profiling: flush stdout before writing profile to stderr On py3, stdout and stderr appear to be buffered and this causes my command's output to be intermixed with the profiling output. Differential Revision: https://phab.mercurial-scm.org/D8024

rust: re-format with nightly rustfmt This fixes test-check-rust-format.t. Differential Revision: https://phab.mercurial-scm.org/D8025

tests: stablize test-rename-merge1.t on Windows This goes with d7622fdec3b5. Differential Revision: https://phab.mercurial-scm.org/D8036

rust-cpython: make sure PySharedRef::borrow_mut() never panics Since it returns a Result, it shouldn't panic depending on where the borrowing fails. PySharedRef::borrow_mut() will be renamed to try_borrow_mut() by the next patch.

rust-cpython: remove useless wrappers from PyLeaked, just move by map() This series prepares for migrating to the upstreamed version of PySharedRef. I found this last batch wasn't queued while rewriting the callers. While Option<T> was historically needed, it shouldn't be required anymore. I wasn't aware that each filed can be just moved.

rust-node: avoid meaningless read at the end of odd prefix This should be heavily factored out by the CPU branch predictor anyway. Differential Revision: https://phab.mercurial-scm.org/D8019

rust-nodemap: generic NodeTreeVisitor This iterator will help avoid code duplication when we'll implement `insert()`, in which we will need to traverse the node tree, and to remember the visited blocks. The structured iterator item will allow different usages from `lookup()` and the upcoming `insert()`. Differential Revision: https://phab.mercurial-scm.org/D7794

rust-nodemap: mutable NodeTree data structure Thanks to the previously indexing abstraction, the only difference in the lookup algorithm is that we don't need the special case for an empty NodeTree any more. We've considered making the mutable root an `Option<Block>`, but that leads to unpleasant checks and `unwrap()` unless we abstract it as typestate patterns (`NodeTree<Immutable>` and `NodeTree<Mutated>`) which seem exaggerated in that case. The initial copy of the root block is a very minor performance penalty, given that it typically occurs just once per transaction. Differential Revision: https://phab.mercurial-scm.org/D7793

rust-nodemap: abstracting the indexing In the forthcoming mutable implementation, we'll have to visit node trees that are more complex than a single slice, although the algorithm will still be expressed in simple indexing terms. We still refrain using `#[inline]` indications as being premature optimizations, but we strongly hope the compiler will indeed inline most of the glue. Differential Revision: https://phab.mercurial-scm.org/D7792

rust-nodemap: NodeMap trait with simplest implementation We're defining here only a small part of the immutable methods it will have at the end. This is so we can focus in the following changesets on the needed abstractions for a mutable append-only serializable version. The first implementor exposes the actual lookup algorithm in its simplest form. It will have to be expanded to account for the missing methods, and the special cases related to NULL_NODE. Differential Revision: https://phab.mercurial-scm.org/D7791

rust-node: handling binary Node prefix Parallel to the inner signatures of the nodetree functions in revlog.c, we'll have to handle prefixes of `Node` in binary form. Another motivation is that it allows to convert from full Node references to `NodePrefixRef` without copy. This is expected to be by far the most common case in practice. There's a slight complication due to the fact that we'll be sometimes interested in prefixes with an odd number of hexadecimal digits, which translates in binary form by a last byte in which only the highest weight 4 bits are considered. This is totally transparent for callers and could be revised once we have proper means to measure performance. The C implementation does the same, passing the length in nybbles as function arguments. Because Rust byte slices already have a length, we carry the even/odd informaton as a boolean, to avoid introducing logical redundancies and the related potential inconsistency bugs. There are a few candidates for inlining here, but we refrain from such premature optimizations, letting the compiler decide. Differential Revision: https://phab.mercurial-scm.org/D7790

rust-revlog: a trait for the revlog index As explained in the doc comment, this is the minimum needed for our immediate concern, which is to implement a nodemap in Rust. The trait will be later implemented in `hg-cpython` by the index Python object implemented in C, thanks to exposition of the corresponding functions as a capsule. The `None` return cases in `node()` match what the `index_node()` C function does. Differential Revision: https://phab.mercurial-scm.org/D7789

pathauditor: drop a redundant call to bytes.lower() `_lowerclean(s)` calls `s.lower()`, so we don't need to do that before calling it. Differential Revision: https://phab.mercurial-scm.org/D8001

