revset: stub to add extra data to baseset for better inspection We sometimes construct a baseset from filtering result. In that case, a baseset can provide more precise information how it is constructed.

revset: add inspection data to all filter() calls This is useful for debugging revset construction.

revset: add extra data to filteredset for better inspection A filteredset is heavily used, but it cannot provide a printable information how given set is filtered because a condition is an arbitrary callable object. This patch adds an optional "condrepr" object that is used only by repr(). To minimize the maintaining/runtime overhead of "condrepr", its type is overloaded as follows: type example -------- --------------------------------- tuple ('<not %r>', other) str '<branch closed>' callable lambda: '<branch %r>' % sorted(b) object other

zeroconf: replace reduce+add with itertools.chain

zeroconf: replace has_key with in

zeroconf: compare singleton using is

zeroconf: remove camelcase in identifiers

setup: switch to with open as We're leaving the modulepolicy bit alone, because it's being rewritten in the next commit.

win32mbcs: use absolute_import

notify: use absolute_import

patchbomb: use absolute_import

convert: __init__ use absolute_import

convert: cvs use absolute_import

convert: transport use absolute_import

convert: bzr use absolute_import

convert: common use absolute_import

convert: convcmd use absolute_import

convert: subversion use absolute_import

blackbox: guard against recursion from dirty check

tests: divorce blackbox test from test-dispatch.py I used test-dispatch.py to demonstrate what would happen if a log file changed from being readonly to writable, by having it replace a directory (proxy for readonly/not-writable) with a log file in between transactions of a running python process (proxy for Mercurial). This commit makes it easier for people to follow what the test is doing, by creating a real file that people can read.

tests: test-dispatch use print_function

tests: test-dispatch use absolute_import

templater: ignore orig/rej files If your mercurial/templates/ directory is dirty, then the template system would otherwise import duplicate templates from the .orig files and potentially try to parse .rej files. Since editing/reverting these templates isn't an unexpected action, and since they're in .hgignore, it's best that the template system know to skip them."

resolve: when pats do not match, hint about path: Suggest a command that would probably work.

shelve: changes getting opts values by get method When shelve is used by another extension that doesn't provide all necessary values in opts shelve raises KeyError exception. This patch fixes this by getting values from opts dictionary with get method.

contrib: remove references to 2to3 The custom porting fixers are removed. A comment related to 2to3 has been removed from the import checker. After this patch, no references to 2to3 remain.

check-code: remove redundant import style check We have a dedicated tool that checks for import conventions. Remove a redundant and less powerful check.

setup: remove support for 2to3 We want to run unaltered source on multiple Python versions. We won't be using 2to3 for Python 3 support.

run-tests: remove 2to3 support Our goal is to have 1 code base that works on 2.6, 2.7, and 3.5+. 2to3 won't be used. Stop passing it to setup.py.

histedit: adds hint how to reorder changesets at editor (issue3766)

revset: replace predicate by revsetpredicate of registrar To make all built-in predicates be known to hggettext, loading built-in predicates by loadpredicate() should be placed before fixing i18nfunctions but after all of predicate decorating.

revset: replace extpredicate by revsetpredicate of registrar This patch consists of changes below (these can't be applied separately). - replace revset.extpredicate by registrar.revsetpredicate in extensions - remove setup() on an instance named as revsetpredicate in uisetup()/extsetup() of each extensions registrar.revsetpredicate doesn't have setup() API. - put new entry for revsetpredicate into extraloaders in dispatch This causes implicit loading predicate functions at loading extension. This loading mechanism requires that an extension has an instance named as revsetpredicate, and this is reason why largefiles/__init__.py is also changed in this patch. Before this patch, test-revset.t tests that all decorated revset predicates are loaded by explicit setup() at once ("all or nothing"). Now, test-revset.t tests that any revset predicate isn't loaded at failure of loading extension, because loading itself is executed by dispatch and it can't be controlled on extension side.

