perf: drop an unnecessary `pass` Differential Revision: https://phab.mercurial-scm.org/D7743

perf: fix the time measurement for pathcopies relative to p2 PyCharm flagged this as unused. But it seems more likely that it was intended to reset the time for the p2 path. Differential Revision: https://phab.mercurial-scm.org/D7742

githelp: drop unused variable assignments Caught by PyCharm. Differential Revision: https://phab.mercurial-scm.org/D7741

debug: drop unused variable assignments Caught by PyCharm. Differential Revision: https://phab.mercurial-scm.org/D7740

changegroup: drop an unused variable assignment Caught by PyCharm. The variable is assigned again below before it is used. Differential Revision: https://phab.mercurial-scm.org/D7739

bugzilla: drop an unused variable assignment Caught by PyCharm. Differential Revision: https://phab.mercurial-scm.org/D7738

ancestor: drop another unused variable assignment Also caught by PyCharm. The only time this is used is immediately after another assignment below this. Differential Revision: https://phab.mercurial-scm.org/D7737

ancestor: drop an unused local variable assignment Caught by PyCharm. Differential Revision: https://phab.mercurial-scm.org/D7736

cleanup: drop unused import from pywatchman Caught by PyCharm. Differential Revision: https://phab.mercurial-scm.org/D7735

narrow: move `testedwith` after module imports PyCharm flagged this as a PEP 8 violation, for module level imports not being at the top of the file. Differential Revision: https://phab.mercurial-scm.org/D7734

match: don't util.normpath() cwd The problem here is that `util.normpath()` calls `util.pconvert()`, which switches to Unix style separators. That results in two test failures like this since 5685ce2ea3bf: --- c:/Users/Matt/hg/tests/test-globalopts.t +++ c:/Users/Matt/hg/tests/test-globalopts.t.err @@ -89,7 +89,7 @@ [255] $ hg -R b ann a/a abort: a/a not under root '$TESTTMP/b' - (consider using '--cwd b') + (consider using '--cwd ..\$TESTTMP\b') [255] $ hg log abort: no repository found in '$TESTTMP' (.hg not found)! ERROR: test-globalopts.t output changed Martin originally had `os.path.normpath()` (which *would* work here too), but changed it during review. He didn't remember why he thought any form is needed here. Most uses simply pass '' or `repo.getcwd()`, so these should generally be in local format anyway. It seems better if `cwd` and `root` use consistent styles here. Differential Revision: https://phab.mercurial-scm.org/D7725

tests: convert the `root` arg of matchmod.match() to local path separators This fixes tests that broke with 8b1a9ba375e5, complaining that "X not under root /repo". The vast majority of real uses are to pass `repo.root`, which is normalized by `wdirvfs.base` being set to the result of `os.path.realpath()`. Failure to convert looks like this: --- c:/Users/Matt/hg/tests/test-match.py.out +++ c:/Users/Matt/hg/tests/test-match.py.err @@ -0,0 +1,48 @@ +ERROR: testVisitchildrensetGlob (__main__.IncludeMatcherTests) + +Traceback (most recent call last): + File "c:\Users\Matt\hg\tests\test-match.py", line 180, in testVisitchildrensetGlob + m = matchmod.match(b'/repo', b'', include=[b'glob:dir/z*']) + File "c:\Users\Matt\hg\mercurial\match.py", line 271, in match + kindpats = normalize(include, b'glob', root, cwd, auditor, warn) + File "c:\Users\Matt\hg\mercurial\match.py", line 322, in _donormalize + pat = pathutil.canonpath(root, cwd, pat, auditor=auditor) + File "c:\Users\Matt\hg\mercurial\pathutil.py", line 251, in canonpath + _(b"%s not under root '%s'") % (myname, root), hint=hint +Abort: dir/z* not under root '/repo' +ERROR: testVisitdirGlob (__main__.IncludeMatcherTests) Differential Revision: https://phab.mercurial-scm.org/D7724

commitablectx: fix the default phase The code was ignoring the config option. Since the config option will control the actual phase of the committed result, we better take it in account early. Differential Revision: https://phab.mercurial-scm.org/D7726

