Pulkit Goyal <7895pulkit@gmail.com> [Wed, 16 Dec 2020 15:04:17 +0530] rev 46212
actions: introduce function to calculate downgrades
An upgrade operation can also downgrade/remove some format variants. Before this
patch there was no clean way to find out all such variants which will be
removed. This patch adds a function for that.
It will be used in next patch.
Differential Revision: https://phab.mercurial-scm.org/D9618
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 16 Dec 2020 14:55:27 +0530] rev 46211
debugupgraderepo: minor documentation fix
When we specify `--no-changelog --no-manifest --no-filelog` we skip all revlog
optimization instead of all filelog optimization.
Also while I was here, for consistency, I did `optimisation` -> `optimization`
to make it consistent with rest of occurrences.
Note: I am not native speaker and I only changed it because of consistency. I
don't know which one is correct.
Differential Revision: https://phab.mercurial-scm.org/D9617
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 16 Dec 2020 14:16:10 +0530] rev 46210
upgrade: rename actions to upgrade_actions
The `actions` were a list of optimizations and format upgrades which will be
upgraded. This does not include things which will be downgraded.
Let's rename the variable for clarity. It now makes obvious that we don't have
any concrete information on what things are downgraded.
Differential Revision: https://phab.mercurial-scm.org/D9616
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 16 Dec 2020 14:06:24 +0530] rev 46209
upgrade: move optimization addition to determineactions()
The documentation of `determineactions()` mention that it is given a list
returned from `findoptimizations()` however it was not true before this patch.
The code extending actions with optimizations also mentioned about it that this
should be in determineactions.
So let's do what comments at couple of places say.
Differential Revision: https://phab.mercurial-scm.org/D9615
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 16 Dec 2020 12:39:15 +0530] rev 46208
upgrade: drop support for old style optimization names
Old style optimization names like `redeltaparent` were converted into
`re-delta-parent` more than two years ago. The old names were kept around for
sometime because of BC reasons.
Refer 5608b5a6c3231c4ec771171cc3079142c7672be6. The commit states the map is
there for a while and we can drop them as the underlying command is a debug
command.
Differential Revision: https://phab.mercurial-scm.org/D9614
Pulkit Goyal <7895pulkit@gmail.com> [Mon, 14 Dec 2020 16:15:01 +0530] rev 46207
upgrade: add a missing space in status message
Found while reading the code to refactor.
Differential Revision: https://phab.mercurial-scm.org/D9583
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 30 Dec 2020 14:56:34 +0530] rev 46206
actions: rename DEFICIENCY constant to FORMAT_VARIANT
It was not obvious what does deficieny means and every format change can't be a
deficiency. There are some format changes like compression levels, share-safe
which can't be understood at deficiencies.
This patch renames the constant to make things clearer.
Differential Revision: https://phab.mercurial-scm.org/D9664
Pulkit Goyal <7895pulkit@gmail.com> [Mon, 14 Dec 2020 16:03:15 +0530] rev 46205
upgrade: rename finddeficiences() to find_format_upgrades()
It was not obvious what does deficieny means and every format upgrade can't be a
deficiency. There are some format upgrades like compression levels, share-safe
which can't be understood at deficiencies.
A change can be an upgrade or downgrade, however this `finddeficiences()` only
used to find upgrades.
This patch renames the function and related variables to make things more
clearer. The ui message also got improved which is a good thing.
Next patch will rename deficiency in general across the upgrade code.
Differential Revision: https://phab.mercurial-scm.org/D9582
msuozzo@google.com [Sat, 02 Jan 2021 01:48:12 -0500] rev 46204
beautifygraph: change the current commit symbol
The existing "circle with vertical fill" character renders as full width in
many terminals making the beautified symbols have uneven size. Additionally,
the proposed "fisheye" character is the logic 'active' version of the current
merge symbol, "bullseye."
Differential Revision: https://phab.mercurial-scm.org/D9678
Augie Fackler <augie@google.com> [Mon, 11 Jan 2021 14:10:31 -0500] rev 46203
merge with stable
Joerg Sonnenberger <joerg@bec.de> [Mon, 28 Dec 2020 01:21:58 +0100] rev 46202
statichttprepo: explicitly convert error message to str (issue6247)
For Python 2.7, the implicit conversion of the HTTPError instance to
str was good enough. For Python 3.x, this fails later when hitting the
str to bytes conversion logic.
Differential Revision: https://phab.mercurial-scm.org/D9661
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 07 Dec 2020 22:27:43 +0100] rev 46201
debugsetparents: add various warning in the help message
I am tired to see people shooting themself in the foot with this. So lets add
more warning. At that point we should probably rename it or add extra
confirmation flag. This is a debug command anyway we can break BC on it.
Differential Revision: https://phab.mercurial-scm.org/D9534
Joerg Sonnenberger <joerg@bec.de> [Mon, 28 Dec 2020 03:18:16 +0100] rev 46200
largefiles: redo heads interception
The largefiles extension replaces the "heads" wire command and tries to
redirect all uses towards the custom "lheads" wire command. As seen in
issue6384, this doesn't currently work for ssh. Instead of hooking into
the _callstream interface, properly register the command for the peer
instance and monkeypatch the executor to do the redirection. This works
transparently for both all kinds of peers and both for the batch and
non-batch case.
Differential Revision: https://phab.mercurial-scm.org/D9663
Joerg Sonnenberger <joerg@bec.de> [Thu, 07 Jan 2021 11:51:18 +0100] rev 46199
nodemap: match comment to actual code
REV_OFFSET constant is 2, not 10.
Differential Revision: https://phab.mercurial-scm.org/D9688
Dan Villiom Podlaski Christiansen <danchr@gmail.com> [Sun, 20 Dec 2020 15:38:50 +0100] rev 46198
clone: update to active bookmark, if set
This cannot happen during a normal Mercurial clone, but is useful for extensions like hg-git where we know what to check out — the Git HEAD — but do not wish to track it later on using the `@` bookmark.
Test Plan:
I have not included an explicit test, as this is much more easily tested within the `hg-git` test suite.
Differential Revision: https://phab.mercurial-scm.org/D9638