Summary:
The "cache" refers to branch and tag caches. Both of them are planned for
removal. Therefore remove the test.
Reviewed By: singhsrb
Differential Revision: D13978569
fbshipit-source-id: 8593f54af4ecd3fb759d2a3a0fe00eb9d8505112
Summary:
`test-patchbomb.t` uses named branches and does not look simple to fix.
We don't use email patches internally, and it's not hard to write scripts
around `hg export`. Therefore drop the extension and its tests.
Reviewed By: singhsrb
Differential Revision: D13978577
fbshipit-source-id: 19867ae68c19c996bfce064eb2234705939db9ea
Summary:
The test uses named branches so it will break without modification in the
future. Since bundle1 and ssh are "unloved" features, remove the test instead
of fixing it.
Reviewed By: singhsrb
Differential Revision: D13978573
fbshipit-source-id: d5446685056457766d1752446de9d842f40f3ebf
Summary:
The test uses named branches extensively, partically because there are multiple
ways to specify where to update:
hg clone URL#NAME1 -b NAME2 -u NAME3
^^^^^^ ^^^^^ ^^^^^
branch name | |
branch name, again! |
revset, can also be a branch name
In the long run, we'll kill NAME1 (`#...` syntax) and NAME2 (`clone -b` flag).
So the (undesirable) complexity will be gone. Remove the test to make the
transition easier.
The `clone -u REV` feature is still tested in `test-clone.t`.
Reviewed By: singhsrb
Differential Revision: D13978578
fbshipit-source-id: 839ad39779c4526e4be48e133610bdb46f57fc93
Summary: This makes it compatible with future releases.
Reviewed By: singhsrb
Differential Revision: D13978572
fbshipit-source-id: 9394a811b848f1160aaa54be0d09dfc2ea142e81
Summary: Named branches are going away. Remove the logic around it.
Reviewed By: phillco
Differential Revision: D13978575
fbshipit-source-id: d6e28d7cadffa612f74a2afc12800829d6113dfa
Summary:
The test uses branches to name commits server-side and isolate them from the
bookmark namespace. Branches are going away. Use local tags instead.
Reviewed By: phillco
Differential Revision: D14057253
fbshipit-source-id: 33b8ab63027042eb5e244187874190bdd31e62d1
Summary:
The test uses named branches heavily. It seems the test just wants to make
sure remotenames work with the schemes extension. Rewrite it to stop using
named branches.
Reviewed By: singhsrb
Differential Revision: D13978570
fbshipit-source-id: 3d0f284bb2398da0283b3bf5a06d879182a70d84
Summary:
Purging of packs may happen before the repack operation, and when loosefiles
are also present, repack will reconstruct the packs from them. By forcing a
packsonly, we won't have a chance of reconstructing the purged ones.
Differential Revision: D14013788
fbshipit-source-id: d77d9ec2b9c5a086caaf9164a8ac56dd7d1e1255
Summary:
The refresh function is intended to load on-disk packfiles that aren't yet
present in the cache, but it had several issues that resulted in unexpected
behavior. Mainly, the new files would be added in the wrong order in the cache,
and runonpacks would not update the last used packs.
Differential Revision: D14013785
fbshipit-source-id: da4151b859441f0ba51863f3907852922c255f82
Summary:
Previously, the logic for chosing between the rust repack, and the python one
was deeply integrated with the python repack. This resulted in work being
performed unnecessarily.
The drawbacks of this method is that in the case of fallback to Python, the
repacklock will be dropped and re-obtained, as well as the prerepack hook will
be ran twice. Since the python code will soon be entirely replaced this is
probably not important.
Differential Revision: D14013790
fbshipit-source-id: 7c754db3fb984cc9e6b7df1a4e32b72fa4a1531b
Summary: Most of the code is identical between these 2 functions, so let's merge them.
Differential Revision: D14013787
fbshipit-source-id: f0469e05fcf723db02d9f4c28097e61f958284a5
Summary:
Repacking one packfile will yield the same packfile, so we can save some IO by
not trying to repack.
