Jason R. Coombs <jaraco@jaraco.com> [Thu, 20 Apr 2023 16:07:47 -0400] rev 50568
hg: move unreachable code to where it could be reached
Arseniy Alekseyev <aalekseyev@janestreet.com> [Tue, 23 May 2023 01:39:47 +0200] rev 50567
stream-clone: support streamv3 on the cli [hg bundle]
We add support and test for this. The support is still experimental, so the
various name and identifier will eventually need to be renamed when stream-v3
gets out of experimental.
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 23 May 2023 01:28:56 +0200] rev 50566
stream-clone: add the `-exp` prefix to the bundle part
We forget to do so in 58adcabc295f, however this is important to prevent
"current" client to send incompatible version to future client.
Joerg Sonnenberger <joerg@bec.de> [Mon, 21 Feb 2022 14:44:22 +0100] rev 50565
zstd: hack include order to ensure that our zstd.h is found
If the regular Python CFLAGS include directories that already have the
zstd headers available, a different and possible incompatible version
can be picked up otherwise. Sadly, it seems like Python has no easy way
to prefix flags before the rest.
Martin von Zweigbergk <martinvonz@google.com> [Thu, 18 May 2023 17:07:43 -0700] rev 50564
exchange: allow passing no includes/excludes to `pull()`
I would expect that `exchange.pull(includepats=[])` results in an empty list of
include patterns to be passed to the remote, but it doesn't currently because we
check for any truthy value instead of checking specifically for `not None`.
pacien <pacien.trangirard@pacien.net> [Tue, 16 May 2023 12:31:07 +0200] rev 50563
stabletailgraph: add test cases challenging the open merge stack
This adds three more complex test cases with situations requiring tricky
state update in the stack-based iteration (arriving soon).
pacien <pacien.trangirard@pacien.net> [Fri, 21 Apr 2023 14:33:33 +0200] rev 50562
stabletailgraph: naive version of leap computation
This adds a naive reference implementation of the computation of leap and
specific leap sets (described in the code documentation).
The existing tests are enriched accordingly.
pacien <pacien.trangirard@pacien.net> [Fri, 21 Apr 2023 16:19:32 +0200] rev 50561
stabletailgraph: extract _parents util func
pacien <pacien.trangirard@pacien.net> [Mon, 22 May 2023 19:04:05 +0200] rev 50560
stabletailgraph: clarify excl part size computation
pacien <pacien.trangirard@pacien.net> [Fri, 21 Apr 2023 14:32:58 +0200] rev 50559
stabletailgraph: clarify naiveness of current implementation
Both the naive and the actual versions of the algorithms are going to co-exist
for the tests. This makes is clearer that this one is naive.
Pierre-Yves David <pierre-yves.david@octobus.net> [Fri, 19 May 2023 14:49:50 +0200] rev 50558
stream-clone: introduce the notion of an experimental "v3" version
We introduce a new experimental "v3" stream protocol, disabled by default. In
practice the "v3-exp" protocol introduced in this changeset is identical to v2,
but this changeset, lay the groundwork for having a new protocol:
configuration, capability exchange, test coverage, etc.
The actual protocol work will starts in the coming changesets.
Pierre-Yves David <pierre-yves.david@octobus.net> [Sat, 20 May 2023 01:39:13 +0200] rev 50557
stream-clone: check is a compatible protocol can be found
The previous code was explicitly checking if "v2" is listed in the "stream"
bundle2 capability. The new code is checking is there is anything common
between the versions supported client side and server side overlaps.
This prepare the introduction of more stream version than "v2".
Pierre-Yves David <pierre-yves.david@octobus.net> [Sat, 20 May 2023 01:22:49 +0200] rev 50556
stream-clone: bail-out earlier if stream clone is not requested
The `canperformstreamclone` function is bit messy. However it seems clearer to
me to check if a stream-clone have been requested by the client or the server
at all, before checking if a compatible protocol can be negotiated with the
server.
So I am doing some gratuitous movement so reorder conditional.
Pierre-Yves David <pierre-yves.david@octobus.net> [Sat, 20 May 2023 01:19:26 +0200] rev 50555
stream-clone: bail-out earlier if pull is partial
The `canperformstreamclone` function is bit messy. However it seems clearer to
me to process the very generic condition about "can we consider a stream-clone
at all", before checking if a stream-clone is requested and if a compatible
protocol can be negotiated with the server.
So I am doing some gratuitous movement so reorder conditional.
Pierre-Yves David <pierre-yves.david@octobus.net> [Sat, 20 May 2023 01:17:27 +0200] rev 50554
stream-clone: bail-out earlier if destination repo is not empty
The `canperformstreamclone` function is bit messy. However it seems clearer to
me to process the very generic condition about "can we consider a stream-clone
at all", before checking if a stream-clone is requested and if a compatible
protocol can be negotiated with the server.
So I am doing some gratuitous movement so reorder conditional.