test-evolve-bumped: add new test case for merge Before this patch we were not testing our warning message for bumped changesets that are the product of a merge. This patch adds a test case for that.

test-evolve-bumped: improve tests legibility This patch introduces in test-evolve-bumped some of the test helpers that we use in other tests to make the code more legible.

evolve: drop len comparison in prune and fold Computing the length of a smartest may be significantly more expensive than computing it length.

evolve: add an help topic about evolution The text is probably bad, but this have been laying around for quite sometime (with Mercurial core in mind) lets move forward and add this to evolve.

evolve: remove _picknexttroubled It was not used anymore

evolve: warn about every skipped evolution When we could not evolve a revision, we used to silently ignore it. We now inform the user that we ignored something and tell him why (with more or less informative message).

evolve: some style fix

evolve: unify revision handling and rework error message This changesets do a massive refactoring of the way evolve select changesets to be evolve. We now use a two stages approach: 1) read command line argument and produce a list of revision to evolve. 2) evolve all these revisions. This allow a much cleaner, robust and extensible code. In the process the error message issued when there is nothing to evolve have been updated to informs about other troubles in the repository and point at useful option to solve them. The 'update' case is handled independently at the start of the function.

evolve: move the 'update' if '.' is obsolete earlier A big refactoring of the computation of the revision to evolve is coming. This case won't be handled by it, so we extract it earlier.

evolve: skip unstable changesets with multiple successorssets We were previously crashing when encountering them, but we want to be able to solve the other solvable troubles instead of stopping right there.

evolve: allow multiple --rev argument This is standard behavior for Mercurial commands.

test-evolve: fix tests failing on some platform We change from echo to printf to print newlines so that evolve's tests work cross-platforms.

evolve: prevent using --rev and --any together Before this patch it was possible to use evolve --rev --any and it meant evolve --rev. This patch throws an error when using evolve --rev --any.

evolve: add selector for trouble types This patch adds command line flags to select the kind of troubles to solve. Before this patch evolve was solving any kind of troubled. With this patch evolve --all and evolve --rev can be configured to only solve only some type of troubles (divergent, bumped or unstable). To give this choice to the user we add three new command line flags to the evolve command: --bumped, --divergent and --unstable.

evolve: refresh the documentation Multiple behavior change are coming, a perfect timing to refresh the doc before we rework it more.

evolve: check for uncommited change earlier Before this patch we were checking for uncommited change multiple times and quite late for uncommited change. This patch moves the check at the beginning of the evolve method to avoid these shortcomings.

merge with some old heads

merge with some older backport

template: add a troubles keyword This keyword list all evolution troubles that affect a changeset. Evolution troubles can be "unstable", "divergent" and "bumped".

evolve: make fold aware of allowunstable Before this patch, the fold command was performing the same way regardless of the value of experimental.evolution. With this patch if the configuration does not allow unstability, fold won't create instability.

evolve: make prune respect allowunsable Before this patch, the prune command was performing the same way regardless of the value of experimental.evolution. With this patch if the configuration does not allow unstability, prune won't create instability.

merge with stable

evolve: add a hint about how to abort an evolve with conflict The whole evolve --continue is hacky, but the way to abort it have been arcane knowledge, carried from sith lord to its apprentice for centuries. We add a hint to help the use discover it. One day a clean solution will come.

merge with branch closing commit

close 3.1 compat branch

evolve: marks the extension as tested with 4.1 Because it is.

merge with stable

evolve: fix default value for --confirm Before this changes, the default was yes, unlike advertised. The test were actually catching that so I go them updated.

prune: use the bookmark wrapper for 3.4 compatibility

merge with stable

evolve: update readme regarding the preview perf improvement

evolve: move 'allsuccessors()' to smarted usage The "new" smarted can be significantly more efficient than the old was. We convert our revset to take advantage of that.

evolve: move 'successors()' to smarted usage The "new" smarted can be significantly more efficient than the old was. We convert our revset to take advantage of that.

evolve: move 'allprecursors()' to smarted usage The "new" smarted can be significantly more efficient than the old was. We convert our revset to take advantage of that.

evolve: move 'precursors()' to smarted usage The "new" smarted can be significantly more efficient than the old was. We convert our revset to take advantage of that.

evolve: move 'suspended()' to smarted usage The "new" smarted can be significantly more efficient than the old was. We convert our revset to take advantage of that.

evolve: move 'troubled()' to smarted usage The "new" smarted can be significantly more efficient than the old was. We convert our revset to take advantage of that.

evolve: avoid creating changectx object in _allsuccessors The repo lookups and object creations have a significant performance overhead.

evolve: avoid creating changectx object in _successors The repo lookups and object creations have a significant performance overhead.

evolve: avoid creating changectx object in _allprecursors The repo lookups and object creations have a significant performance overhead.

evolve: avoid creating changectx object in _precursors The repo lookups and object creations have a significant performance overhead.