fix: fix handling of merge commits by using overlayworkingctx Most of this code was conceptually copied from what rebase does, with one small difference: hgext.rebaserev.rebase uses branchmerge=True, while I had to use branchmerge=False, or else it got really confused about updating to the same revision in some situations. I believe that the difference is that rebase is always dealing with *some* form of update - it never gets to mergemod.update if the source and destination are the same, while we can encounter that situation with fix. This may imply that this code has some issues with named branches that should be investigated. Differential Revision: https://phab.mercurial-scm.org/D7703

rust-index: add a struct wrapping the C index Implementing the full index logic in one go is journey larger than we would like. To achieve a smoother transition, we start with a simple Rust wrapper that delegates allwork to the current C implementation. Once we will have a fully working index object in Rust, we can easily start using more and more Rust Code with it. The object in this patch is functional and tested. However, multiple of the currently existing rust (in the `hg-cpython` crate) requires a `Graph`. Right now we build this `Graph` (as cindex::Index) using the C index passed as a PyObject. They will have to be updated to be made compatible. Differential Revision: https://phab.mercurial-scm.org/D7655

rust-index: add a `inner` method to the Index struct We are about to introduce a Rust wrapper to the C index. It will be exposed as a Python object. As a start all call will be forwarded to an internal index object (as used now). For this forwarding to works, we need to access that index object (from the C extensions). We adds a method doing to on the existing `Index` rust struct. Differential Revision: https://phab.mercurial-scm.org/D7654

revlog: made C Capsule an array of function pointers Although it's perfectly valid to put a function pointer in a capsule, as we've been doing since the start of rust/hg-cpython, an array of function pointers has several advantages: - it can hold several functions. That's our main motivation here. We plan to expose index_length() and index_node(), which will be needed for a Rust implementation of nodemap. - it could also have data - (probably minor in the case of Mercurial) proper support for architectures for which data and code pointers don't have the same size. Differential Revision: https://phab.mercurial-scm.org/D7543

lfs: fix a discrepancy with a function wanting a filelog, but calling it rlog This conceptually broke in 1541e1a8e87d when the filelog isa revlog relationship was changed to containment of the revlog. It was made more obvious in 62a532045e71 and related API simplification. It's resolved in favor of passing a revlog because the revlog verification code doesn't have a reference to a filelog. Differential Revision: https://phab.mercurial-scm.org/D7711

revlog: split the content verification of a node into a separate method This will be used by LFS to tune what is skipped. In the future, this could also be used by LFS to indicate which nodes tagged with `skipread` are simply in need of a blob fetch, so that they can be done in a batch later. (Currently, `skipread` also indicates censored data and errors.) Additionally, it could be used to cache the sha1 hash value for each blob so that large blobs don't need to be re-read and re-hashed if they are used by multiple nodes. Differential Revision: https://phab.mercurial-scm.org/D7710

verify: update comment to say that lfs doesn't need fulltext to check renames The reason is that `filelog.renamed()` indirectly calls `filelog.revision()`, which is what accesses the full text. However, LFS wraps `filelog.renamed()` and completely handles the case where an LFS blob is in play by using rawdata. I've got a test to demonstrate that this is the case, and prevent regressions. But the `skipread` flag is set on all lfs revisions when using `--no-lfs`, regardless of whether or not the blobs are local. Just above this, that flag is consulted, causing the rename checks to be skipped. That will need to be loosened up first. Differential Revision: https://phab.mercurial-scm.org/D7709

resourceutil: use `from importlib import resources` Without this patch, we get the following error from when trying to run hg with PyOxidizer: module 'importlib' has no attribute 'resources' I don't know what why that happens, but `from importlib import resources` is the form I would prefer anyway, so let's use that now that the impoort-checker has been fixed. Differential Revision: https://phab.mercurial-scm.org/D7701

import-checker: allow all absolute imports of stdlib modules Before this patch, we didn't allow imports like from importlib import resources (That's the reason I used `import importlib.resources` in D7629.) I think that form is still an absolute import, so I don't think we forbade on purpose. Differential Revision: https://phab.mercurial-scm.org/D7700

help: drop a reference to Windows 9x Differential Revision: https://phab.mercurial-scm.org/D7694

help: clarify that the Windows registry key for hgrc files is systemwide Since there's no version or path info here to distinguish between installations, it is effectively systemwide (unless splitting hairs about the WoW64 registry redirection). Differential Revision: https://phab.mercurial-scm.org/D7693

