branchmap: stop writing cache for uncommitted data If we are about to write the branch while a transaction is active. we delay that write. After the transaction is closed, we flush all the write we delayed (unless they have been written in between). Differential Revision: https://phab.mercurial-scm.org/D12128

transaction: add a way to know a transaction has been finalized This will be useful to fix the timing of the branchmap on disk caching. Differential Revision: https://phab.mercurial-scm.org/D12127

branchmap: Add a test about writing branchmap and aborted transaction It turn out that we update the branchmap on disk whenever we recompute it… including in a middle of a transaction. This means writing the new branchmap too early (the changeset are not visible yet) and worse… it mean writing branchmap for transaction we rollback. so we introduce a test to highlight the issue (and prevent it to regress in the future). Differential Revision: https://phab.mercurial-scm.org/D12126

dirstate-v2: rename the configuration to enable the format The rename of the old experimental name was overlooked before the 6.0 release. We rename everything to use the new name (and keep the released name as an alias for compatibility). Differential Revision: https://phab.mercurial-scm.org/D12129

stream-clone: filter possible missing requirements using all supported one The `supportedformat` requirements is missing some important requirements and it seems better to filter out with all requirements we know, not just an "arbitrary" subset. The `supportedformat` set is lacking some important requirements (for example `revlog-compression-zstd`). This is getting fixed on default (for Mercurial 6.1) However, fixing that in 6.1 means the stream requirements sent over the wire will contains more items. And if we don't apply this fix on older version, they might end up complaining about lacking support for feature they actually support for years. This patch does not fix the deeper problem (advertised stream requirement lacking some of them), but focus on the trivial part : Lets use the full set of supported requirement for looking for unsupported ones. This patch should be simple to backport to older version of Mercurial and packager should be encouraged to do so. This is a graft of d9017df70135 from default. Differential Revision: https://phab.mercurial-scm.org/D12091

helptext: add missing newline to Rust helptext This makes it so the list renders properly. Differential Revision: https://phab.mercurial-scm.org/D12007

unamend: error out when running on merge commit It's better to error out than to produce a bad commit. We do that same in `hg uncommit`. I haven't looked into how much work it is to make them work instead. Differential Revision: https://phab.mercurial-scm.org/D12087

tests: demonstrate how `hg unamend` fails on merge commits When `hg unamend` is run on a merge commit, it seems that it loses the changes from the second parent. Differential Revision: https://phab.mercurial-scm.org/D12086

revlog: fix a bug where transaction can be aborted partially Fix a repo corruption bug caused by a partial transaction rollback. Differential Revision: https://phab.mercurial-scm.org/D12009

revlog: demonstrate a bug where transaction can be aborted partially Differential Revision: https://phab.mercurial-scm.org/D12008

rust: fix build errors on darwin I'm not all _that_ versed in Rust, but I think the root cause is that some constants are u16 rather than u32 on Darwin. I checked that the code still compiles on the latest Ubuntu. Differential Revision: https://phab.mercurial-scm.org/D11955

subrepo: make -S work again on Windows for incoming/outgoing to remote repos While it worked fine for the top level repo, the remote path for the subrepo got mangled to something like "https://server/prefix\repo\subrepo", which I've seen result in both a 400 and a 404, depending on the server. We need to `normpath` at least the `subpath` because of "http://user:***@localhost:$HGPORT/main/../sub" in `test-subrepo-relative-path.t`. Keep the `os.path` flavor for handling filesystem based remote repos, since this string is also displayed. This is one case where the automatic substitution of '\' for '/' and rematching done by the test runner is unfortunate- I don't see how to write a test to catch this. Differential Revision: https://phab.mercurial-scm.org/D11971

Added signature for changeset 799fdf4cca80

Added tag 6.0.1 for changeset 799fdf4cca80

docs: update Rust readme with a mention of `rhg` Differential Revision: https://phab.mercurial-scm.org/D11811

docs: fix typos and wording for the persistent-nodemap feature Differential Revision: https://phab.mercurial-scm.org/D11810

docs: add missing dirstate-v2 documentation A user trying out the dirstate-v2 feature without the Rust extensions is currently faced with an error message suggesting this help section, which does not exist. Differential Revision: https://phab.mercurial-scm.org/D11809

docs: add documentation about Rust This makes the features more discoverable for users. Differential Revision: https://phab.mercurial-scm.org/D11808

heptapod-ci: make Windows jobs manual again The dance continues, we make the Windows tests optional once again since we don't have Windows runners available anymore. We're working on getting new ones in a more stable fashion, hopefully soon. Differential Revision: https://phab.mercurial-scm.org/D11957

tests: allow for variation in zstd output as seen on s390x and powerpc

sparse: lock the store when updating requirements config Differential Revision: https://phab.mercurial-scm.org/D11817

sparse: demonstrate a bug when used with safe-share sparse: lock the store when updating requirements config Differential Revision: https://phab.mercurial-scm.org/D11766

hghave: fix the check for suid on platforms lacking support The mac tests were raising an AttributeError without the default arg. Differential Revision: https://phab.mercurial-scm.org/D11906

upgrade: byteify a few error messages These were flagged by pytype (which is currently disabled on this file due to another failure). Differential Revision: https://phab.mercurial-scm.org/D11905

dirstate-v2: fix upgrade on an empty repository This used to crash as the dirstate file does not exist in this case. Differential Revision: https://phab.mercurial-scm.org/D11866

dirstate-v2: test upgrade without rust too There is no reason to gate this anymore. Differential Revision: https://phab.mercurial-scm.org/D11865

upgrade: don't use `dd status=none` in the test This breaks on Windows, so lets use `>2 /dev/null` like in the other tests. Differential Revision: https://phab.mercurial-scm.org/D11873

test-upgrade: narrow the `debugformat` call when upgrading to dirstate-v2 This will make the test less noisy when new requirement are added. Differential Revision: https://phab.mercurial-scm.org/D11864