Joey Hess [Mon, 19 Oct 2020 16:16:15 +0000 (12:16 -0400)]
close
Joey Hess [Mon, 19 Oct 2020 15:14:01 +0000 (11:14 -0400)]
switch to Haskell2010
When I put in Haskell98 this spring, I was under the mistaken
apprehension that ghc defaulted to that. But it actually its default
is a third mode, which is closer to Haskell2010 but with some differences.
The manual says "By default, GHC mainly aims to behave (mostly) like a
Haskell 2010 compiler"
Fixed two cases where the Haskell98 do indentation flexability let
wrongly indented code build. That is one of the places where
ghc does not behave like Haskell2010 by default.
The other place that I think I was concerned about, is GHC manual
section 19.1.1.3. Expressions and patterns. But that only seems to
affect code using bottoms, so would only affect pure functions throwing
an error, which I don't think git-annex does in many places as it's
pretty horrid style. And it would only affect rare cases like shown in
that section. If it did happen, it would mean that the error was not
thrown before specifying Haskell98, and then was. Haskell2010 behaves
the same as Haskell98.
This commit was sponsored by Denis Dzyubenko on Patreon.
Joey Hess [Tue, 13 Oct 2020 22:57:30 +0000 (18:57 -0400)]
comment
Joey Hess [Tue, 13 Oct 2020 22:41:47 +0000 (18:41 -0400)]
devblog
Joey Hess [Tue, 13 Oct 2020 20:46:32 +0000 (16:46 -0400)]
devblog
Joey Hess [Tue, 13 Oct 2020 20:31:01 +0000 (16:31 -0400)]
Fix a memory leak introduced in the last release
The problem was this line:
cleanup = and <$> sequence (map snd v)
That caused all of v to be held onto until the end, when the cleanup action
was run.
I could not seem to find a bang pattern that avoided the leak, so I
resorted to a IORef, rather clunky, but not a performance problem because
it will only be written once per git ls-files, so typically just 1 time.
This commit was sponsored by Mark Reidenbach on Patreon.
Joey Hess [Tue, 13 Oct 2020 19:36:01 +0000 (15:36 -0400)]
enable profiling in dev build
Joey Hess [Tue, 13 Oct 2020 19:33:59 +0000 (15:33 -0400)]
blew a day investigating this and still don't understand it fully
Joey Hess [Tue, 13 Oct 2020 16:24:13 +0000 (12:24 -0400)]
comment
Joey Hess [Tue, 13 Oct 2020 16:19:36 +0000 (12:19 -0400)]
add a few missing copyrights
Joey Hess [Mon, 12 Oct 2020 20:14:51 +0000 (16:14 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com into master
Joey Hess [Mon, 12 Oct 2020 20:12:24 +0000 (16:12 -0400)]
comment
Joey Hess [Mon, 12 Oct 2020 19:47:46 +0000 (15:47 -0400)]
comment
This commit was sponsored by Ethan Aubin.
Maximilian [Mon, 12 Oct 2020 19:53:27 +0000 (19:53 +0000)]
Added a comment
Joey Hess [Mon, 12 Oct 2020 19:47:46 +0000 (15:47 -0400)]
comment
This commit was sponsored by Ethan Aubin.
Joey Hess [Mon, 12 Oct 2020 19:10:57 +0000 (15:10 -0400)]
comment
Joey Hess [Mon, 12 Oct 2020 19:03:18 +0000 (15:03 -0400)]
comment
Joey Hess [Mon, 12 Oct 2020 18:50:38 +0000 (14:50 -0400)]
close as notabug
Joey Hess [Mon, 12 Oct 2020 18:49:36 +0000 (14:49 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com into master
Joey Hess [Mon, 12 Oct 2020 18:48:26 +0000 (14:48 -0400)]
profiling
Currently a bit stuck, but at least starting to get some clues on this
memory leak.
This commit was sponsored by Ethan Aubin.