registrar: define revsetpredicate to decorate revset predicate revsetpredicate is used to replace revset.predicate and revset.extpredicate in subsequent patches. This patch also adds loadpredicate() to revset, because this combination helps to figure out how the name of safe predicate is put into safesymbols. This patch still uses safesymbols set to examine whether the predicate corresponded to the 'name' is safe from DoS attack or not, because just setting func._safe property needs changes below for such examination. before: name in revset.safesymbols after: getattr(revset.symbols.get(name, None), '_safe', False) "automatic registration" described in help doc of revsetpredicate class will be achieved by the subsequent patch, which lists loadpredicate() up in dispatch.extraloaders.

registrar: introduce new class for registration to replace funcregistrar _funcregistrarbase differs from funcregistrar in points below: - every code paths should use same class derived from _funcregistrarbase to register functions in a same category funcregistrar expects (3rd party) extensions to use (a class derived from) delayregistrar. - actual extra setup should be executed in another function For example, marking revset predicate as "safe" is executed in a class derived from _funcregistrarbase, but putting name of "safe" predicate into safesymbols is executed in another function for it. funcregistrar expects derived classes to do so. New class is named as module private one, because code paths, which register functions, should use not it directly but one derived from it.

dispatch: make loading extra information from extension extensible This patch makes loading extra information from extension module at dispatching extensible. Factoring 'loadcmdtable()' into commands.py is a part of generalization of loading extra information. This extensibility assumes registration of new function like below, for example: - revset predicate - fileset predicate - template keyword - template filter - template function - internal merge tool - web command This patch requires not loader function itself but container module and the name of it, because listing loader function directly up implies actual loading module of it, even if it isn't used at runtime (for example, extensions don't always define revset predicate)

obsolete: fix n^2 marker computation behavior Previously, if you ran obsolete.createmarkers with a bunch of markers that did not have successors (like when you do a prune), it encountered a n^2 computation behavior because the loop would read the changelog (to get ctx.parents()), then add a marker, in a loop. Adding a marker invalidated the computehidden cache, and reading the changelog recomputed it. This resulted in pruning 150 commits taking 150+ seconds in a large repo. The fix is to break the reading part of the loop to be separate from the writing part.

bdiff: (pure) support array.array arrays (issue5130)

mq: restrict generated patch name to 75 characters (issue5117) Super long first line in description lead to very long file name that windows is unhappy about. We restrict the name to 75 char to avoid the issue. 75 seems fine and leave some extra room for '__#' suffix in case of conflict. I does not seems worthwhile to add a dedicated config option to configure the length. It can be done in the future if there is an actual user demand for it.

files: don't recurse into subrepos without a path or -S (issue5127) The 'm.always()' check was needed for when a path to 'sub1' is given, and 'sub1' contains a subrepo itself. But that also caused the automatic recursion when no path was given. Instead, force -S when printing a subrepo if the subpath is an exact match (which will unconditionally recurse once in the nested subrepo).

parsers: optimize filtered headrevs logic The old native head revs logic would iterate over every node, starting from 0, and check if every node was filtered (by testing it against the filteredrevs python set). On large repos with hundreds of thousands of commits, this could take 150ms. This new logic iterates over the nodes in reverse order, and skips the filtered check if we've seen an unfiltered child of the node. This saves approximately a bagillion filteredrevs set checks, which shaves the time down from 150ms to 20ms during every branch cache write.

destutil: choose non-closed branch head at first (BC) Before this patch, destupdate() returns the tipmost (descendant) branch head regardless of closed or not. But updating to closed branch head isn't reasonable for ordinary workflow, because: - "hg heads" doesn't show closed heads (= updated parent itself) by default - subsequent committing on it re-opens closed branch even if inactivation of closed head is needed, update destination isn't it, because it should be merged into to another branch in such case. This patch chooses non-closed descendant branch head as default update destination at first. If all descendant branch heads are closed, destupdate() returns the tipmost closed branch head. For simplicity, this patch chooses adding _destupdatebranchfallback() instead largely changing _destupdatebranch(). This patch changes not only normal lookup code path, but also the "no default branch" code path, for consistency.

