run-tests: send the test result after freeing the channel Sending the message about the test being "done" signals to the main thread that a new test can be started. Before this changeset, we sent this signal before freeing the channel, there is room for a race condition where a new test would search for a channel before the old test freed the one it used. This is an example of the failure it would produce: https://foss.heptapod.net/mercurial/mercurial-devel/-/jobs/552404

run-tests: also send a message in the Keyboard interrupt case The next patch will do something equivalent, so lets do the change in an independant changeset first in case we need to bisect something in the future.

revlog: use appropriate format char for int ("i" instead of "I") From https://docs.python.org/3.8/c-api/arg.html#numbers : i (int) [int] Convert a Python integer to a plain C int. I (int) [unsigned int] Convert a Python integer to a C unsigned int, without overflow checking.

revlog: use %d to format int instead of %lu (issue6565) The issue says gcc warns that the data types don't match. I couldn't reproduce the warning locally for some reason, but this patch shouldn't break things. Maybe %lu was simply a copy-paste error from 6b1eae313b2f (https://phab.mercurial-scm.org/D10625).

rhg: correctly handle the case where diffs are encoded relative to nullrev returning a valid entry for nullrev fix chain that delta against nullrev.

test-revlog: test a repository that contains a diff against nullrev We are witnessing a crash in the rust code, so we lets make sure this case is tested.

worker: adapt _blockingreader to work around a python3.8.[0-1] bug (issue6444) Python 3.8.0 is the latest I can load on Ubuntu 18.04, and I regularly hit the TypeError because this function is missing. While it can be avoided by disabling worker usage via config option, that's a bit obscure. I'm limiting the function definition to the narrow range of affected pythons because there were other bugs in this area that were worked around, that I don't fully understand. See the bug report for discussions on why the narrow range, and related commits working around other bugs. Differential Revision: https://phab.mercurial-scm.org/D12627

filelog: show the passed argument on error The error now do more than stating what it need. It also state what it got.

manifest: improve error message in case for tree manifest In the case where tree manifest is not enabled but we still receive an sub directory information for the manifest. The error now inform which sub-directory was passed.

doc: use an absolute path in sys.path The idea and rationale is similar to https://phab.mercurial-scm.org/D12599 (landed as 1b6e381521c5). Differential Revision: https://phab.mercurial-scm.org/D12622

check-py3-compat: use an absolute path in sys.path The idea and rationale is similar to https://phab.mercurial-scm.org/D12599 (landed as 1b6e381521c5). Differential Revision: https://phab.mercurial-scm.org/D12621

branchmap: use a context manager when writing the branchmap This is cleaner and safer. The previous code date from long before we had context manager available.

rhg: align the dirstate v2 writing algorithm with python Use the same algorithm of file append as python does, where we do a manual seek instead of relying on O_APPEND. (see the reasons in the inline comment)

test-dirstate: actually test the append code path in dirstate v2 Apparently it's not sufficient to modify a file to force the dirstate write-out, so the append code path was untested. By removing a file instead of changing we're forcing append to happen.

auto-upgrade: skip the operation if the repository cannot be locked This seems like a fine default behavior for now. If some users wants something more aggressive we can make the behavior configurable in the future. Differential Revision: https://phab.mercurial-scm.org/D12619

auto-upgrade: add a test case where the repository is already locked This show the current behavior when the repository to auto-upgrade is already locked. The current behavior is to abort, which is probably not great. Now that we have a proper test, we can think about the behavior we wants in a later tests. Differential Revision: https://phab.mercurial-scm.org/D12618

wait-on-file: properly wait on any files and symlink This make the utility more useful, for example to wait on a lock file. We also add an explicit -L check since the lock are "weird" symlink. Differential Revision: https://phab.mercurial-scm.org/D12617

debuglock: make the command more useful in non-interactive mode The existing prompt mode simply release the lock immediately in non-interactive mode. That is quite useless in the test so now the non-interactive mode simply wait for a signal. Differential Revision: https://phab.mercurial-scm.org/D12616

auto-upgrade: add a test case with no permission to lock the repository This show the current behavior when the repository is unlockable. The current behavior is to abort, which is probably not great. Now that we have a proper test, we can think about the behavior we want in a later changeset. Differential Revision: https://phab.mercurial-scm.org/D12615

auto-upgrade: introduce a way to auto-upgrade to/from dirstate-v2 This is similar to what we introduced for `share-safe`, but apply to the tracked-hint feature. Differential Revision: https://phab.mercurial-scm.org/D12614

auto-upgrade: introduce a way to auto-upgrade to/from tracked-hint This is similar to what we introduced for `share-safe`, but apply to the tracked-hint feature. Differential Revision: https://phab.mercurial-scm.org/D12613

upgrade: split some logic from UpgradeOperation The logic for automatic-upgrade and the upgrade-repo should be able to use the same code. However that code often need an UpgradeOperation object to function. So we start spliting the Operation into a minimal component that we will be able to reuse outside of the "classic" upgrade path. We will put the base-class to use in the next changeset. Differential Revision: https://phab.mercurial-scm.org/D12612

auto-upgrade: introduce a way to auto-upgrade to/from share-safe This is the first "automatic-upgrade" capability. In the following commits, similar features are coming for other "fast to upgrade" formats. This is different from the `safe-mismatch.source-not-safe` and `safe-mismatch.source-safe` configuration that deal with mismatch between a share and its share-source. Here we are dealing with mismatch between a repository configuration and its actual format. We will need further work for cases were the repository cannot be locked. A basic protection is in place to avoid a infinite loop for now, but it will get proper attention in a later changeset. Differential Revision: https://phab.mercurial-scm.org/D12611

rust: make requirements public These can be used by any client crates (including `rhg`), no need to make them private to the crate. Differential Revision: https://phab.mercurial-scm.org/D12610

clone: use better names for temp files Before this commit, the file names are /tmp/tmpn8smvlr8 After this commit, they are more like /tmp/hg-clone-n8smvlr8/00manifest.ndb3qj52v6, which makes it much clearer what these files correspond to. Differential Revision: https://phab.mercurial-scm.org/D12623

tests: make sure .js files stay in ASCII encoding (issue6559) Differential Revision: https://phab.mercurial-scm.org/D12620

ci: have rust-cargo-test inherit from all This help changing configuration for everything at the same time. This was initially the case before being dropped by mistake in 0ddd5e1f5f67.

copies-sdc: mark upgrade action as changelog only We don't need to recompute the other revlog to add the changelog-v2 feature. This simplify upgrade that use copies-sdc (as shown in the tests).

changelog-v2: mark upgrade action as changelog only We don't need to recompute the others revlog to add the changelog-v2 feature. This does not have much effect in practice as the `copies-sdc` upgrade still triggers the other revlogs. This will be fixed in the next changeset.

copies-sdc: no longer suppress the upgrade output To make sure the upgrade simplification we need are taken into account, we need to see more of the output of `debugupgraderepo`. The --quiet flag simplify the output a lot and globing the `preserved` field mean this output should remains stable across (future) unrelated changes.