merge: replace a repo.lookup('.') by more typical repo['.'].node() The `repo.lookup('.')` form comes from b3311e26f94f (merge: fix --preview to show all nodes that will be merged (issue2043)., 2010-02-15). I don't know why that commit changed from `repo['.']`, but I don't think there's any reason to do that. Note that performance should not be a reason (anymore?), because repo.lookup() is implemented by first creating a context object. Differential Revision: https://phab.mercurial-scm.org/D7998

merge: drop now-unused "abort" argument from hg.merge() Differential Revision: https://phab.mercurial-scm.org/D7997

merge: don't auto-pick destination with `hg merge 'wdir()'` If the user doesn't specify a commit to merge with, we'll have `node==None` in `commands.merge()`. We'll then try to find a good commit to merge with. However, if the user, for some strange reason, runs `hg merge 'wdir()'`, we'll also have `node==None` and we'll do that same. That's clearly not the intent, so let's not do that. It turns out we'd instead crash on that command after this patch, so I added special handling of it too. Differential Revision: https://phab.mercurial-scm.org/D7996

merge: call hg.abortmerge() directly and return early It's seem really weird to go through a lot of unrelated code before we call `hg.merge(..., abort=True)` when we can just call `hg.abortmerge()` and return early. Differential Revision: https://phab.mercurial-scm.org/D7995

merge: check that there are no conflicts after --abort Same idea as in abcc82bf0717 (clean: check that there are no conflicts after, 2020-01-24). We should reuse more code here, but that will come later. Differential Revision: https://phab.mercurial-scm.org/D7994

merge: use check_incompatible_arguments() for --abort Differential Revision: https://phab.mercurial-scm.org/D7993

rebase: move some variables after an error cases where they're not needed Differential Revision: https://phab.mercurial-scm.org/D7905

rebase: clarify a little by calculating a set in Python instead of in revset By calculating the set in Python, we can give it a name, which helps readability. Differential Revision: https://phab.mercurial-scm.org/D7904

merge: avoid a negation in the definition of updatedirstate We only use `partial` in one place: the definition of `updatedirstate`. Let's simplify that a little. Differential Revision: https://phab.mercurial-scm.org/D7900

merge: move definition of `partial` closer to where it's used Differential Revision: https://phab.mercurial-scm.org/D7983

copies: extract function for finding directory renames The directory rename code is logically quite isolated, so it makes sense to make it physically isolated as well. Differential Revision: https://phab.mercurial-scm.org/D7977

copies: avoid calculating debug-only stuff without --debug `renamedeleteset` and `divergeset` is only used with `repo.ui.debugflag`, so let's avoid calculating them otherwise. While at it, I also added a `del renamedeleteset` for consistency. Differential Revision: https://phab.mercurial-scm.org/D7976

copies: move early return in mergecopies() earlier It wasn't obvious that the early return happened only when there are no copies. That is the case, however, because if `fullcopy` is empty, then so is `copies1` and `copies2`, and then so is `inversecopies1` and `inversecopies2`, and then so is `allsources`, and then so is `copy`, `diverge` and `renamedelete`. By moving the early return earlier, we also avoid calculating the set of added files from the base to each side. Differential Revision: https://phab.mercurial-scm.org/D7975

tests: test merge of renames of different sources to same target This is a really obscure scenario, but let's still have it tested so we know when it changes. Differential Revision: https://phab.mercurial-scm.org/D7985

clean: check that there are no conflicts after As noted by Pulkit, there should never be any conflicts after doing a clean update, so `hg.clean()` should never return `True`. Let's check that assertion instead to clarify the code. The callers will now get a `None` instead of a `False` returned, but that should be fine (both result in a 0 exit status). Differential Revision: https://phab.mercurial-scm.org/D7984

progress: delete deprecated ui.progress() Differential Revision: https://phab.mercurial-scm.org/D7991

rust-dependencies: update rayon This is just to make sure we use the latest version and also makes it easier to peruse the docs. Differential Revision: https://phab.mercurial-scm.org/D7926

merge with stable

merge: define updatedirstate a little earlier and reuse it Differential Revision: https://phab.mercurial-scm.org/D7899

merge: don't call update hook when using in-memory context I'm pretty sure many hook implementors will assume that they can inspect the working copy and/or dirstate parents when the hook is called, so I don't think we should call the hook when using an in-memory context. The new behavior matches that of the preupdate hook. Differential Revision: https://phab.mercurial-scm.org/D7898