formatter: make labels work with templated output To describe the bug this fix is addressing, one can do ``$ hg status -T "{label('red', path)}\n" --color=debug`` and observe that the label is not applied before my fix and applied with it.

hghave: improve description of Hypothesis

purge: use absolute_import

record: use absolute_import

relink: use absolute_import

schemas: use absolute_import

shelve: use absolute_import

strip: use absolute_import

graphmod: augment the graph to include more information about the edges The walker knows when an edge leads to a direct parent, a grandparent (skipping revisions not part of the revset) and parents that are missing altogether (neither it nor a grandparent is in the revset). Add this information to the parents sequence yielded.

graphmod: refactor state handling Move ASCII graph state to a dictionary, to clarify what is being tracked. Move the 'seen' state (tracking currently active edges) into this structure.

templater: move label() function from color extension ui.label() is no-op by default, so we can just call ui.label() by label() template function no matter if the color is enabled or not.

templater: make label() take unknown symbol as color literal Instead of the mapping hack introduced by b775a2029e8d, this patch changes the way how a label symbol is evaluated. This is still hackish, but should be more predictable in that it doesn't depend on the known color effects. This change is intended to eliminate the reference to color._effects so that color.templatelabel() can be merged with templater.label().

convert: monotone use absolute_import

convert: p4 use absolute_import

convert: hg use absolute_import

convert: cvsps use absolute_import

convert: darcs use absolute_import

convert: filemap use absolute_import

convert: gnuarch use absolute_import

convert: git use absolute_import

branchmap: check node against changelog instead of repo Testing 'node in repo' requires constructing a changectx, which is a little expensive. Testing 'repo.changelog.hasnode(node)' is notably faster. This saves 10-20ms off of every command, when testing a few thousand nodes from the branch cache. I considered changing the implementation of localrepository.__contains__ so every place would benefit from the change, but since localrepository.__contains__ uses changectx to check if the commit exists, it means it supports a wider range of possible inputs (like revs, hashes, '.', etc), so it seemed unnecessarily risky.

changegroup: clear progress callback after changelog processing The progress callback is replaced by one for manifests after changelog processing is done, but let's not depend on manifests replacing the value and instead explicitly clear it.

tests: adjust for code move in Hypothesis 2.0.0 It appears that the Settings class was renamed to settings, and because of this, the settings module was renamed to configuration. https://github.com/DRMacIver/hypothesis/commit/a0e663b

changegroup: progress for added files is not measured in "chunks" The "prog" class cg1unpacker.apply() has the unit set to "chunks". This is not correct for files, where the file itself is the unit. The unit is not usually printed, which is probably why this has not been fixed yet. It can be show with e.g. "--config progress.format='topic number unit'".

changegroup: exclude submanifests from manifest progress The progress callback for manifests is cleared outside of _unpackmanifests(), which means it will remain in effect while pulling subdirectory manifests when using treemanifests. Since the total number of revisions used for the progress is the number of changesets, the total number of treemanifest revisions is usually larger than that. One effect of this is that the ETA is negative. It's hard to estimate the number of subdirectory revisions, so let's just exclude them from progress for now.

commit: block amend while histedit is in progress (issue4800)

chg: limit reconnect attempts Some users may have hg as a wrapper script which sets sensitive environment variables (like setting up virtualenv). This will make chg redirect forever because the environment variables are never considered up to date. This patch adds a limit (10) for reconnect attempts and warn the user with a possible solution if the limit is exceeded.

chg: use validate to make sure the server is up to date This patch uses the newly added validate method to make sure the server has loaded the up-to-date config and extensions. If the server cannot validate itself, the client will receive instructions and follow them to try to reach another server that is more likely to validate itself. The instructions can be a redirect (connect to another server address) and/or an unlink (stops an out-dated server).

chg: implement validate in hgclient This patch implements the corresponding validate method in hgclient. It will return instruction strings as is without taking any real action.

