chg: define PRINTF_FORMAT_ for non gnu C compiler Before this patch, if __GNUC__ is not defined, PRINTF_FORMAT_ will not be defined and will cause compilation error. This patch solves the issue by making sure PRINTF_FORMAT_ is defined. It allows chg to be compiled with tcc (http://bellard.org/tcc/) by: tcc -o chg *.c

chgserver: drop bundle.mainreporoot config Before this patch, although chgserver drops repo object by setting it to None, there is another side effect loading a repo: setting bundle.mainreporoot. This patch explicitly sets it to empty to undo the side effect. It will make chg pass test-strip.t.

chgserver: drop old hack to recreate ui on HGPLAIN change It's been superseded by config/env hash. A log message is simplified as we no longer need a diff of environment variables. Also, _loadnewui() is changed to require args since it can't copy old --config values without args.

graphmod: allow edges to end early Rather than draw an edge all the way to the bottom of the graph, make it possible to end an edge to parents that are not part of the graph early on. This results in a far cleaner graph. Any edge type can be set to end early; set the ui.graphstyle.<edgetype> parameter to the empty string to enable this. For example, setting the following configuration: [ui] graphstyle.grandparent = : graphstyle.missing = would result in a graph like this: o changeset: 32:d06dffa21a31 |\ parent: 27:886ed638191b | : parent: 31:621d83e11f67 | : o : changeset: 31:621d83e11f67 |\: parent: 21:d42a756af44d | : parent: 30:6e11cd4b648f | : o : changeset: 30:6e11cd4b648f |\ \ parent: 28:44ecd0b9ae99 | ~ : parent: 29:cd9bb2be7593 | / o : changeset: 28:44ecd0b9ae99 |\ \ parent: 1:6db2ef61d156 | ~ : parent: 26:7f25b6c2f0b9 | / o : changeset: 26:7f25b6c2f0b9 |\ \ parent: 18:1aa84d96232a | | : parent: 25:91da8ed57247 | | : | o : changeset: 25:91da8ed57247 | |\: parent: 21:d42a756af44d | | : parent: 24:a9c19a3d96b7 | | : | o : changeset: 24:a9c19a3d96b7 | |\ \ parent: 0:e6eb3150255d | | ~ : parent: 23:a01cddf0766d | | / | o : changeset: 23:a01cddf0766d | |\ \ parent: 1:6db2ef61d156 | | ~ : parent: 22:e0d9cccacb5d | | / | o : changeset: 22:e0d9cccacb5d |/:/ parent: 18:1aa84d96232a | : parent: 21:d42a756af44d | : | o changeset: 21:d42a756af44d | |\ parent: 19:31ddc2c1573b | | | parent: 20:d30ed6450e32 | | | +---o changeset: 20:d30ed6450e32 | | | parent: 0:e6eb3150255d | | ~ parent: 18:1aa84d96232a | | | o changeset: 19:31ddc2c1573b | |\ parent: 15:1dda3f72782d | ~ ~ parent: 17:44765d7c06e0 | o changeset: 18:1aa84d96232a parent: 1:6db2ef61d156 parent: 15:1dda3f72782d The default configuration leaves all 3 types set to |. This is part of the work towards moving smartlog upstream; currently smartlog injects extra nodes into the graph to indicate grandparent relationships (nodes elided).

graphmod: allow for different styles for different edge types Rather than draw all edges as solid lines, allow for using different styles for different edge types. For example you could use dotted lines for edges that do not connect to a parent, and dashed lines when connecting to a grandparent (implying missing nodes in between). For example, setting the following configuration: [ui] graphstyle.grandparent = : graphstyle.missing = . would result in a graph like this: o changeset: 32:d06dffa21a31 |\ parent: 27:886ed638191b | : parent: 31:621d83e11f67 | : o : changeset: 31:621d83e11f67 |\: parent: 21:d42a756af44d | : parent: 30:6e11cd4b648f | : o : changeset: 30:6e11cd4b648f |\ \ parent: 28:44ecd0b9ae99 | . : parent: 29:cd9bb2be7593 | . : o . : changeset: 28:44ecd0b9ae99 |\ \ \ parent: 1:6db2ef61d156 | . . : parent: 26:7f25b6c2f0b9 | . . : o . . : changeset: 26:7f25b6c2f0b9 |\ \ \ \ parent: 18:1aa84d96232a | | . . : parent: 25:91da8ed57247 | | . . : | o-----+ changeset: 25:91da8ed57247 | | . . : parent: 21:d42a756af44d | | . . : parent: 24:a9c19a3d96b7 | | . . : | o . . : changeset: 24:a9c19a3d96b7 | |\ \ \ \ parent: 0:e6eb3150255d | | . . . : parent: 23:a01cddf0766d | | . . . : | o---+ . : changeset: 23:a01cddf0766d | | . . . : parent: 1:6db2ef61d156 | | . . . : parent: 22:e0d9cccacb5d | | . . . : | o-------+ changeset: 22:e0d9cccacb5d | . . . . : parent: 18:1aa84d96232a |/ / / / / parent: 21:d42a756af44d | . . . : | . . . o changeset: 21:d42a756af44d | . . . |\ parent: 19:31ddc2c1573b | . . . | | parent: 20:d30ed6450e32 | . . . | | +-+-------o changeset: 20:d30ed6450e32 | . . . | parent: 0:e6eb3150255d | . . . | parent: 18:1aa84d96232a | . . . | | . . . o changeset: 19:31ddc2c1573b | . . . .\ parent: 15:1dda3f72782d | . . . . | parent: 17:44765d7c06e0 | . . . . | o---+---+ | changeset: 18:1aa84d96232a . . . . | parent: 1:6db2ef61d156 / / / / / parent: 15:1dda3f72782d . . . . . Edge styles can be altered by setting the following one-character config options:: [ui] graphstyle.parent = | graphstyle.grandparent = : graphstyle.missing = . The default configuration leaves all 3 types set to |, leaving graph styles unaffected. This is part of the work towards moving smartlog upstream; currently smartlog injects extra nodes into the graph to indicate grandparent relationships (nodes elided).

