tests: rewrite test-rebase-detach.t using drawdag This makes it much clearer because each test case can have exactly the graph it wants (no extra cruft just because other tests want it). Differential Revision: https://phab.mercurial-scm.org/D94

tests: simplify config in test-rebase-detach.t Differential Revision: https://phab.mercurial-scm.org/D93

drawdag: include files from both parents in merge commits Consider a graph like this: D |\ B C |/ A drawdag will add a file called A in commit A, file B in B, file C in C. That's fine and expected. In merge commits like D, I would expect the files and their contents to be taken from the parent commits, so commit D in this example would have files A, B, and C. However, drawdag will instead add the file D compared to the first parent. Depending on whether B or C got a smaller nodeid, the contents of D would be {A, B, D} or {A, C, D}. This patch changes it to to be {A, B, C}. Differential Revision: https://phab.mercurial-scm.org/D92

gitweb: preserve whitespace in description Without this, multiple spaces or tabs in the commit message aren't preserved and things like tables don't align properly. As part of adding the CSS rule, we had to cuddle the content with the <div> to not introduce leading and trailing whitespace. The "addbreaks" filter was also removed because it would insert an additional newline, effectively double spacing content. Differential Revision: https://phab.mercurial-scm.org/D113

sparse: add a requirement when a repository uses sparse (BC) The presence of a sparse checkout can confuse legacy clients or clients without sparse enabled for reasons that should be obvious. This commit introduces a new repository requirement that tracks whether sparse is enabled. The requirement is added when a sparse config is activated and removed when the sparse config is reset. The localrepository constructor has been taught to not open repos with this requirement unless the sparse feature is enabled. It yields a more actionable error message than what you would get if the lockout were handled strictly at the requirements verification phase. Old clients that aren't sparse aware will see the generic "repository requires features unknown to this Mercurial" error, however. The new requirement has "exp" in its name to reflect the experimental nature of sparse. There's a chance that the eventual non-experimental feature won't change significantly and we could have squatted on the "sparse" requirement without ill effect. If that happens, we can teach new clients to still recognize the old name. But I suspect we'll sneak in some BC and we'll want a new requirement to convey new meaning. Differential Revision: https://phab.mercurial-scm.org/D110

sparse: consolidate common code for writing sparse config In 3 functions we were writing the sparse config and updating the working directory. In two of them we had a transaction-like process for restoring the sparse config in case of wdir update fail. Because the pattern is common, we've already made mistakes, and the complexity will increase in the near future, let's consolidate the code into a reusable function. As part of this refactor, we end up reading the "sparse" file twice when updating it. This is a bit sub-optimal. But I don't think it is worth the code complexity to pass around the variables to avoid the redundancy. Differential Revision: https://phab.mercurial-scm.org/D109

revset: pass repo when passing ui The repo instance is currently only used to provide a changeset lookup function as part of parsing revsets. I /think/ this allows node fragments to resolve. I'm not sure why we wouldn't want this to always "just work" if parsing a revset string. Plus, an upcoming commit will introduce a new consumer that needs a handle on the repo. So passing it more often will make that code work more. Passing a repo instance in all callers of revset.match* results in a bunch of test changes. Notably, branch and tags caches get populated as part of evaluating revsets. I'm not sure if this is desirable. So this patch takes the conservative approach and only passes the repo if we're passing a ui instance. Differential Revision: https://phab.mercurial-scm.org/D97

win32: copy-edit debugssl messages to match prevailing style

run-tests: make sure to check if pygments is installed before using it e80041832e introduced support to color the output of tests but used pygments without checking whether it's installed or not. That breaks test-run-tests.t for machines which don't have pygments installed. This patch conditionalize the color test in test-run-tests.t and also add a check to make sure pygments is installed before using that.

sparse: require [section] in sparse config files (BC) Previously, [include] was implicit and pattern lines before a [section] were added to includes. Because the format may change in the future and explicit behavior, well, more explicit, this commit changes the config parser to reject pattern lines that don't occur in a [section]. Differential Revision: https://phab.mercurial-scm.org/D96

sparse: use set for capturing profiles Order doesn't need to be preserved. A set is acceptable. Differential Revision: https://phab.mercurial-scm.org/D95

util: remove dead code which used to be for old python2 versions Differential Revision: https://phab.mercurial-scm.org/D107

status: add a flag to terse the output (issue4119) This adds an experimental flag -t/--terse which will terse the output. The terse flag will respect other flags which filters the output. The flag takes a string whose value can be a subsequence of "marduic" (the order does not matter here.) Ignored files are not considered while tersing unless -i flag is passed or 'i' is there is the terse flag value. The flag is experimental for testing as there may be cases which will produce strange results with the flag. We can set the terse on by default by simply passing 'u' to the cmdutil.tersestatus(). This patch also adds a test file with tests covering the new feature.

commitextras: check the format of the arguments and no internal key is used This patch adds check to make the arguments are passed as KEY=VALUE and no key which is used internally is passed. This patch also adds test for the extension.

commitextras: move fb extension to core which add extras to a commit This patch moves the Facebook extension to add extra fields to a commit to a in-core extension.

archive: add an experimental config to control the metadata file template Experimental because given the possible complexity, it may be worth figuring out how to load this from a file, similar to the style files for the log command, instead of trying to stuff it on the command line.

archive: use a templater to build the metadata file There are no visible changes here. I'm starting to wonder if adding the '+' to the 'node' line instead of a separate key line in 3047167733dc was the right thing to do. The '{node}' keyword never includes '+' elsewhere, and the way setup.py works, it would truncate it anyway. Additionally, the file is missing '{p2node}' when 'wdir()' merges are archived. I thought about adding an 'identify' line that would correspond to `hg id -n`. But the other nodes are the full 40 characters, and the output most useful for versioning is the short form. All of this cries out for customization via templating. (Although maybe having the short identify line by default is still a good idea.)