hgclient: use absolute_import and print_function

fixpax: use absolute_import and print_function

debugcmdserver: use absolute_import and print_function

check-config: use absolute_import and print_function

casesmash: use absolute_import

chgserver: implement validate command validate will load the repo config and check if the server has up-to-date config to continue serve the client. In case it does not, the server will send instructions to the client about what to do next, including to retry with a different address or to unlink an outdated socket file to stop an old server.

templater: handle exception when applying map operator to non-iterable object Before this, "{noniterable % template}" raised an exception. This tries to provide a better indication for the common case, where a left-hand-side expression is a keyword.

templater: factor out thin helper that evaluates argument as string This function is trivial, but it can make it clear that args[i] is a (func, data) pair.

color: make label() template function stringify "expr" argument first Since label() should return a string (or a thunk to be evaluated to a string), this change is okay. This helps porting to evalstring() helper. See the next patch for details.

templater: fix shortest() to evaluate int argument and handle error

templater: fix pad() to evaluate int argument and handle error

templater: drop redundant type conversion when evaluating integer argument A function argument may be an integer. In this case, it isn't necessary to convert a value to string and back to integer. Because an argument may be an arbitrary object (e.g. date tuple), TypeError should be caught as well.

templater: factor out function that evaluates argument as integer We have more bare int()s that may raise ValueError or TypeError. This function will be used to fix them.

chgserver: use basename for socket symlink Previously we use full path and the symlink may point to outside (unsafe) world if the directory is moved. This patch fixes it by only linking to basename of the target. Therefore the symbolic link and socket files will always stay in the same directory.

patch: when importing from email, RFC2047-decode From/Subject headers Reported at https://bugs.debian.org/737498

histedit: reword message when a changeset produces no changes There are various ways to use histedit such that an item in the list of things to perform will not result in a change relative to the previous repository state. When that happens, histedit does not keep the commit/message. This changes the note to try to explain to the user that it will not be present in their history.

tests: replace cat.py with cat in test-histedit-fold-non-commute.t

tests: Solaris grep doesn't add a trailing newline when it's missing The bad-extension tests emits a list of not-loaded extensions, and pipes that output through grep. On Solaris, the test-output gets "(no-eol)" appended because although the message has no trailing newline, GNU grep adds it. If we simply add the newline to the message, the problem goes away for both versions of grep.

tests: Solaris cmp complains about empty files, even with -s When you compare an empty file, such as /dev/null, with a non-empty file, Solaris cmp complains on stderr with "cmp: EOF on /dev/null", even if the -s argument is present. GNU cmp makes the complaint, but silences it with -s. We can change the pdiff utility to simply redirect stderr to /dev/null so that we don't have to worry about this difference in the test files.

tests: Solaris sed does not support "\n" meaning newline in the RHS of s/// The blackbox test rewrites a copy of test-dispatch.py on the fly, and adds a couple of lines with the s/// command. GNU sed supports the use of the \n escape to represent a newline, but not Solaris sed. Using a literal newline, prefixed by a backslash, works with both versions of the utility.

tests: Solaris cp doesn't support the -T option The treemanifest tests use the -T option to cp in order to ensure that the two directories named on the commandline are treated as peers, rather than the usual behavior when the final argument is a directory. GNU cp has this option, but other implementations may not. Thankfully, there's no pressing reason to use it. We can simply copy the contents of the first directory into the target directory, since we know that the target directory already exists.

templater: make date() use helper function to evaluate argument A date argument should never be a generator, but using evalfuncarg() should be good for consistency.

templater: fix revset() to evaluate format arguments eagerly See the previous patch for why. This patch also removes redundant list() construction from a list.

templater: fix ifcontains() to evaluate items argument eagerly See the previous patch for why. An "items" argument may be a string, a generator, or an arbitrary container object.

templater: fix get() to evaluate arguments eagerly If a key is constructed from a template expression, it may be a generator. In that case, a key have to be stringified. A dictarg should never be a generator, but this patch also changes it to call evalfuncarg() for consistency.

