue, 16 Sep 2025 19:34:27 +0000 (19:34 +0000)]
Added a comment

10 months agoAdded a comment
yarikoptic [Tue, 16 Sep 2025 19:33:40 +0000 (19:33 +0000)]
Added a comment

10 months agoannex.assistant.allowunlocked
Joey Hess [Tue, 16 Sep 2025 18:58:26 +0000 (14:58 -0400)]
annex.assistant.allowunlocked

Sponsored-by: the NIH-funded NICEMAN (ReproNim TR&D3) project
10 months agofix bug in recent commit
Joey Hess [Tue, 16 Sep 2025 18:32:04 +0000 (14:32 -0400)]
fix bug in recent commit

10 months agotag repronim based on https://git-annex.branchable.com/forum/Is_there_a_way_to_have_a...
Joey Hess [Tue, 16 Sep 2025 17:45:36 +0000 (13:45 -0400)]
tag repronim based on https://git-annex.branchable.com/forum/Is_there_a_way_to_have_assistant_add_files_locked__63__/#comment-096bedb2d22d5aae6a51a53179372d4f

10 months agoimprove example
Joey Hess [Tue, 16 Sep 2025 17:34:54 +0000 (13:34 -0400)]
improve example

10 months agoclose
Joey Hess [Tue, 16 Sep 2025 17:27:25 +0000 (13:27 -0400)]
close

10 months agouse giveup not error
Joey Hess [Tue, 16 Sep 2025 17:27:01 +0000 (13:27 -0400)]
use giveup not error

This is not debugging a git-annex bug, so no stack trace needed.

10 months agoimprove error message when SETCREDS overwrites git-annex config
Joey Hess [Tue, 16 Sep 2025 17:22:14 +0000 (13:22 -0400)]
improve error message when SETCREDS overwrites git-annex config

That is not allowed, so it's not a bug in git-annex when it happens and
instead tell the special remote developer how it's messed up.

Note that currently only Remote.External can overwrite the parsed remote
config with a PassedThrough value. PassedThrough values are otherwise
only generated for configs that are not parsed by the remote config
parser.

Sponsored-by: Joshua Antonishen
10 months agofixed
Joey Hess [Tue, 16 Sep 2025 16:47:21 +0000 (12:47 -0400)]
fixed

10 months agoclose
Joey Hess [Tue, 16 Sep 2025 15:00:30 +0000 (11:00 -0400)]
close

10 months agocomment
Joey Hess [Tue, 16 Sep 2025 14:44:04 +0000 (10:44 -0400)]
comment

10 months agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Tue, 16 Sep 2025 14:36:19 +0000 (10:36 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

10 months agoset locale encoding after fdTohandle
Joey Hess [Tue, 16 Sep 2025 01:33:10 +0000 (21:33 -0400)]
set locale encoding after fdTohandle

fdToHandle does not set the usual system locale encoding,
so when the Handle is used for any String IO, it needs to be done
manually for correctness.

I don't know if this fixes any bugs. It might eg, fix a bug with
multicast receive of a file.

Sponsored-by: Leon Schuermann
10 months agoAdded a comment
yarikoptic [Tue, 16 Sep 2025 00:40:22 +0000 (00:40 +0000)]
Added a comment

10 months agoboot libs
Joey Hess [Mon, 15 Sep 2025 23:58:54 +0000 (19:58 -0400)]
boot libs

10 months agowork around file-io not setting locale encoding when opening a Handle
Joey Hess [Mon, 15 Sep 2025 23:25:03 +0000 (19:25 -0400)]
work around file-io not setting locale encoding when opening a Handle

Works around this bug https://github.com/haskell/file-io/issues/45

The fix is in Utility.FileIO.CloseOnExec because all use of file-io is
already wrapped through that module. Although perhaps that ought to be
refactored at this point.

I'd hope that file-io will eventually fix this bug, and also provide
CloseOnExec variants of its functions. That would allow depending on the
fixed version, and removing this ugly code.

Note that, functions like readFile that don't care about the encoding
due to reading/writing a ByteString were kept optimally fast by not
setting the encoding. This avoids an IORef read and write per open.

Sponsored-by: Graham Spencer
10 months agobug
Joey Hess [Mon, 15 Sep 2025 22:50:42 +0000 (18:50 -0400)]
bug

10 months agorequire_OsPath branch
Joey Hess [Mon, 15 Sep 2025 20:37:53 +0000 (16:37 -0400)]
require_OsPath branch

10 months agocomment
Joey Hess [Mon, 15 Sep 2025 20:13:25 +0000 (16:13 -0400)]
comment

10 months agoupdate
Joey Hess [Mon, 15 Sep 2025 18:27:45 +0000 (14:27 -0400)]
update

10 months agocomment
Joey Hess [Mon, 15 Sep 2025 18:25:08 +0000 (14:25 -0400)]
comment

10 months agosmall optimisation
Joey Hess [Mon, 15 Sep 2025 17:49:14 +0000 (13:49 -0400)]
small optimisation

10 months agodrop problem end characters from filename operating on String not RawFilePath
Joey Hess [Mon, 15 Sep 2025 17:42:50 +0000 (13:42 -0400)]
drop problem end characters from filename operating on String not RawFilePath

Fix bug that could cause an invalid utf-8 sequence to be used in a
temporary filename when the input filename was valid utf-8.