merge with stable

packaging: add configparser to inno requirements file This dependency is missing and pip complains about it in strict hashing mode. How this was missed, I have no clue. Differential Revision: https://phab.mercurial-scm.org/D7973

python-zstandard: blacken at 80 characters I made this change upstream and it will make it into the next release of python-zstandard. I figured I'd send it Mercurial's way because it will allow us to drop this directory from the black exclusion list. # skip-blame blackening Differential Revision: https://phab.mercurial-scm.org/D7937

tests: move non-collapse test out of test-rebase-collapse The test case was added in 76630fbbf4fa (test-rebase-collapse: Add test for rebase regression introduced in 12309c09d19a, 2012-01-23). I think `hg rebase --collapse` was involved in either the regression or in the fix that caused the regression, but the test that was added doesn't use `--collapse`, so it doesn't seem to belong in test-rebase-collapse.t. The test case is about copies, so I moved it to test-rebase-rename.t. Differential Revision: https://phab.mercurial-scm.org/D7968

debugcommands: add Python implementation to debuginstall This seems like a useful detail to print. Differential Revision: https://phab.mercurial-scm.org/D7979

run-tests: remove --py3-warnings This Python 2 only mode was to help Python 2 alert when doing things not supported on Python 3. Now that we have test coverage with Python 3, I don't think we need it. Differential Revision: https://phab.mercurial-scm.org/D7978

rust-node: binary Node ID and conversion utilities The choice of type makes sure that a `Node` has the exact wanted size. We'll use a different type for prefixes. Added dependency: hexadecimal conversion relies on the `hex` crate. The fact that sooner or later Mercurial is going to need to change its hash sizes has been taken strongly in consideration: - the hash length is a constant, but that is not directly exposed to callers. Changing the value of that constant is the only thing to do to change the hash length (even in unit tests) - the code could be adapted to support several sizes of hashes, if that turned out to be useful. To that effect, only the size of a given `Node` is exposed in the public API. - callers not involved in initial computation, I/O and FFI are able to operate without a priori assumptions on the hash size. The traits `FromHex` and `ToHex` have not been directly implemented, so that the doc-comments explaining these restrictions would stay really visible in `cargo doc` Differential Revision: https://phab.mercurial-scm.org/D7788

rust-nodemap: building blocks for nodetree structures This is similar to `nodetreenode` in `revlog.c`. We give it a higher level feeling for ease of handling in Rust context and provide tools for tests and debugging. The encoding choice is dictated by our ultimate goal in this series, that is to make an append-only persistent version of `nodetree`: the 0th Block must be adressed from other Blocks. Differential Revision: https://phab.mercurial-scm.org/D7787

lfs: move the initialization of the upload request into the try block This (almost) guarantees that the file is closed in the case of an exception. The one hole is if the `seek(SEEK_END)`/`tell()`/`seek(0)` sequence fails. But that's going to go away when subclassing `httpconnection.httpsendfile` to fix the worker problem, so I'm not going to worry too much. (And that class appears to have the same problem.) Differential Revision: https://phab.mercurial-scm.org/D7959

lfs: drop an unnecessary r'' prefix No longer necessary since the source transformer was removed. # skip-blame for changing string prefixes Differential Revision: https://phab.mercurial-scm.org/D7958

lfs: explicitly close the file handle for the blob being uploaded The previous code relied on reading the blob fully to close it. The obvious problem is if an error occurs before that point. But there is also a problem when using workers where the data may need to be re-read, which can't happen once it is closed. This eliminates the surprising behavior before attempting to fix the worker problem. Differential Revision: https://phab.mercurial-scm.org/D7957

lfs: drop the unused progressbar code in the `filewithprogress` class This has been unused since f98fac24b757, which added worker based transfers for concurrency, shifting the progressbar maintenance to the single thread waiting on the worker to complete. Since the name no longer fits, rename the class. Differential Revision: https://phab.mercurial-scm.org/D7956

rust-filepatterns: remove bridge code for filepatterns-related functions These functions will be used internally by `hg-core` without needed to be exposed to Python. Differential Revision: https://phab.mercurial-scm.org/D7868

