packaging: isolate invocation of WiX to own function Like we did for Inno, we want to split out the building of Mercurial from invoking the packaging tool so that we can introduce an alternate build mechanism. As part of this refactor, there are inconsequential changes to file layouts. Before, some shared files such as the WiX binaries and merge modules would be installed under build/. Now, they are installed under build/wix-*. This is to keep implementation simpler. But it also helps keep build state more isolated. Differential Revision: https://phab.mercurial-scm.org/D8474

packaging: support building Inno installer with PyOxidizer We want to start distributing Mercurial on Python 3 on Windows. PyOxidizer will be our vehicle for achieving that. This commit implements basic support for producing Inno installers using PyOxidizer. While it is an eventual goal of PyOxidizer to produce installers, those features aren't yet implemented. So our strategy for producing Mercurial installers is similar to what we've been doing with py2exe: invoke a build system to produce files then stage those files into a directory so they can be turned into an installer. We had to make significant alterations to the pyoxidizer.bzl config file to get it to produce the files that we desire for a Windows install. This meant differentiating the build targets so we can target Windows specifically. We've added a new module to hgpackaging to deal with interacting with PyOxidizer. It is similar to pyexe: we invoke a build process then copy files to a staging directory. Ideally these extra files would be defined in pyoxidizer.bzl. But I don't think it is worth doing at this time, as PyOxidizer's config files are lacking some features to make this turnkey. The rest of the change is introducing a variant of the Inno installer code that invokes PyOxidizer instead of py2exe. Comparing the Python 2.7 based Inno installers with this one, the following changes were observed: * No lib/*.{pyd, dll} files * No Microsoft.VC90.CRT.manifest * No msvc{m,p,r}90.dll files * python27.dll replaced with python37.dll * Add vcruntime140.dll file The disappearance of the .pyd and .dll files is acceptable, as PyOxidizer has embedded these in hg.exe and loads them from memory. The disappearance of the *90* files is acceptable because those provide the Visual C++ 9 runtime, as required by Python 2.7. Similarly, the appearance of vcruntime140.dll is a requirement of Python 3.7. Differential Revision: https://phab.mercurial-scm.org/D8473

packaging: split Inno installer building from Mercurial building We want to make the logic for producing the installer agnostic about how Mercurial is built to allow for alternate build methods (like PyOxidizer). Differential Revision: https://phab.mercurial-scm.org/D8472

packaging: remove pyoxidizer.bzl from packaging directory We have another version in rust/hgcli that is more modern and is already associated with our Rust CLI project. Differential Revision: https://phab.mercurial-scm.org/D8471

contrib: install PyOxidizer in Linux and Windows environments For Linux, this was trivial. For Windows, we need to teach the powershell script to install Rust as well. This was also pretty straightforward. Differential Revision: https://phab.mercurial-scm.org/D8468

diff: re-establish linear runtime performance The previous method with sum() and list() creates a new list object for every hunk. Then sum() is used to flatten out this sequence of lists. The sum() function is not "lazy", but creates a new list object for every "+" operation and so this code had quadratic runtime behaviour.

copies: fix the changeset based algorithm regarding merge In 99ebde4fec99, we changed the list of files stored into the `files` field. This lead to the changeset centric copy algorithm to break in various merge situation involving merge. Older information could reach the merge through `p1`, and while information from `p2` was strictly fresher, it would get overwritten anyway. We update the situation with more details about which revision introduces rename information. This help use making the right decision in case of merge. We are now running a more comprehensive suite of test with include this kind of situation. The behavior differ slightly from the filelog based in a couple of instance. There is mostly two distinct cases: 1) there are conflicting rename information in a merge (different rename history on each side). In this case the filelog based implementation arbitrarily pick a side based on the file-revision-number. So it depends on a local factor. The changeset centric algorithm will use a deterministic approach, by picking the information coming from the first parent of the merge. This is stable across different clone. 2) rename information related to file that exist in both source and destination. The filelog based implementation do not even try to detect these, however the changeset centric one get them for "free" (it is simpler to detect them than not). The new implementation focus on correctness. Performance improvement will come later. Differential Revision: https://phab.mercurial-scm.org/D8244

merge with stable

rust-chg: clean up excessive indents Differential Revision: https://phab.mercurial-scm.org/D8450

rust-chg: do not terminate tokio runtime until pager exits We no longer need to spawn a task just to keep the pager handle. The pager handle can be held by ChgUiHandler since the handler itself is not consumed and recreated across async calls. Differential Revision: https://phab.mercurial-scm.org/D8449

rust-chg: modernize entry function Finally the entire build passes. There's a bug that run() no longer waits for the spawned pager, which will be fixed by the next patch. Differential Revision: https://phab.mercurial-scm.org/D8448

rust-chg: reimplement locator by using async/await and tokio-0.2 connect_spawned() is rewritten from scratch by using std::process. Before, it would select completion of either connection or server process. New code could be implemented as such, but it's much simpler to occasionally run try_wait() to detect server death. Differential Revision: https://phab.mercurial-scm.org/D8447