import-checker: report local with stdlib late warning Without this, developers have to figure it out on their own

tests: update test output for test written on stable branch (issue5104) The changed test lines were added in d493d64757eb, which landed on stable. When this changeset merged to the default branch, the test failed because a4692267bc2d (on default but not stable) changed the order of working directory update when performing a share-based clone from pooled storage. The changes in this patch are similar to test changes in a4692267bc2d.

chgserver: pass hashstate and base server address to chgcmdserver In order to detect a hash change from a request handler, chg must know the original hashstate. It also needs the base server address to figure out redirect addresses.

chg: drop progress.assume-tty config It was necessary to go through progress.uisetup() to set up the progressui wrapper. Since the progress extension has got into the core, progress.assume-tty is no longer necessary.

chgserver: mangle server address to include confighash Before this patch, chgserver will use the address provided by the client. The new design is one server per confighash. This patch appends "-$confighash" to the address the client provides. To maintain the compatibility and make sure the client can connect to the server, a symbolic link is created at the original address pointing to the new address. The address is intentionally mangled at the server, instead of being pre- calculated by some other process (eg. a previous server). In this way, we can avoid file system race conditions.

chgserver: update docs Update the docstring to reflect the latest changes

merge with stable

commands: use absolute_import All mercurial.* modules are now using absolute_import \o/

cmdutil: use absolute_import Now that @command doesn't write back into commands when it is being executed during the loading of commands.py itself, we are unblocked from converting cmdutil to absolute_import.

keyword: use absolute_import

pager: use absolute_import

pager: add tests We've never had tests for pager, and I want to start experimenting with ways to clean up the implementation and make it a bit more graceful.

tests: flag Windows specific lines about background closing as optional

run-tests: defer leftover (?) cleanup until after all output is exhausted Previously, after matching a single line, any contiguous subsequent lines ending with (?) would be added to the output and removed from the expected output. This is a problem if the subsequent test output would have matched the consumed (?) line, because it kept the optional line and then added a duplicate without the (?) [1]. Instead, wait until there is nothing more to match before handling the leftovers. [1] https://www.mercurial-scm.org/pipermail/mercurial-devel/2016-February/080197.html

test-run-tests: pad the failure test to preserve the run order Test size seems to dictate the order in which the tests are run, and the next patch will add to test-success.t.

changelog: backed out changeset 86de91c56355

changelog: backed out changeset 1778770e1982 We want to avoid leaking UTF-8 to main body of code wherever possible.

dispatch: store norepo/optionalrepo/inferrepo attributes in function (API) This can eliminate import cycles and ugly push/pop of global variables at _checkshellalias(). Attributes of aliascmd are directly accessible. Because norepo/optionalrepo/inferrepo lists aren't populated, extensions examining them no longer work. That's why this patch removes these lists to signal the API incompatibility. This breaks 3rd-party extensions that are yet to be ported to @command decorator.

extensions: copy extra __dict__ of original function Future patches will make @command decorator set properties such as "norepo" to a function object. This patch makes sure these properties never be lost by wrapcommand() or wrapfunction(). This change won't be crazy as the standard functools.wraps() copies __dict__.

extensions: copy attributes to wrapper by wrapfunction() Before this patch, new partial function "wrap" had no useful docstring. It makes sense to copy __doc__ and __module__ as we do for wrapcommand().

extensions: extract function that copies function attributes to wrapper wrapfunction() will be changed to copy these attributes. See the next patch for details.

tests: fix section description Copy and paste error

zeroconf: import ui as uimod per test-check-module-imports

changelog: lazy decode user (API) This appears to show a similar speedup as the previous patch.

changelog: lazy decode description (API) Currently, changelog reading decodes read values. This is wasteful because a lot of times consumers aren't interested in some of these values. This patch changes description decoding to occur in changectx as needed. revsets reading changelog entries appear to speed up slightly: revset #7: author(lmoscovicz) plain 0) 0.906329 1) 0.872653 revset #8: author(mpm) plain 0) 0.903478 1) 0.878037 revset #9: author(lmoscovicz) or author(mpm) plain 0) 1.817855 1) 1.778680 revset #10: author(mpm) or author(lmoscovicz) plain 0) 1.837052 1) 1.764568