chgserver: use global ui instead of repo ui for dispatch.request.ui Before this patch, chgserver will use repo ui as dispatch.request.ui, while req.ui is designed to be global ui without repo config. Passing repo ui as dispatch.request.ui leads to repo.ui being incorrect, which can lead to unwanted results. For example, if the repo config has [extensions], it could affect which localrepository.featuresetupfuncs get executed and the repo may have an incorrect list of supported requirements. This patch changes _renewui to return both global ui and repo ui. The global ui is passed to req.ui, and the repo ui is used to calculate confighash. It will make chg pass test-largefiles-misc.t and test-requires.t, which are both related to repo requirements.

progress: update comment to reflect implementation Progress has been on by default in core for a while, the comment was written before this happened.

tests: remove lines that enable progress extension progress has not been an extension for a long time, it has been integrated into core, thus these lines have not done anything for a while.

run-tests: add support for automatically bisecting test failures

check-code: also ban strcat We're not using it now, so it's easy to ban.

check-code: prevent use of strcpy

osutil: stop using strcpy strcpy is a security vulnerability masquerading as a utility function. Replace it with memcpy since we know how much to copy anyway.

histedit: add a hint about enabled dropmissing to histedit edit comment Adds a hint to histedit comment reminding user about enabled dropmissing. This will make the enabled dropmissing more visible. The example comment: # Edit history between b592564a803c and b54649a8a63f # # Commits are listed from least to most recent # # You can reorder changesets by reordering the lines # # Commands: # # e, edit = use commit, but stop for amending # m, mess = edit commit message without changing commit content # p, pick = use commit # b, base = checkout changeset and apply further changesets from there # d, drop = remove commit from history # f, fold = use commit, but combine it with the one above # r, roll = like fold, but discard this commit's description # # Deleting a changeset from the list will DISCARD it from the edited history!

hghave: change ssl check to just check ssl module Previously, the "ssl" check effectively looked for PyOpenSSL or Python 2.7.9. After this patch, we simply look for just the "ssl" module. After d962e955da08, there have been no references to PyOpenSSL in the tree (the previous usage of PyOpenSSL was to implement ssl support on old, no longer supported Python versions that didn't have an ssl module (e.g. Python 2.4). So, the check for PyOpenSSL served no purpose. Pythons we support ship with the ssl module. Although it may not be available in all installations. So, we still need the check for whether the ssl module imports, hence the hghave check. The main side-effect of this change is that we now run test-https.t (the only test requiring the "ssl" hghave feature) on Python <2.7.9 when PyOpenSSL is not installed (which is probably most installations) and the ssl module is available. Before, we wouldn't run this test on these older Python versions. I confirmed that test-https.t passes with Python 2.6.9 and 2.7.8 on OS X 10.11.

help: document sharing of revlog header with revision 0 The previous docs were incorrect about there being a discrete header on revlogs.

mpatch: move collect() to module level This helps the code read a little more clearly.

mpatch: un-nest the move() method This helps the code read a little more clearly.

mpatch: move pull() method to top level There was no need for this to be nested.

chgserver: use old ui.system if fout is not stdout or needs to be captured Before this patch, chgui will override the system method, forwarding every process execution to the client so sessions and process groups can work as expected. But the chg client will just use stdout, if ui.fout is not stdout or if the output is set to be captured to safe._buffers, the client will not behave correctly. This can happen especially with code prepending "remote:". For example, bundle2 uses ui.pushbuffer, and sshpeer sets fout to ferr. We may have trouble with interactive commands in the fout set to ferr case but if it really bites us, we can always send file descriptors to the client. This patch adds a check to detect the above situations and fallback to the old ui.system if so. It will make chg happy with test-bundle2-exchange.t, test-phases-exchange.t, test-ssh-bundle1.t and test-ssh.t.

node: use byte literals to construct nullid and wdirid Python 3's hex() insists on operating on bytes. This patch gives it what it wants. '' and b'' in Python 2 are equivalent, so this has no impact on Python 2.