windows: add a global equivalent to /etc/mercurial for *.rc processing This follows the Unix model of processing this directory immediately after <internal>/*.rc, and prior to the installation relative files. Since the Unix processing supports both a directory and a file (the former overriding the latter), and since %HOME% supports both `*.ini` and `.hgrc` (again, the former overriding the latter), this does too. The Unix file doesn't have a `.` prefix, so it's not used here either. Note that this is the opposite order of processing the exe relative paths. But since it's in agreement with Unix, %HOME% and %USERPROFILE%, it seems reasonable to ignore that. Maybe we can change that and take a BC, because that's something the installer should be controlling, and I can't imagine people having both paths *and* conflicting settings. Differential Revision: https://phab.mercurial-scm.org/D7692

match: normalize `cwd` early By having cwd in absolute form, we won't have to adjust it when passing it to subrepo matchers. This will matter for a coming patch. Differential Revision: https://phab.mercurial-scm.org/D7650

match: make sure `root` argument is always an absolute path (API) The `root` argument should already be an absolute path, but we had tests that passed a relative path. This patch fixes up the tests and adds an assertion. This assumes that `os.path.isabs('/repo')` will be `True` on all platforms we care to run tests on. Augie tested for me that it does work on Windows, so that's good enough for me. Differential Revision: https://phab.mercurial-scm.org/D7649

tests: show that fileset patterns don't work with `fix` when not in repo root Differential Revision: https://phab.mercurial-scm.org/D7569

py3: force bytestr conversion of "reason" in scmutil.callcatch() For instance, reason may be an InvalidURL as shown in added test.

tests-pure: fixing test-parseindex2 This is a followup to D7603 (49fa0b31ee1d) which broke the tests for pure Python implementation. There are two divergences between pure and C implementations: - the pure implementation would accept only -1 as slice end, whereas C accepts both -1 and len(index) - in pure Python, `headrevs` is provided by revlog.py, not by the index. Differential Revision: https://phab.mercurial-scm.org/D7707

rust-index: add a function to convert PyObject index for hg-core Function in hg-core need something implementing the `Graph` trait. Right now, the `hg-cpython` entry points directly turn the PyObject passed as argument into a `cindex::Index`. However, if we start having the option to use an Index in Rust, we need to dispatch between the different possible PyObject we could receive. So move the "duplicate" call into a unified function. When time come. It will be easy to update the logic of all interface when the time come. Differential Revision: https://phab.mercurial-scm.org/D7653

test: extract some generic data and utility from test-rust-ancestor.py We will reuse this for more tests related to revlog index. In pratice this series of changesets add an index implementation provided from Rust and we want to be able to test it. Differential Revision: https://phab.mercurial-scm.org/D7652

patch: fix header.__repr__() to not have `b''` prefixes in file names Spotted by Denis Laxalde. Differential Revision: https://phab.mercurial-scm.org/D7695

cmdutil: allow native string as input to check_at_most_one_arg() We seem to always convert **opts args to use bytes as keys early on in Mercurial core, but I'm not sure we have good reason to do that, and not all extensions do that. It's therefore helpful to be able to pass in a native string to check_at_most_one_arg(). Differential Revision: https://phab.mercurial-scm.org/D7699

cmdutil: return underscore-separate name from check_at_most_one_arg() As noticed by Yuya, when I changed the function (during review) to work with underscore-separated names as input, I forgot to make sure the returned name was also underscore-separated. We don't have any cases where it matters yet, but it should still clearly be fixed. Instead of converting the hyphen-separated value we already have in `previous`, I'm changing it so we convert to the underscore-separated values to be hyphen-separated only when we need to display them. This will also help a coming change where we allow the inputs to native strings instead only bytes. Differential Revision: https://phab.mercurial-scm.org/D7698

rebase: restore i18n of a hint message As noticed by Yuya, I lost the _() call in 71fee4564410 (rebase: use rewriteutil.precheck() instead of reimplementing it, 2019-12-18). Differential Revision: https://phab.mercurial-scm.org/D7697

status: make unresolved files always be in the morestatus structured output We don't know the status of those files, only that they're unresolved, so we don't output the status for those - any code parsing this will have to be tolerant to that. Differential Revision: https://phab.mercurial-scm.org/D7668

status: make morestatus call out unresolved conflicts after update Differential Revision: https://phab.mercurial-scm.org/D7704

status: extract active-merge state for reuse Differential Revision: https://phab.mercurial-scm.org/D7667