Differential Revision: D14013789
fbshipit-source-id: 8069840cc7cb1837eb94cea97e50b3bbaa548873
Summary:
This should auto recover from a state where watchman returns too many files,
which makes "hg status" slow.
The default config value 200 is not chosen very scientifically, but it should
cover the 6s status case where there are 90k changed files reported.
Differential Revision: D14036494
fbshipit-source-id: 368f53e99e2e54343a6ac9145cbb86e4ac65e4ac
Summary:
akushner encountered a case where "hg status" takes 5+ seconds and cannot
self-recover. That is because watchman constantly returns huge number of files
and fsmonitor didn't choose to update the clock to address that.
Reviewed By: DurhamG
Differential Revision: D14036493
fbshipit-source-id: 0bd8f7c9e3a51b6b4d2a1584b68883f8579e3f0f
Summary:
D13875656 made a config path change that breaks tests without using HGRCPATH,
or local build runs.
Reviewed By: DurhamG
Differential Revision: D14034919
fbshipit-source-id: 80de214f1769a8f40e79dc0ab1dbba4d55f506a7
Summary:
We've settled on the following grouping for imports:
- standard library
- 3rd party crates
- internal crates
- modules within same crate
This diff updates revisionstore accordingly.
Reviewed By: singhsrb
Differential Revision: D14030243
fbshipit-source-id: 74a7897342e39eb1d80202c8aae8c149bf08fc41
Summary: This is caused by D13516212, and not covered by tests.
Reviewed By: DurhamG, StanislavGlebik
Differential Revision: D14025817
fbshipit-source-id: 8a529085c40c5d2ae17ccf3de1ccea22c8d7dbd2
Summary: It is unavailable on OSX, but we have a custom implementation.
Reviewed By: DurhamG
Differential Revision: D14028103
fbshipit-source-id: 4b6916792fc9da43795b0f9db57ec2a057c92283
Summary: It turns out that `hyper-tls` does not support ALPN for negotiating HTTP/2 connections, and only supports HTTP/2 prior knowledge. (This is a limitation of the underlying TLS library, `native-tls`.) Unfortunately, while the Mononoke API server itself is fine with HTTP/2 prior knowledge for non-TLS connections, the Mononoke VIPs require TLS, and thus per the HTTP/2 spec require ALPN negotiation from an HTTP/1.1 initial connection. As a result, we need to revert back to using HTTP/1.1 for now in order to use TLS.
Reviewed By: singhsrb
Differential Revision: D14015335
fbshipit-source-id: b78197d4cfecf184479162c5b14ba54cbef66ee7
Summary:
To connect to the memcache client it used a path from a loose files data store (`remotefilelogcontentstore._path`), which is a path to the hg cache, but pack files store doesn't have the same field.
I replaced the path with cache path from shallowutil.
Reviewed By: singhsrb, liubov-dmitrieva
Differential Revision: D14020122
fbshipit-source-id: 29bc8a01fc42d43eddc6ae67cf4e41fc552e117c
Summary:
This back ports Yuya's patch from upstream:
changeset: 443029011990c75c533ec0454fd0f5c1060d4690
branch: stable
user: Yuya Nishihara <yuya@tcha.org>
date: Sun, 01 Jul 2018 06:36:53 -0800
encoding: alias cp65001 to utf-8 on Windows
As far as I can tell, cp65001 is the Windows name for UTF-8. I don't know
how different it is from the UTF-8, but Python 3 appears to have introduced
new codec for cp65001, so the alias is enabled only for Python 2.
https://bugs.python.org/issue13216
This patch is untested, but hopefully fixes the following issue.