Lukey [Mon, 12 Oct 2020 17:20:37 +0000 (17:20 +0000)]
Added a comment
mhauru [Mon, 12 Oct 2020 17:07:58 +0000 (17:07 +0000)]
Added a comment
Joey Hess [Mon, 12 Oct 2020 16:32:53 +0000 (12:32 -0400)]
comment
yarikoptic [Mon, 12 Oct 2020 16:00:32 +0000 (16:00 +0000)]
Added a comment
Joey Hess [Mon, 12 Oct 2020 15:41:58 +0000 (11:41 -0400)]
comment
Joey Hess [Mon, 12 Oct 2020 15:35:40 +0000 (11:35 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com into master
Lukey [Sun, 11 Oct 2020 14:17:47 +0000 (14:17 +0000)]
Added a comment
Lukey [Sun, 11 Oct 2020 14:08:27 +0000 (14:08 +0000)]
Added a comment
Lukey [Sun, 11 Oct 2020 14:01:35 +0000 (14:01 +0000)]
Added a comment
yarikoptic [Sun, 11 Oct 2020 13:51:28 +0000 (13:51 +0000)]
Added a comment
Joey Hess [Sat, 10 Oct 2020 14:40:35 +0000 (10:40 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com into master
Maximilian [Sat, 10 Oct 2020 09:11:21 +0000 (09:11 +0000)]
Joey Hess [Sat, 10 Oct 2020 00:24:56 +0000 (20:24 -0400)]
comment
Joey Hess [Thu, 8 Oct 2020 18:36:36 +0000 (14:36 -0400)]
this was closed a long time ago, but a rename made it re-appear
Joey Hess [Thu, 8 Oct 2020 14:50:39 +0000 (10:50 -0400)]
Fix build on Windows with network-3
inet_addr was removed, but all this needs is localhost, so hardcoding it
should work fine.
It may be that this windows ifdef is no longer needed. It was added in 2013
with a note that getAddrInfo didn't work on windows, but it seems likely
such a problem would have been fixed since.
Joey Hess [Thu, 8 Oct 2020 14:38:15 +0000 (10:38 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com into master
Joey Hess [Thu, 8 Oct 2020 14:35:16 +0000 (10:35 -0400)]
change from deprecated and removed aNY_PORT to defaultPort
(Both are just 0 internally.)
Joey Hess [Thu, 8 Oct 2020 14:31:03 +0000 (10:31 -0400)]
remove unused define
Joey Hess [Thu, 8 Oct 2020 14:09:12 +0000 (10:09 -0400)]
design for finally slaying this beast
This commit was sponsored by Ethan Aubin.
jwodder [Thu, 8 Oct 2020 13:02:20 +0000 (13:02 +0000)]
jwodder [Thu, 8 Oct 2020 12:44:52 +0000 (12:44 +0000)]
Added a comment
Joey Hess [Wed, 7 Oct 2020 18:18:57 +0000 (14:18 -0400)]
add news item for git-annex 8.
20201007
Joey Hess [Wed, 7 Oct 2020 18:10:56 +0000 (14:10 -0400)]
releasing package git-annex version 8.
20201007
Joey Hess [Wed, 7 Oct 2020 16:12:53 +0000 (12:12 -0400)]
renamed some filenames to not end with a period
Windows does not allow such filenames, which prevented cloning git-annex
on windows.
Joey Hess [Wed, 7 Oct 2020 16:04:54 +0000 (12:04 -0400)]
Fix a build failure on Windows.
Joey Hess [Tue, 6 Oct 2020 15:25:17 +0000 (11:25 -0400)]
avoid depending on cmp
If it was not available, the caches would get deleted unncessarily.
This should be just as fast as using it.
Joey Hess [Tue, 6 Oct 2020 15:23:55 +0000 (11:23 -0400)]
shorten buildid
No need to end it with " git-annex"
Joey Hess [Tue, 6 Oct 2020 14:54:10 +0000 (10:54 -0400)]
response
Joey Hess [Tue, 6 Oct 2020 14:24:09 +0000 (10:24 -0400)]
close
Joey Hess [Tue, 6 Oct 2020 14:07:48 +0000 (10:07 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com into master
Joey Hess [Tue, 6 Oct 2020 14:07:27 +0000 (10:07 -0400)]
update
Joey Hess [Tue, 6 Oct 2020 12:46:35 +0000 (08:46 -0400)]
move cache cleanup back to before cache creation
It was moved to avoid a race, but that's now avoided in another way.
I prefer having it here, because this way if it somehow fails and
deletes the locpath that is going to be used, at least it will get
re-created.
Joey Hess [Tue, 6 Oct 2020 12:44:05 +0000 (08:44 -0400)]
fix another race
Use a different tmp directory so the cache cleanup won't delete the
locpath directory while it's being populated.