Sponsored-by: k0ld
10 months agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Mon, 15 Sep 2025 17:15:41 +0000 (13:15 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

10 months agofix mispaste of comment
Joey Hess [Mon, 15 Sep 2025 17:14:56 +0000 (13:14 -0400)]
fix mispaste of comment

10 months agoAdded a comment
yarikoptic [Mon, 15 Sep 2025 17:11:51 +0000 (17:11 +0000)]
Added a comment

10 months agocomment
Joey Hess [Mon, 15 Sep 2025 16:15:51 +0000 (12:15 -0400)]
comment

10 months agoreopen
Joey Hess [Mon, 15 Sep 2025 16:07:29 +0000 (12:07 -0400)]
reopen

10 months agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Mon, 15 Sep 2025 16:04:06 +0000 (12:04 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

10 months agofix p2phttp worker thread leak with deleted repository LOCKCONTENT
Joey Hess [Mon, 15 Sep 2025 15:59:37 +0000 (11:59 -0400)]
fix p2phttp worker thread leak with deleted repository LOCKCONTENT

p2phttp: Fix a hang that could occur when used with --directory, and a
repository in the repository got removed.

It could leak up to -J number of worker threads, but this only affected a
client trying to access the deleted repository.

It may be that this could also affect a non-deleted repository, and also
leak a worker thread, if invalid p2p protocol is sent.

11 months agoAdded a comment
yarikoptic [Thu, 11 Sep 2025 15:46:53 +0000 (15:46 +0000)]
Added a comment

11 months agoAdded a comment
yarikoptic [Thu, 11 Sep 2025 12:30:50 +0000 (12:30 +0000)]
Added a comment

11 months ago(no commit message)
git-annex.branchable.com-2746523870@d0bf050902d9409ff6e93b060e84e482d7e91e7f [Thu, 11 Sep 2025 12:25:06 +0000 (12:25 +0000)]

11 months ago(no commit message)
git-annex.branchable.com-2746523870@d0bf050902d9409ff6e93b060e84e482d7e91e7f [Thu, 11 Sep 2025 12:23:36 +0000 (12:23 +0000)]

11 months agoAdded a comment: git annex bundle - questions
psxvoid [Thu, 11 Sep 2025 08:02:08 +0000 (08:02 +0000)]
Added a comment: git annex bundle - questions

11 months agoAdded a comment: resolved
psxvoid [Thu, 11 Sep 2025 05:12:30 +0000 (05:12 +0000)]
Added a comment: resolved

11 months agonoCreateProcessWhile to fix close-on-exec races
Joey Hess [Wed, 10 Sep 2025 18:29:15 +0000 (14:29 -0400)]
noCreateProcessWhile to fix close-on-exec races

Sponsored-by: the NIH-funded NICEMAN (ReproNim TR&D3) project
11 months agoalways import Utility.Process rather than System.Process
Joey Hess [Wed, 10 Sep 2025 17:53:50 +0000 (13:53 -0400)]
always import Utility.Process rather than System.Process

This is groundwork for Utility.Process doing some extra locking, for
which it's important that it's always used for process creation.

11 months agowindows build fixes
Joey Hess [Wed, 10 Sep 2025 17:34:17 +0000 (13:34 -0400)]
windows build fixes

11 months agoRemoved support for building with cryptonite, use crypton.
Joey Hess [Wed, 10 Sep 2025 17:21:06 +0000 (13:21 -0400)]
Removed support for building with cryptonite, use crypton.

cryptonite is not maintained and has security problems.
Debian stable includes crypton by now.

11 months agoImprove performance when used with a local git remote that has a large working tree
Joey Hess [Wed, 10 Sep 2025 16:05:55 +0000 (12:05 -0400)]
Improve performance when used with a local git remote that has a large working tree

git write-tree was being run once per file git-annex acts on when eg,
getting files, which is slow when the remote repository has a large
tree.

onLocal calls quiesce after each action, and quiesce closes the keys db
since [[!commit ba7ecbc6a9c]]. Which has a relevant comment about
performance. I have not addressed that, the keys db still gets closed and
reopened after each file.

Turns out that, since git write-tree was run by each call to
reconcileStaged, the .git/annex/keysdb.cache value was never the
same as the git index's inode. Because git write-tree updates the index's
mtime even when no changes have been made.

And so, when the database got closed and reopened, reconcileStaged would
see a changed index, and run git write-tree again. Over and over.

I considered writing the index's new inodecache after write-tree to the
keysdb.cache, but that would be vulnerable to a race, if the index was
changed just after write-tree.

The fix was to stop using keysb.cache at all. When the database is closed
and later reopened by the same process, avoid re-doing reconcileStaged.

Now that .git/annex/keysdb.cache is no longer used. It could be removed,
but the time overhead of removing it would be more than the space overhead
of keeping it. Defferred removal to the v11 upgrade.