rust-chg: reimplement ChgClientExt as ChgClient wrapper ChgClient is no longer an extension trait because: a. Client object is not consumed and recreated in future-0.3 world, which unblocks writing a simple wrapper struct. b. async fn isn't allowed in trait. Overall, the API should become simpler. Differential Revision: https://phab.mercurial-scm.org/D8446

rust-chg: reimplement run_command operation as async function The crafted state machine is no longer needed thanks to async/await. The state machine is basically rewritten as follows: - Ready(..) -> return .. - PollAgain(..) -> run .. and await - Err(..) -> return Err(..) Differential Revision: https://phab.mercurial-scm.org/D8445

rust-chg: reimplement uihandler by using async-trait and tokio-0.2 We no longer have to consume self and arguments. Differential Revision: https://phab.mercurial-scm.org/D8444

rust-chg: have attach_io() simply take reference of AsRawFd object We no longer have to deal with the restriction of the Future type. Before, these file objects couldn't be references and that's the only reason why we had to make stderr an Option<T>. This fixes future type deduction issue of stderr = None, where rustc would complain that T of Option<T> couldn't be deduced. Differential Revision: https://phab.mercurial-scm.org/D8443

rust-status: check for '.hg' regardless of file type (issue6300) '.hg' would show up in `hg status` if were a symlink. Differential Revision: https://phab.mercurial-scm.org/D8461

rust: remove extra empty line This is a gratuitous cleanup. Differential Revision: https://phab.mercurial-scm.org/D8460

upgrade: properly filter action depending on planned work The `determineactions` function filters out deficiency that are not scheduled to be fixed by the target repository configuration. However it only did so for requirement we currently support, letting other actions for unsupported requirement through even if the target repo configuration did not request it. As a result the output of the command was easily polluted by experimental feature with no upgrade support. We rework the code to still filter out requirement based action without the faulty filtering. Differential Revision: https://phab.mercurial-scm.org/D8427

nodemap: skip persistent nodemap warming for revlog not using it Before this patch, the usual checking (especially, inline-ess) were not performed when warming the cache through `hg debugupdatecache`. This is now fixed. Differential Revision: https://phab.mercurial-scm.org/D8408

wait-on-file: adjust the timer counter The wait performed in increment of 0.01 second, but the timer was expressed in second. So we did not wait 20s, we waited 0.2 second. Internally we adjust the counter to be expressed in centisecond. This prevent some flackyness in the test. Differential Revision: https://phab.mercurial-scm.org/D8453

packaging: add docutils as dependency The previous commit revealed that attempting to run `python setup.py build_doc` from the packaging virtualenv failed due to missing docutils package. We didn't notice this before because py2exe Windows packaging appears to use a Python from another virtualenv (which does include docutils) to invoke setup.py. I discovered this as part of implementing packaging outside of that virtualenv environment. Differential Revision: https://phab.mercurial-scm.org/D8470

setup: use sysstr() on process output Otherwise we get a str-bytes mismatch on Python 3 if an error occurs. Differential Revision: https://phab.mercurial-scm.org/D8469

automation: install latest Python versions in Linux Staying up to date. Keeping parity with the Windows environment. Differential Revision: https://phab.mercurial-scm.org/D8467

contrib: update to latest Python 2.7, 3.7, and 3.8 We would ideally update the 3.5 and 3.6 versions as well. But Python didn't generate exe installers for newer versions for some reason. Differential Revision: https://phab.mercurial-scm.org/D8466

automation: always use latest Windows AMI The old AMI isn't available any more. We seem to run into this problem every few months. Amazon (or Microsoft) appears to be removing the old AMIs when they are superseded or something. Let's give up on tracking known images and switch the image selection logic to use the latest published image. Differential Revision: https://phab.mercurial-scm.org/D8465

rust-chg: reimplement attach_io operation as async function In short, MessageLoop<Connection> was redesigned as Protocol<Connection>, and the protocol methods no longer consume self. API changes are briefly documented in the following page: https://docs.rs/tokio-hglib/0.3.0/tokio_hglib/struct.Protocol.html Differential Revision: https://phab.mercurial-scm.org/D8442

rust-chg: upgrade to futures-0.3 based libraries And do some trivial fixes: - BytesMut::put_u32_be() -> put_u32() - tokio_process -> tokio::process, CommandExt -> Command, spawn_async() -> spawn(), stdin() -> stdin - tokio_timer::sleep() -> tokio::time::delay_for() Differential Revision: https://phab.mercurial-scm.org/D8441

rust-chg: exclude futures-dependent modules from build and break things It's impractical to upgrade the codebase incrementally since futures 0.1 and 0.3 APIs are fundamentally different. So this patch temporarily excludes futures-dependent modules from the build. These modules will be upgraded and re-enabled one by one. Differential Revision: https://phab.mercurial-scm.org/D8440

tests: stabilize test-log.t on Windows I think this is because Windows doesn't recognize single quote. Other ssh based clones use this clunky style too. Differential Revision: https://phab.mercurial-scm.org/D8459