commandserver: do not handle EINTR for selector.select selectors2 library handles EINTR transparently so we don't need to handle select.error ourselves.

debugobsolete: also report the number of obsoleted changesets This seems useful to have the number of obsoleted changesets when calling debugobsolete.

transaction-summary: display the summary for all transactions Now that we records "all" changes happening in a transaction (in tr.changes) we will be able to provide better report on various changes (phases turned public, changeset obsoleted, branch merged or created, etc..) This is far too late in the cycle to play with this, but having this existing method called more widely will help extensions to play around with various options during the 4.4 cycle. Instead of calling registersummarycallback only for transactions we want, we always call it and use the transaction name to decide when to report (eg: we do not want `hg amend` to report new obsoleted changesets). Filtering on transaction name does not seems great, but seems good enough for the moment. We can change the API during the next cycle. The previous manual call during unbundling of the bundle2 "obsmarkers" part is no longer necessary and has been dropped.

test-strip: fix call to super class in crashstriprepo The 'self' argument is already taken care of by the 'super(...)' call. Instead we were passing the repository instance as the 'desc' value. This got discovered by breaking the next changeset.

share: share 'cachevfs' with the source clone (issue5108) Share extension now also share caches reads and writes. Not sharing caches results in costly caches recomputations which can takes up to minutes when using shares on large repositories. There are a couple of file in the '.hg/cache/' that depends of the current visibility. Visibility can be affected by the working copy location, something which is specific to each share. We ignores them for this series because they: * are the minority, * already have a good fallback to other precomputed caches, * are only affected when people use the experimental evolution feature.

cachevfs: add a devel warning for cache access though 'vfs' This will help third party extensions to migrate to the new 'cachevfs'.

cachevfs: migration the tags fnode cache to 'cachevfs' This will help sharing the cache between shares.

cachevfs: migrate tagscache to 'cachevfs' This will help sharing the cache between shares.

cachevfs: migration the revbranchcache to 'cachevfs' This will help sharing the cache between shares.

cachevfs: use the new vfs in when computing branchmap cache This will help sharing the cache between shares.

cachevfs: add a vfs dedicated to cache Most of the cache content lives in '.hg/cache/'. Moreover they are computed exclusively from data in the '.hg/store' directory. This creates issues with the share extension as the '.hg/store' directory is shared but the '.hg/cache' is not. On large repositories, this makes this prevent some usage of the share extension inefficient as some caches can take minutes to be recomputed. To improve the situation, we introduce a new 'cachevfs' that will be dedicated to cache reading and writing. In the next patches of this series, we'll migrate the 4 existing caches to it and update the share extension.

vfsward: register 'write with no lock' warnings as 'check-locks' config Update 'write with no lock' warnings in order to be better controlled by the config. We reuse the option used for lock order for these other lock related message. The message can now be disabled using 'devel.check-locks = no' (in addition to the usual 'devel.all-warnings = no').

demandimport: prefer loaded module over package attribute (issue5617) In general, the attribute of the same name is overwritten by executing an import statement. import a.b print(a.b.c) # 'c' of a/b/__init__.py from a.b.c import d print(a.b.c) # a/b/c.py However, this appears not true for the scenario described in the test case, and surprisingly, "from a.b.c import d" works even if "a.b.c" is not a module. This patch works around the problem by taking the right module from sys.modules if available.

demandimport: drop Py3 workarounds from Py2 implementation

demandimport: drop hack for old Pythons which had no level argument _import() was added by 868282fa29d8, but it is identical to _origimport() now.

tests: drop "absimport" conditionals which should be always true

color: drop the now useless color extension all the extension features are provided by core since 4.2.

extensions: expand the builtins extensions declaration This will future updates of the set cleaner and more readable.

debugextensions: drop the color from the test We are about to remove the extension so we remove trace of it in this specific test to prevent it to be confused.

configitems: register the 'bugzilla.notify' config

configitems: register the 'ui.mergemarkertemplate' config

configitems: register the 'ui.color' config

color: drop the _enabledbydefault module variable Since color is on by default, cleanup the now useless variable in both core and color extension.

configitems: register the 'ui.forcecwd' config

configitems: register the 'ui.fallbackencoding' config

bookmark: deprecate direct del of a bookmark value We want all bookmark deletion to go through 'applychanges', so lets deprecate legacy ways of doing bookmark deletion.

bookmark: deprecate direct set of a bookmark value We want all bookmark update to go through 'applychanges', so lets deprecate legacy ways of doing bookmark update.

bookmark: track bookmark changes at the transaction level The transaction has now a 'bookmarks' dictionary in tr.changes. The structure of the dictionary is {BOOKMARK_NAME: (OLD_NODE, NEW_NODE)}. If a bookmark is deleted NEW_NODE will be None. If a bookmark is created OLD_NODE will be None. If the bookmark is updated multiple time, the initial value is preserved.

bookmark: deprecate 'recordchange' in favor of 'applychanges' Now that we have migrated all in-core caller of 'recordchange' to 'applychanges', deprecate 'recordchange' so external callers will move to the new unified method.

bookmark: drop deletedivergent It has no caller anymore.

bookmark: use 'divergent2delete' in checkconflict checkconflict used to also do some bookmark deletion in case of divergence. It is a bit suspicious given the function name, but it's not the goal of this series. In order to unify bookmarks changing, checkconflict now return the list of divergent bookmarks to clean up and the callers must clean them by calling applyphases.