fix: use rewriteutil.precheck() instead of reimplementing it Differential Revision: https://phab.mercurial-scm.org/D7687

split: use rewriteutil.precheck() instead of reimplementing it As you can see from the patch, I disagree with most of the comment saying that `rewriteutil.precheck()` is not worthwhile. Differential Revision: https://phab.mercurial-scm.org/D7686

rebase: use rewriteutil.precheck() instead of reimplementing it After this patch, there's still another place in `rebase.py`, in the `--stop` code path, that reimplements `rewriteutil.precheck()`. I couldn't fix that place because it `rewriteutil.precheck()` checks that there is only one dirstate parent, which fails because we have two parents at that point. I think it's incorrect that rebase leaves the user with two parents during conflicts, but changing that is way out of scope for this series. Differential Revision: https://phab.mercurial-scm.org/D7685

histedit: use rewriteutil.precheck() instead of reimplementing it We have had this nice utility since 490df753894d (rewriteutil: add a precheck function to check if revs can be rewritten, 2017-11-24). We got two callers soon thereafter, in 98f97eb20597 (rewriteutil: use precheck() in uncommit and amend commands, 2017-11-28). It's about time we use it in other places. rewriteutil.precheck() looks unfortunate for i18n (it seems to rely on an English verb for the action making sense in other languages), but hopefully not bad enough that we should avoid using it. Differential Revision: https://phab.mercurial-scm.org/D7684

rebase: inline single-use variables passed to _definedestmap() Differential Revision: https://phab.mercurial-scm.org/D7646

rebase: use cmdutil.check_at_most_one_arg() for action+revision Differential Revision: https://phab.mercurial-scm.org/D7645

rebase: use cmdutil.check_incompatible_arguments() for --auto-orphans Differential Revision: https://phab.mercurial-scm.org/D7644

rebase: use cmdutil.check_at_most_one_arg() for -b/-s/-r Differential Revision: https://phab.mercurial-scm.org/D7643

rebase: inline single-use `dryrun` and `confirm` variables Differential Revision: https://phab.mercurial-scm.org/D7642

rebase: use cmdutil.check_incompatible_arguments() for action+confirm/dryrun Differential Revision: https://phab.mercurial-scm.org/D7661

rebase: use cmdutil.check_at_most_one_arg() for --confirm/--dry-run I've also updated the helper to work with the hyphenated --dry-run option. Differential Revision: https://phab.mercurial-scm.org/D7641

windows: factor the hgrc directory scan into a function Differential Revision: https://phab.mercurial-scm.org/D7691

windows: don't return early from building the hgrc search path This will minimize the changes needed to add other sources. Differential Revision: https://phab.mercurial-scm.org/D7690

windows: clarify a comment about the hgrc search path The exe relative hgrc.d and the registry paths are not mutually exclusive. Differential Revision: https://phab.mercurial-scm.org/D7689

windows: drop detection of Windows 95/98/ME Support was removed in python 2.6. Differential Revision: https://phab.mercurial-scm.org/D7688

examples: add an example configuration for go source files Tested by timeless. Differential Revision: https://phab.mercurial-scm.org/D7683

util: move common proxyobserver attributes to the base class Fixes the following pytype warnings: line 791, in _writedata: No attribute 'logdata' on baseproxyobserver [attribute-error] line 792, in _writedata: No attribute 'logdataapis' on baseproxyobserver [attribute-error] line 793, in _writedata: No attribute 'fh' on baseproxyobserver [attribute-error] line 794, in _writedata: No attribute 'fh' on baseproxyobserver [attribute-error] line 799, in _writedata: No attribute 'logdataapis' on baseproxyobserver [attribute-error] line 800, in _writedata: No attribute 'fh' on baseproxyobserver [attribute-error] line 802, in _writedata: No attribute 'fh' on baseproxyobserver [attribute-error] line 803, in _writedata: No attribute 'name' on baseproxyobserver [attribute-error] line 805, in _writedata: No attribute 'fh' on baseproxyobserver [attribute-error] line 809, in _writedata: No attribute 'logdataapis' on baseproxyobserver [attribute-error] line 810, in _writedata: No attribute 'fh' on baseproxyobserver [attribute-error] line 814, in _writedata: No attribute 'fh' on baseproxyobserver [attribute-error] line 815, in _writedata: No attribute 'name' on baseproxyobserver [attribute-error] line 817, in _writedata: No attribute 'fh' on baseproxyobserver [attribute-error] Differential Revision: https://phab.mercurial-scm.org/D7675