Sponsored-by: unqueued
11 months agoMerge branch 'master' of ssh://git-annex.branchable.com
Joey Hess [Wed, 10 Sep 2025 14:22:10 +0000 (10:22 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com

11 months agobug report
Joey Hess [Wed, 10 Sep 2025 14:16:52 +0000 (10:16 -0400)]
bug report

11 months agoinitial report on problem with # in the path
yarikoptic [Tue, 9 Sep 2025 14:36:29 +0000 (14:36 +0000)]
initial report on problem with # in the path

11 months agoAdded a comment
yarikoptic [Tue, 9 Sep 2025 12:47:26 +0000 (12:47 +0000)]
Added a comment

11 months ago(no commit message)
jkrebian [Tue, 9 Sep 2025 07:02:04 +0000 (07:02 +0000)]

11 months agoAdded a comment
babudarabu@232a9694ce5401143f6210561371f887dd15cd61 [Mon, 8 Sep 2025 17:13:02 +0000 (17:13 +0000)]
Added a comment

11 months agorename forum/notes_and_enhancements_git_annex_on_android.mdwn to forum/notes_and_enha...
waldi5001 [Mon, 8 Sep 2025 10:00:45 +0000 (10:00 +0000)]
rename forum/notes_and_enhancements_git_annex_on_android.mdwn to forum/notes_and_enhancements_for_git_annex_on_android.mdwn

11 months ago(no commit message)
waldi5001 [Mon, 8 Sep 2025 09:55:03 +0000 (09:55 +0000)]

11 months agoAdded a comment
waldi5001 [Mon, 8 Sep 2025 08:47:56 +0000 (08:47 +0000)]
Added a comment

11 months agocomments
Joey Hess [Sat, 6 Sep 2025 17:38:17 +0000 (13:38 -0400)]
comments

11 months agochangelog for close-on-exec work
Joey Hess [Fri, 5 Sep 2025 20:19:50 +0000 (16:19 -0400)]
changelog for close-on-exec work

11 months agoupdate
Joey Hess [Fri, 5 Sep 2025 20:18:07 +0000 (16:18 -0400)]
update

11 months agoopen feed file with close-on-exec bit set
Joey Hess [Fri, 5 Sep 2025 20:02:17 +0000 (16:02 -0400)]
open feed file with close-on-exec bit set

parseFeedFromFile does not set the bit, so open and read the file
ourselves.

Versioned dependency on utf8-string should not cause any issues,
that version is available in all all versions of debian that package it.

Sponsored-by: the NIH-funded NICEMAN (ReproNim TR&D3) project
11 months agoclarify
Joey Hess [Fri, 5 Sep 2025 19:49:33 +0000 (15:49 -0400)]
clarify

11 months agoupdate
Joey Hess [Fri, 5 Sep 2025 19:48:43 +0000 (15:48 -0400)]
update

11 months agoconvert withFile and withBinaryFile to close-on-exec safe versions
Joey Hess [Fri, 5 Sep 2025 19:44:43 +0000 (15:44 -0400)]
convert withFile and withBinaryFile to close-on-exec safe versions

At this point, every call in git-annex to withFile, openBinaryFile,
withBinaryFile, appendFile, and openTempFile, readFile, and writeFile
have been converted.

Sponsored-by: the NIH-funded NICEMAN (ReproNim TR&D3) project
11 months agoconvert all readFile, writeFile, and appendFile to close-on-exec safe versions
Joey Hess [Fri, 5 Sep 2025 19:37:13 +0000 (15:37 -0400)]
convert all readFile, writeFile, and appendFile to close-on-exec safe versions

Even in the Build system. This allows grepping to make sure that there
are none left un-converted:

git grep "writeFile" |grep -v F\\.| grep -v doc/|grep -v writeFileString | grep -v writeFileProtected |grep -v Utility/FileIO
git grep "readFile" |grep -v F\\.| grep -v doc/|grep -v readFileString |grep -v Utility/FileIO
git grep "appendFile" |grep -v F\\.| grep -v doc/|grep -v appendFileString |grep -v Utility/FileIO

Might be nice to automate that to prevent future mistakes...

Sponsored-by: the NIH-funded NICEMAN (ReproNim TR&D3) project
11 months agoadd readFileString, writeFileString, appendFileString
Joey Hess [Fri, 5 Sep 2025 18:51:26 +0000 (14:51 -0400)]
add readFileString, writeFileString, appendFileString

Exported by Common, so they will be available everywhere.
These are the same as readFile, writeFile, appendFile.
But have two benefits:

* They take OsPath, so using them avoids converting back and forth
  unncessarily.
* They use the close-on-exec flag so can't leak FDs to child processes.
  Unlike the standard Haskell versions which unfortunately have that
  wart currently. (I do hope the standard versions get fixed
  eventually.)

Sponsored-by: the NIH-funded NICEMAN (ReproNim TR&D3) project
11 months agoupdate
Joey Hess [Fri, 5 Sep 2025 18:07:40 +0000 (14:07 -0400)]
update

11 months agoupdate
Joey Hess [Fri, 5 Sep 2025 18:06:15 +0000 (14:06 -0400)]
update