rust-utils: add Rust implementation of Python's "os.path.splitdrive" I also wrote the NT version although I didn't mean to at first, so I thought I would keep it, so that any further effort to get the Rust code working on Windows is a little easier. Differential Revision: https://phab.mercurial-scm.org/D7864

setup: link osutil.so to libsocket on Solaris/illumos (issue6299)

discovery: avoid wrongly saying there are nothing to pull We can get in a situation where a revision passed through `hg pull --rev REV` are available on the server, but not a descendant of the advertised server heads. For example the server could lying be during heads advertisement, to hide some pull request. Or obsolete/hidden content could be explicitly pulled. So in this case the lookup associated to `REV` returned successfully, but the normal discovery will find all advertised heads already known locally. This flip a special boolean `anyinc` that will prevent any fetch attempt, preventing `REV` to be pulled over. We add three line of code to detect this case and make sure a pull actually happens. My main target is to make some third party extensions happy (I expect the associated test to move upstream with the extension). However this fix already make some of the `infinitepush` test happier.

Added signature for changeset 8fca7e8449a8

Added tag 5.3.2 for changeset 8fca7e8449a8

histedit: add missing b prefix to a string If i18n is disabled (such as via HGPLAIN=1), `_()` doesn't convert from str to bytes, so this raises a TypeError on py3. Differential Revision: https://phab.mercurial-scm.org/D8354

py3: require values in changelog extras to be bytes I don't know what happened here because b436059c1cca (py3: use pycompat.bytestr() on extra values because it can be int, 2019-02-05) came about b44a47214122 (py3: use string for "close" value in commit extras, 2018-02-11). Whatever happened, we shouldn't need to convert the values to bytes now. It's better to not convert because that might cover up bugs where someone sets a unicode value in the extras and that works until the unicode value happens to contain non-ascii (at which point it will fail because `bytestr()` expects its argument to be ascii if it's unicode). Differential Revision: https://phab.mercurial-scm.org/D8332

py3: make setup.py's hgcommand() consistently return bytes Before this patch, it returned unicode when the command failed. That made e.g. `make local PYTHON=python3` fail on an obsolete commit. Differential Revision: https://phab.mercurial-scm.org/D8331

darwin: use vim, not vi, to avoid data-loss inducing posix behavior Apple's version of vim, available at opensource.apple.com/release/macos-1015.html (for Catalina, but this behavior has been there for a while) has several tweaks from the version of vim from vim.org. Most of these tweaks appear to be for "Unix2003" compatibility. One of the tweaks is that if any ex command raises an error, the entire process will (when you exit, possibly minutes/hours later) also exit non-zero. Ex commands are things like `:foo`. Luckily, they only enabled this if vim was executed (via a symlink or copying the binary) as `vi` or `ex`. If you start it as `vim`, it doesn't have this behavior, so let's do that. To see this in action, run the following two commands on macOS: ``` $ vi -c ':unknown' -c ':qa' ; echo $? 1 $ vim -c ':unknown' -c ':qa' ; echo $? 0 ``` We don't want to start ignoring non-zero return types from the editor because that will mean you can't use `:cquit` to intentionally exit 1 (which, shows up as 2 if you combine an ex command error and a cquit, but only a 1 if you just use cquit, so we can't differentiate between the two statuses). Since we can't differentiate, we have to assume that all non-zero exit codes are intentional and an indication of the user's desire to not continue with whatever we're doing. If this was a complicated `hg split` or `hg histedit`, this is especially disastrous :( Differential Revision: https://phab.mercurial-scm.org/D8321

cext: move variable declaration to the top of the block for C89 support Not sure if we still care about C89 in general, but MSVC requires this style too. Differential Revision: https://phab.mercurial-scm.org/D8304

byteify-string: resolve symlink before byteifying Otherwise the script turns symlinks into regular files.

cext-index: propagate inline_scan error in `index_deref` Before this change, revlog index corruption could be silently ignored in some situation. Differential Revision: https://phab.mercurial-scm.org/D8276

heptapod-ci: fix test paths in the listing file Now what we run the test from the root, we need to list test name from the root. Differential Revision: https://phab.mercurial-scm.org/D8275

hg-core: add a compilation error if trying to compile outside of Linux For now, we can only provide support for Linux in `hg-core`, so we have to be explicit about it in case anyone wonders why their Dirstate is suddenly broken, or why the crate does not compile (on Windows for example). Differential Revision: https://phab.mercurial-scm.org/D8246