corruption: backout changeset 49fd21f32695 (issue6528) inverting the parent is masking copy information leading to bad content being fetched and bad status result. Since 49fd21f32695, exchange can actively swap these parent corrupting existing changesets and triggering the corruption. Data corruption are considered critical so backing this out and doing and unscheduled release seems in order. Differential Revision: https://phab.mercurial-scm.org/D10995

corruption: add a test for issue6528 The initial reproduction script was provided by Charles Chamberlain from Jane Street. Differential Revision: https://phab.mercurial-scm.org/D10996

stream: double check that self.vfs is *not* in the vfsmap The stream clone logic allows for writing any content to any file under various vfs. This is *not* suitable for *vfs*, since writing in `.hg/` directly allow to modify the configuration and is a great and simple gateway for remote code execution. Differential Revision: https://phab.mercurial-scm.org/D10905

dirstate: split dirstatemap in its own file The dirstate file is large enough and the dirstatemap is quite insulated logic already. Differential Revision: https://phab.mercurial-scm.org/D10934

run-tests: stop writing a `python3` symlink pointing to python2 Having `python3` actually pointing to `python2` is bad. So we stop doing so. In addition we need to re-introduce a `python` executable since some of the script really need to be able to say "current python" in their shbang. For example, `hghave` is one of such script. The faulty changes where introduced by c102b704edb5. Differential Revision: https://phab.mercurial-scm.org/D10943

check-code: stop forbidding return code result There is no explication of what is the intend of that check and what is the alternative. I suspect this comes from the transition to the "unified test" format circa 2010. With the non zero return explicitly listed in the output explicit $? checking became Redundant. However there is valid use case for checking $? so I am dropping this check. Differential Revision: https://phab.mercurial-scm.org/D10994

run-tests: avoid an early return To fix the "python3 pointing to python2" we will also need to create a "python" pointer. So we will need to create multiple pointer. So we need to stop using early return. We replace the early return with a loop and a continue, since the next changeset will introduce that loop anyway. Differential Revision: https://phab.mercurial-scm.org/D10942

run-tests: clarify pyexename assignement That assignement is wrong, we clarify that it is wrong before starting to fix it. This will make the fix clearer. Differential Revision: https://phab.mercurial-scm.org/D10941

run-tests: factor appdata out This is a gratuitous readability change. Differential Revision: https://phab.mercurial-scm.org/D10940

run-tests: unroll a non-so-one-liner This is shorter and clearer. Differential Revision: https://phab.mercurial-scm.org/D10939