This does change the hash used for the locpath directory, but it already
changed in
f0ec725234463a7165a4339c8a65486a9b04ab3a
yarikoptic [Tue, 6 Oct 2020 01:27:01 +0000 (01:27 +0000)]
Added a comment
yarikoptic [Tue, 6 Oct 2020 00:04:38 +0000 (00:04 +0000)]
Added a comment: FTR
yarikoptic [Mon, 5 Oct 2020 23:17:08 +0000 (23:17 +0000)]
Added a comment
yarikoptic [Mon, 5 Oct 2020 23:11:32 +0000 (23:11 +0000)]
Added a comment: just comments
Joey Hess [Mon, 5 Oct 2020 19:19:48 +0000 (15:19 -0400)]
comment
Joey Hess [Mon, 5 Oct 2020 18:51:19 +0000 (14:51 -0400)]
comment
Joey Hess [Mon, 5 Oct 2020 18:38:29 +0000 (14:38 -0400)]
comment
Joey Hess [Mon, 5 Oct 2020 18:37:33 +0000 (14:37 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com into master
Joey Hess [Mon, 5 Oct 2020 18:36:36 +0000 (14:36 -0400)]
fix confirmed
Joey Hess [Mon, 5 Oct 2020 18:17:46 +0000 (14:17 -0400)]
fully atomic LOCPATH populating
This fixes a race between two runshells from two different
bundles. One could have run the cache cleanup code, seen the
LOCPATH the other one was in the process of populating, which didn't
have a base or a buildid file written yet, and so the cache cleanup code
would delete it out from under the other process.
Also, doing it fully atomically simplifies where the races between two
runshell processes from the same bundle. Now that needs to be
dealt with to only the mv that puts it in place.
Note that, if the same bundle has 2 runshells run first thing, they will
both generate locales, which is unncessary work, but that should be a
very unusual circumstance and after the LOCPATH is set up, it won't
happen again anyway.
kyle [Mon, 5 Oct 2020 18:09:07 +0000 (18:09 +0000)]
Added a comment
Joey Hess [Mon, 5 Oct 2020 18:07:27 +0000 (14:07 -0400)]
move cache cleanup to avoid a race
This should fix doc/bugs/standalone_runshell_can_race_and_fail_to_remove___96____126____47__.cache__47__git-annex__47__locales__47____96___dirs
where 2 runshells were running and the second one tried to clean up
LOCPATH while the first one was still populating it.
By moving the cleanup until after LOCPATH is populated, we guarantee
it's populated, so don't need to worry about such a race with another
process populating our same LOCPATH.
Joey Hess [Mon, 5 Oct 2020 17:53:43 +0000 (13:53 -0400)]
include buildid in LOCPATH
This avoids the possibility that the bundle could be updated in place,
leading to LOCPATH existing but containing locales for the old version,
which needed to be tested for with code that was not race-free.
LOCPATH/buildid is still written and checked when cleaning up stale caches.
That is not actually necessary, except old versions of the standalone
bundle expect to see it, and this prevents them cleaning up the locale
cache of a new version. And still checking it prevents the new version
cleaning up the locale cache of the old version while the old version is
still in use.
Added explicit tests before creating LOCPATH and the base and buildid files.
The buildid file no longer needs to be updated every time, because it's
stable for the given LOCPATH directory.
And the base file actually did not need to be updated every time,
because the LOCPATH is derived from base, so if the bundle is moved
elsewhere, a different LOCPATH will be used.
Transitioning to this will mean that two git-annex builds that otherwise
have the same buildid -- the same git-annex md5sum -- will use different
LOCPATH values, but that's handled fine by the cache cleanup code, so at
most it will mean one extra generation of the locale files.
yarikoptic [Mon, 5 Oct 2020 18:02:51 +0000 (18:02 +0000)]
Added a comment
Joey Hess [Mon, 5 Oct 2020 17:45:27 +0000 (13:45 -0400)]
switch to md5sum for buildid
This does not need to be cryptogrphically secure, and a little shorter
will be better when using the buildid in a path in subsequent changes.
Joey Hess [Mon, 5 Oct 2020 17:42:43 +0000 (13:42 -0400)]
runshell: fix some mess left in a race
This seems to be the best way to deal with the race; if the first and
second runshell are running very close together, the first will generate
the locale directory, and a second test -d would still leave a race
window.
kyle [Mon, 5 Oct 2020 17:42:26 +0000 (17:42 +0000)]
Added a comment
Joey Hess [Mon, 5 Oct 2020 17:38:34 +0000 (13:38 -0400)]
runshell: Update files atomically when preparing to run git-annex
This does not make it entirely idempotent, but it's a start.
Joey Hess [Mon, 5 Oct 2020 17:37:00 +0000 (13:37 -0400)]
small runshell optimisation
avoid some test -d for locale variables that are not set
Joey Hess [Mon, 5 Oct 2020 16:57:16 +0000 (12:57 -0400)]
comment
Joey Hess [Mon, 5 Oct 2020 16:44:40 +0000 (12:44 -0400)]
runshell: Fix a edge case where rm errors were sent to stdout, which could confuse things parsing git-annex output.