evolve: fix the template keyworkd decorator docstring It was still talking about revsets.

evolve: search divergence within all precursors Before this patches we looking for divergence base in immediate successors only. This was a bug leading the divergence base to be reported missing more than it should.

evolve: make sure we use an unfiltered repo when looking for divergence data Let's be double sure

evolve: make sure we use an unfiltered repo when looking for divergence data Let's be double sure

directaccess: remove import from the no warning list removal of the list previously failed for some reason.

directaccess: add some commands to the directaccess list Before this patch we had a limited list of commands in the directaccess list. This patch adds about 20 commands to that list: - all the read-only commands in core supporting a rev as an argument - 'outgoing': since people can use 'outgoing' to know what is going to be pushed, the output of hg outgoing should be consistent with to the output of hg push and must therefore disallow directaccess since hg push forbids it.

evolve: move test for evolve --rev ordering in a separate file evolve --rev reordering is a complicated enough topic to justify a separate test.

evolve: properly evolve stacked unstable with --rev Before this patch, _singlesuccessor was not returning a revision number for the case where the parent of the argument was not obsolete. This resulted in bug when testing membership with the set of revisions. This patch fixes it and adds a test.

directaccess: disable directaccess for push and serve Before this patch, push and serve were subject to directaccess. This patch makes them throw error when trying to access hidden hashes.

directaccess: change rule from opt-in to opt-out Before this patch we would opt-in commands for direct access and the default filter for new repository was 'visible'. With this patch, the default filter for new repos is 'visible-directaccess-warn'. It means that by default all the commands have directaccess with warnings.

inhibit: move transaction wrapping outside of repo setup Before this patch, transaction wrapping code was done in reposetup. It happened to cause stackoverflows in repos with a lot of subreps. This patch moves the wrapping to extsetup and avoids this problem.

tests: ignores other core output in capability testing This will avoid breaking the test anytime core changes.

evolve: make uncommit respect allowunsable Before this patch, the uncommit command was performing the same way regardless of the value of experimental.evolution. With this patch if the configuration does not allow unstability, uncommit won't create instability.

directaccess: use cached filteredrevs Before this patch we were calling directly repoview.computehidden(repo) to compute the revisions visible with direct access, without going through the caching mechanism for the filtered revisions. There was two issues with that: (1) Performance: We were not leverating the cached values of the 'visible' revs (2) Stability: If there were to be a cache inconsistency with the computation of 'visible' we would crash in the branchmap consistency check partial.validfor. Consider the scenario of rebase with bookmarks: - when we delete a bookmark on an obsolete changeset (like what rebase does when moving the bookmark after rebasing the changesets) - then this changes the value returned by repoview.computehidden(repo) as bookmarks are used as dynamic blockers in repoview.computehidden(repo) - as of now, we don't invalidate the cache in the case of bookmark change - if we have a cached value from before the bookmark change, repoview.filterrevs(repo, 'visible') considers the cached value correct and returns something different than repoview.computehidden(repo) - in turn, if we use repoview.computehidden(repo) in directaccess, the subset relationship is broken and the cache consistency assertion (parial.validfor) fails if branchmap.updatecache is called in this time window This patch leverages the caching infrastructure in place to speed up the computation of the filteredrevs for visible-directaccess-nowarn and visible-directaccess-warn. Incidentally it prevents the bug discussed in (2) from crashing when running a rebase with a bookmark. Note that there still needs to be a fix in core for the case discussed in (2). The test for this side of the fix (not core's fix for (2) is very hard to implement without introducing a lot of dependencies and does not belong here. It is much easier to have the test of the fix for the scenario (2) in core along with the fix.

inhibit: improve performance of transaction wrapping Before this patch, transaction wrapping was the most expensive part of inhibit computation wise. This patch changes the revset that we use in the transaction wrapping to make it ~50x faster to compute: revset #0: obsolete() - hidden() 0) wall 0.000214 comb 0.000000 user 0.000000 sys 0.000000 (best of 11209) vs revset #0: (not hidden()) and obsolete() 0) wall 0.010965 comb 0.010000 user 0.010000 sys 0.000000 (best of 237)

evolve: extract the code computing dependencies in a separate function The code to compute dependencies between unstable changeset can be reused to compute the next reveset. This patch extracts it in its own function to make it reusable.

directaccess: add mechanism to load directaccess after some other extensions directaccess needs to load after some extensions to avoid interfering with them. This patch adds a mechanism to specify what extension directaccess needs to load after.

evolve: add another test for evolve --rev This patch adds one test for three conditions not covered before: - evolve --rev on a stack with obsolete commits without successors - evolve --rev on two sets of stacks in parallel - evolve --rev on a set of stack partially solvable (because all of the commits of the stack are not specified with --rev)

evolve: add a more complex test for evolve --rev The previous tests of evolve --rev would have passed with an ordering of revs by increasing revision numbers. This patch adds a test that would fail if that was what we were doing and therefore tests the implementation better.