evolve: fix the code flow pattern of solving obswdir par and troubled revs Now we will go to _handlenotrouble() (which prints messages about no revs to solve) only when there is no troubled revs and working dir parent is not obsolete. This change also saves us from an issue which was about looking into the revset (smartset contains troubled revs to solve) when a rev from the revset gets hidden. This happens in the case when our wdir parent is obsolete. After resolving obswdir parent we were looking into the revset to check if there is any troubled revs to solve but we should have performed this check before performing the obswdir resolution. Changes in test file reflect this fixed behaviour.

evolve: refactor for consistent behavior of evolve when wdp is obsolete This patch make sure that when working directory parent is obsolete `hg evolve` and `hg evolve --all` don't behave differently.

evolve: backout 3027005c42c3 to reintroduce a bug for right fix This patch backout 3027005c42c3 as it was accepted by mistake while it was being "in-discussion" state.

pick: register pickstate as an unfinished state This way pickstate file will indicate that unfinished pick command needs to be dealt with (--continue or --abort) before modifying the repo. Otherwise it would be e.g. possible to commit during an interrupted pick and that's not expected.

pick: rename variable for unfinishedstates

pick: actually delete pickstate if --abort is given Makes pick to be, uh, actually aborted.

evolve: orphans that evolve into nothing don't need successors (issue5967) When continuing to solve an orphan that created no changes (i.e. clean wdir), _completeorphan() used to create an obsmarker that said that the result of that orphan evolution is the currently checked out changeset. That's not a correct obsmarker, because all of the orphan's changes were dropped and so it had no effect on the currently checked out changeset. This is an issue that has only existed when --continu'ing evolve, that's why the fix touches _completeorphan(), but not _solveunstable(). This fix is adapted from a similar "if node is None" block in _finalizerelocate().

tests: demonstrate obsmarker creation after discarding conflicting changes Continued evolve creates an incorrect obsmarker that says 2 is a successor of 1. It's incorrect because 1 was dropped as it created no changes to commit (after conflict resolution that discarded its changes). If evolve does the same thing in one go (e.g. just by using --tool :local and without subsequent need to continue) the obsmarker is correct.

pick: remove transaction on the whole command (issue6037) At its core, pick is a pretty straightforward and well-behaving command, it uses functions already in core hg, it checks that wdir is clean and that changeset to pick is not public, it checks if there happen to be merge conflicts and can be --continue'd later, etc. It is very similar to graft in core (it also uses mergemod.graft function), but it obsoletes the original changeset. However, graft does not experience this incorrect behavior from issue 6037. What happens in the test case for this issue when we pick a revision that touches both "a" and "b": mergemod.graft() takes the original changeset and tries to apply it to the wdir, which results in "b" being marked as newly added and ready to be committed, "a" updated with the new content and being marked as modified, but "a" also has conflicts. Pick correctly notices this and saves its state before asking for user intervention. So far so good. However, when the command raises InterventionRequired to print a user-facing message and exit while being wrapped in repo.transaction() context manager, the latter partially undoes what mergemod.graft() did: it unmarks "b" as added. And when user continues pick, "b" is therefore not tracked and is not included in the resulting commit. The transaction is not useful here, because it doesn't touch wdir (it's still dirty), it doesn't remove pickstate (and other commands will refuse to work until pick --abort or --continue), it just makes "b" untracked. The solution is to use repo.transaction() only to wrap code that writes data to hg store in the final stages of the command after all checks have passed and is not expected to fail on trivial cases like merge conflicts. For example, committing the picked changeset. But since pick uses repo.commit() for that, and because that function already uses a transaction, wrapping it in another transaction doesn't make sense.

tests: demonstrate hg pick forgetting files after conflicts This test currently passes to show that pick is behaving incorrectly.