changegroup: replace changegroup with makechangegroup As part of reducing the number of changegroup creation APIs, let's replace the changegroup function with makechangegroup. This pushes the responsibility of creating the outgoing set to the caller, but that seems like a simple and reasonable concept for the caller to be aware of. Differential Revision: https://phab.mercurial-scm.org/D668

changegroup: delete getlocalchangegroup As part of reducing the number of changegroup creation APIs, let's go ahead and delete this unused function. It appears to have been deprecated in the last release anyway. Differential Revision: https://phab.mercurial-scm.org/D667

changegroup: replace getlocalchangegroupraw with makestream As part of reducing the number of changegroup creation apis, let's replace calls to getlocalchangegroupraw with calls to makestream. Aside from one case of checking if there are no outgoing commits and returning None, this is pretty much a drop in replacement. Differential Revision: https://phab.mercurial-scm.org/D666

changegroup: replace changegroupsubset with makechangegroup As part of getting rid of all the permutations of changegroup creation, let's remove changegroupsubset and call makechangegroup instead. This moves the responsibility of creating the outgoing set to the caller, but that seems like a relatively reasonable unit of functionality for the caller to have to care about (i.e. what commits should be bundled). Differential Revision: https://phab.mercurial-scm.org/D665

changegroup: replace getsubset with makechangegroup The current changegroup APIs are a bit of a mess. Currently you can use getsubsetraw, getsubset, changegroupsubset, getlocalchangegroupraw, getchangegroup, and getlocalchangroup to produce changegroups. This patch is the beginning of a refactor to boil all of that away to just makechangegroup and makestream. The first step adds the new functions and replaces getsubset function with them. Differential Revision: https://phab.mercurial-scm.org/D664

rebase: remove unnecessary '.unfiltered()' calls Now we have a clear centric place to control whether `rbsrt.repo` is unfiltered or not, we can drop `unfiltered()` in other places. Differential Revision: https://phab.mercurial-scm.org/D645

rebase: remove complex unhiding code This is similar to Martin von Zweigbergk's previous patch [1]. Previous patches are adding more `.unfiltered()` to the rebase code. So I wonder: are we playing whack-a-mole regarding on `unfiltered()` in rebase? Thinking about it, I believe most of the rebase code *should* just use an unfiltered repo. The only exception is before we figuring out a `rebasestate`. This patch makes it so. See added comment in code for why that's more reasonable. This would make the code base cleaner (not mangling the `repo` object), faster (no need to invalidate caches), simpler (less LOC), less error-prone (no need to think about what to unhide, ex. should we unhide wdir p2? how about destinations?), and future proof (other code may change visibility in an unexpected way, ex. directaccess may make the destination only visible when it's in "--dest" revset tree). [1]: https://www.mercurial-scm.org/pipermail/mercurial-devel/2017-March/094277.html Differential Revision: https://phab.mercurial-scm.org/D644

rebase: use unfiltered repo when loading state Before this patch, `rebase --abort` may fail to do the cleanup: $ hg rebase --abort rebase aborted (no revision is removed, only broken state is cleared) The added test case makes sure `--abort` works in this case. Differential Revision: https://phab.mercurial-scm.org/D643

rebase: do not crash rebasing merge with a parent having hidden successor The added test will crash with previous code. Differential Revision: https://phab.mercurial-scm.org/D640

changegroup: fix to allow empty manifest parts The current chunk reading algorithm relied on counting the number of empty chunks and comparing it to the number of chunk lists it expected (1 list of files for cg1 and cg2, and 1 list of files + 1 list of trees for cg3). This implicitly assumed that both the changelog part and the manifestlog part were never empty (since them being empty would cause it to count it as one list being done, and screw up the count). In our treemanifest code, the manifest section could be empty, so we need to handle that case. This patches refactors that code to be more explicit about how it counts the expected parts. Differential Revision: https://phab.mercurial-scm.org/D646