https://bitbucket.org/tortoisehg/thg/issues/5127/
More context:
Windows 10 Build 17035 (November 2017) introduced a
"Use Unicode UTF-8 for worldwide language support" checkbox, which makes the
"A" flavored APIs use UTF-8 instead of some non-UTF-8 local encoding. With
that checkbox ticked, hg would crash in cmd.exe complaining
"LookupError: unknown encoding: cp65001"
{F150484107}
Reviewed By: phillco
Differential Revision: D14017361
fbshipit-source-id: a648b2a9bc341d796532c6fd2730c083402b3736
Summary:
Change system config entry point to only `/etc/mercurial/system.rc` (unix) and
`\ProgramData\Facebook\Mercurial\system.rc` (Windows) so they won't overlap
with a vanilla Mercurial installation.
Another goal of this change is to make it easier to drop the directory
`%include` feature. So detecting config changes (for example, edenfs wants to
make sure ignore rules are up-to-date) can be made cheaper by just stating
files without `listdir`.
Reviewed By: markbt
Differential Revision: D13875656
fbshipit-source-id: 314c0bf87ff086dec5b88e232edca0133356484e
Summary:
The super python builtin expects a type as an argument, while a string was
passed, this was causing trouble for some users.
Reviewed By: DurhamG
Differential Revision: D14012716
fbshipit-source-id: 6714eb20745428818721ec3df85588a7b4c7ebb7
Summary:
It would be massy if there are 2 chg servers running: one for fb hg, one for
upstream hg, and they share a same socket path.
Change socket path and the commandserver name so fb-hg chg can only talk to
fb-hg chg servers.
Reviewed By: markbt
Differential Revision: D13869319
fbshipit-source-id: f9d42af9bdfc542207f23c536b478fd5ef8d02a0
Summary:
The test is about treating the "default" remote branch specially.
Named branches are going away, therefore delete the test.
Reviewed By: ikostia
Differential Revision: D13976850
fbshipit-source-id: 0283e78c120efd0cf9c91811b823923780cdbfd9
Summary:
The test uses named branch. Since subrepo is also on the list to kill, just
remove the test now.
Differential Revision: D13954459
fbshipit-source-id: 29e8009376081d964441497796fda3490f2137e3
Summary:
The test was about the connection handling between remotefilelog and
fastannotate. It uses named branch to test certain code paths, which
is not trivial to migrate off.
We'll probably rework the protocol used here and maybe do something very
different regarding on linelog with Mononoke. Therefore just remove the
test for now.
Reviewed By: singhsrb
Differential Revision: D13934435
fbshipit-source-id: 9ba0b760ba17efd3ede6de7612e6532f3d13a4f8
Summary:
The test was relying on the ability to merge a "fast-forward" commit, it's no
longer possible without named branches. So the DAG was changed from two linear
commits to "Y"-ish 3 commits.
This makes the test future-proof.
Differential Revision: D13954460
fbshipit-source-id: 944747404843e985f06d18fe0618843dca64630f
Summary: This will be used by scmmemcache to send history data to memcache
Reviewed By: DurhamG
Differential Revision: D13975346
fbshipit-source-id: f41eaf9a4968072dd07efbcd9d539e6293c3fa4f
Summary:
We can now fetch history data stored in memcache and write it to a history
pack.
Reviewed By: DurhamG
Differential Revision: D13975308
fbshipit-source-id: 2196328ad60a55d1e2b39d88d939f434e496837a
Summary:
The initial get_data/set_data only sent the full-text to memcache, which is
just enough for non-LFS data. Let's use Serde to serialize/deserialize the data
that we send to memcache. This will make it simple to add checksuming, or more
metadata to it.
Reviewed By: DurhamG
Differential Revision: D13974714
fbshipit-source-id: 41a235e1d1e8128b14f00b668745f4f9a070a360
Summary:
Similarly to the get_data, we can now read a datapack and send the proper
deltas to memcache. This change is lacking in the same way the get_data is.
Reviewed By: DurhamG
Differential Revision: D13886026
fbshipit-source-id: a00475e89b7e75dbbe9afa9f9d293a686f969a3f