Joey Hess [Mon, 5 Oct 2020 16:37:43 +0000 (12:37 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com into master
Joey Hess [Mon, 5 Oct 2020 16:36:36 +0000 (12:36 -0400)]
update
yarikoptic [Mon, 5 Oct 2020 15:26:10 +0000 (15:26 +0000)]
Added a comment
Ilya_Shlyakhter [Sun, 4 Oct 2020 15:45:16 +0000 (15:45 +0000)]
Added a comment: standalone build
yarikoptic [Sat, 3 Oct 2020 04:04:28 +0000 (04:04 +0000)]
Initial report on hanging batched processes
yarikoptic [Sat, 3 Oct 2020 03:52:33 +0000 (03:52 +0000)]
initial report on rm failing to remove cache/locales
Joey Hess [Fri, 2 Oct 2020 20:39:10 +0000 (16:39 -0400)]
update to lts-16.16
Joey Hess [Fri, 2 Oct 2020 17:57:10 +0000 (13:57 -0400)]
Enable building with git-annex benchmark by default
Only turning it off when the criterion library is not installed.
Not enabled for osx or i386ancient yet since that will need some
invesitgation to update their respective stack.yaml files.
Joey Hess [Fri, 2 Oct 2020 17:45:09 +0000 (13:45 -0400)]
remove unused import
Joey Hess [Fri, 2 Oct 2020 17:30:24 +0000 (13:30 -0400)]
Fix build with Benchmark build flag
Broke a while ago during optimisation work, and not noticed since the flag
is disabled by default.
This commit was sponsored by Brock Spratlen on Patreon.
kyle [Thu, 1 Oct 2020 19:07:48 +0000 (19:07 +0000)]
Added a comment
yarikoptic [Thu, 1 Oct 2020 17:26:10 +0000 (17:26 +0000)]
Added a comment
Joey Hess [Thu, 1 Oct 2020 17:08:09 +0000 (13:08 -0400)]
response
Joey Hess [Thu, 1 Oct 2020 16:53:24 +0000 (12:53 -0400)]
Merge branch 'master' of ssh://git-annex.branchable.com into master
yarikoptic [Thu, 1 Oct 2020 14:06:34 +0000 (14:06 +0000)]
Added a comment
Joey Hess [Wed, 30 Sep 2020 15:09:09 +0000 (11:09 -0400)]
avoid git check-ignore overhead on importing known files
isKnownImportLocation does a database lookup and there's an index
to make that lookup fast, so it's probably faster than talking to git
check-ignore. Checking the matcher is faster still.
While before the gitignore check was added it did not need to always
check isknown, now it does, because it's that or the more expensive
notignored. But at least we can skip notignored when a file is known,
which will often be the common case: Importing from a remote that's been
exported to, and/or imported from before, only new files will not be
known, so only those will need to check notignored.
At first, I had this:
(matches <&&> (isknown <||> notignored)) <||> isknown
Notice that checks isknown every time, whether it matches or not.
So, it's no slower to instead do this:
isknown <||> (matches <&&> notignored)
That has the benefit that, when it's known, it doesn't need to run
matches, which while faster than isknown, is still going to use some CPU.
And it perhaps more clearly expresses the condition: Any known file is
wanted, otherwise it's down to what matches and is not ignored.
This commit was sponsored by Jack Hill on Patren.
Joey Hess [Wed, 30 Sep 2020 14:41:59 +0000 (10:41 -0400)]
import: Check gitignores when importing trees from special remotes
It seemed best to do this, for consistency with every other way files can
get into a git-annex repo. Although it's just a bit strange that a local
.gitignore file affects the pseudo-commits made for the remote that's
imported from.
This commit was sponsored by Brett Eisenberg on Patreon.
Joey Hess [Wed, 30 Sep 2020 14:10:03 +0000 (10:10 -0400)]
avoid a second traversal of the ImportableContents
Do all filtering in one pass.
Joey Hess [Wed, 30 Sep 2020 13:40:06 +0000 (09:40 -0400)]
typo
Joey Hess [Tue, 29 Sep 2020 21:53:48 +0000 (17:53 -0400)]
Parse youtube-dl progress output
Which lets progress be displayed when doing concurrent downloads.
Amoung other things, like --json-progress etc.
The youtube-dl output is no longer displayed, except for any errors.
This commit was sponsored by Denis Dzyubenko on Patreon.
Joey Hess [Tue, 29 Sep 2020 17:59:51 +0000 (13:59 -0400)]
add a pointer for git-annex repos over http
So users are less likely to try to use httpalso for them, which is not
its purpose.
Joey Hess [Tue, 29 Sep 2020 17:56:27 +0000 (13:56 -0400)]
httpalso: Support being used with special remotes that do not have encryption= in their config.
Joey Hess [Tue, 29 Sep 2020 17:51:51 +0000 (13:51 -0400)]
add import
Joey Hess [Tue, 29 Sep 2020 17:45:14 +0000 (13:45 -0400)]
upgrade: Avoid an upgrade failure of a bare repo in unusual circumstances