config: drop debug messages saying where config was read from `hg config --debug` includes lines like this: set config by: $EDITOR but also lines like this: $EDITOR: ui.editor=emacs -nw The `set config by` messages don't seem to provide much additional information over what we get from the `$EDITOR:`-type message. I could imagine wanting to see which values got overriden by a later entry, but that information is already not present. So let's just remove the first type of output. My next patch would otherwise amplify the redundant output (there would be one `set config by` for each line in `mergetools.rc`). Differential Revision: https://phab.mercurial-scm.org/D7627

rcutil: don't check if defaultrc/ is a directory -- we know it is `mercurial/defaultrc/` is a directory both in the Mercurial repo and once installed on a target platform. The directory was created in c4ce077588d0 (config: introduce "built-in" default configuration settings in default.d, 2014-09-04). That commit has some more information, but it still doesn't seem to say that `defaultrc/` (then called `default.d/`) could be a file. Perhaps the check was there to allow you to run the same code on an older install/repo? Differential Revision: https://phab.mercurial-scm.org/D7624

rust-matchers: add support for `exactmatcher` in `dirstate.status` `exactmatcher` is the name in the Python implementation and corresponds to `FileMatcher` in Rust. Differential Revision: https://phab.mercurial-scm.org/D7531

rust-dirstate-status: update bridge for new rust version of `dirstate.status` Differential Revision: https://phab.mercurial-scm.org/D7530

rust-dirstate-status: add `walk_explicit` implementation, use `Matcher` trait This is the first time we actually use the `Matcher` trait, still for a small subset of all matchers defined in Python. While I haven't yet actually measured the performance of this, I have tried to avoid any unnecessary allocations. This forces the use of heavy lifetimes annotations which I am not sure we can simplify, although I would be happy to be proven wrong. Differential Revision: https://phab.mercurial-scm.org/D7529

rust-matchers: add `FileMatcher` implementation Mercurial defines an `exactmatcher`, I find `FileMatcher` to be clearer, but am not opposed to using the old name. This change also switched the order of `assert_eq` arguments as it is clearer that way for most people. Differential Revision: https://phab.mercurial-scm.org/D7528

exchange: ensure all outgoing subrepo references are present before pushing We've run into occasional problems with people committing a repo, and then amending or rebasing in the subrepo. That makes it so that the revision in the parent can't be checked out, and the problem gets propagated on push. Mercurial already tries to defend against this sort of dangling reference by pushing *all* subrepo revisions first. This reuses the checks that trigger warnings in `hg verify` to bail on the push unless using `--force`. I thought about putting this on the server side, but at that point, all of the data has been transferred, only to bail out. Additionally, SCM Manager hosts subrepos in a location that isn't nested in the parent, so normal subrepo code would complain that the subrepo is missing when run on the server. Because the push command pushes subrepos before calling this exchange code, a subrepo will be pushed before the parent is verified. Not great, but no dangling references are exchanged, so it solves the problem. This code isn't in the loop that pushes the subrepos because: 1) the list of outgoing revisions is needed to limit the scope of the check 2) the loop only accesses the current revision, and therefore can miss subrepos that were dropped in previous commits 3) this code is called when pushing a subrepo, so the protection is recursive I'm not sure if there's a cheap check for the list of files in the outgoing bundle. If there is, that would provide a fast path to bypass this check for people not using subrepos (or if no subrepo changes were made). There's probably also room for verifying other references like tags. But since that doesn't break checkouts, it's much less of a problem. Differential Revision: https://phab.mercurial-scm.org/D7616

procutil: try and avoid angering CoreFoundation on macOS We've seen failures like this: objc[57662]: +[__NSCFConstantString initialize] may have been in progress in another thread when fork() was called. objc[57662]: +[__NSCFConstantString initialize] may have been in progress in another thread when fork() was called. We cannot safely call it or ignore it in the fork() child process. Crashing instead. Set a breakpoint on objc_initializeAfterForkError to debug. I think this is due to forking off some background processes during `hg update` or similar. I don't have any conclusive proof this is the fork() call that's to blame, but it's the most likely one since the regular `hg update` codepath uses the other fork() invocation (via workers) and we don't get this report from non-Google macOS users. Ugh. Differential Revision: https://phab.mercurial-scm.org/D7615