tests: try to import modules with Python 3 All of mercurial.* is now using absolute_import. Most of mercurial.* is able to ast parse with Python 3. The next big hurdle is being able to import modules using Python 3. This patch adds testing of hgext.* and mercurial.* module imports in Python 3. As the new test output shows, most modules can't import under Python 3. However, many of the failures are due to a common problem in a highly imported module (e.g. the bytes vs str issue in node.py).

tests: perform an ast parse with Python 3 Previously, test-check-py3-compat.t parsed Python files with Python 2 and looked for known patterns that are incompatible with Python 3. Now that we have a mechanism for invoking Python 3 interpreters from tests, we can expand check-py3-compat.py and its corresponding .t test to perform an additional AST parse using Python 3. As the test output shows, we identify a number of new parse failures on Python 3. There are some redundant warnings for missing parentheses for the print function. Given the recent influx of patches around fixing these, the redundancy shouldn't last for too long.

run-tests: add --with-python3 to define a Python 3 interpreter Currently, very few parts of Mercurial run under Python 3, notably the test harness. We want to write tests that run Python 3. For example, we want to extend test-check-py3-compat.t to parse and load Python files. However, we have a problem: finding appropriate files requires running `hg files` and this requires Python 2 until `hg` works with Python 3. As a temporary workaround, we add --with-python3 to the test harness to allow us to define the path to a Python 3 interpreter. This interpreter is made available to the test environment via $PYTHON3 so tests can run things with Python 3 while the test harness and `hg` invocations continue to run from Python 2. To round out the feature, a "py3exe" hghave check has been added.

crecord: add docblock to handlekeypressed This information is pretty useful when reading the code.

crecord: fix docblock indentation

crecord: clean up empty lines at ends of docblocks

filemerge: indicate that local/other are p1/p2

sslutil: use preferred formatting for import syntax

largefiles: add some docstrings

largefiles: drop partial support for not having a user cache 971c55ce03b8 introduced support for not having a "global" user cache. In the rare cases where the environment didn't provide the location of the current home directory, the usercachepath function could return None. That functionality has since bitrotten and several code paths did not correctly check for usercachepath returning None: $ HOME= XDG_CACHE_HOME= hg up --config extensions.largefiles= getting changed largefiles abort: unknown largefiles usercache location Dropping the partial support for it is thus not really a backward compatibility breaking change. Thus: consistently fail early if the usercache location is unknown. It is relevant to be able to control where the largefiles are stored and how they propagate, but that should probably be done differently. The dysfunctional code just gets in the way.

largefiles: refactor usercachepath - extract user cache path function It is convenient to have the user cache location explicitly.

shelve: adds restoring newly created branch (issue5048) (BC) Before this patch shelve never preserved branch information, so after applying unshelve branch was the same as it was on working copy no matter in which branch shelve took place. This patch makes bare shelving(with no files specified, without interactive,include and exclude option) remembers information if the working directory was on newly created branch ,in other words working directory was on different branch than its first parent. In this situation unshelving restores branch information to the working directory.

shelve: make non bare shelve not saving branch information in bundle This patch prepares for restoring newly created branch only on bare shelve later because information about new-branch will be preserved only when shelve was bare and working copy branch was different than branch of its parent. In other case information about new-branch will be gone, so unshelve will not recognise that shelve was made on new-branch and it will not restore branch information from the bundle to the working directory.

shelve: preserve newly created branch on non-bare shelve in wctx (BC) Before this patch current branch in working context wasnt preserved after shelve, this patch makes it restore after update.

crecord: rewrite a comment about filtering patches I couldn't parse the previous version. I think the rewrite better expresses the intent of that comment.

run-tests: teach _processoutput to handle multiple lines of churn Instead of treating expected output as happening in a precise order, and assuming that if a line is missing it will never happen, assume that expected output is a prioritized list of likely matching lines. This means that if: foo/bar (glob) baz/bad (glob) changes to: baz/bad foo/bar instead of generating: baz/bad foo/bar For which we've lost both (glob) markers, we will match both lines and generate: baz/bad (glob) foo/bar (glob) This retains any special annotations we have for lines.

run-tests: indent _processoutput to aid readability for next patch The next commit will loop over the expected[pos] list, this change makes that change easier to review.

namespaces: fix name/node confusion There was a lot of copy/paste here.

tests: add test for "log -r wdir() -p" (issue4871) The issue has been fixed by 93bcc73df8d5, "cmdutil.changeset_printer: pass context into showpatch()". This patch adds test to prevent future regression.

tests: include modified/added/removed files in "log -r wdir()" output This patch prepares for the test of "hg log -r wdir() -p".

contrib: revsetbenchmarks use absolute_import and print_function

contrib: synthrepo use absolute_import

contrib: python-hook-examples use absolute_import

contrib: make perf.py use absolute_import

largefiles: replace invocation of os.path module by vfs in lfutil.py Replaces invocations os.path functions to methods in vfs. Unfortunately (in my view) this makes code less readable, because instead of using clear variable names with path it needs to replace them with vfs(..). I need guidance how to make such transition look more readable. For example in this patch there is example with few places with wvfs.join(standindir), standindir before this patch was absolute path, in this it is changed to relative because it is used also in expression wvfs.join(standindir, pat).

largefiles: replace invocation of os.path module by vfs in lfcommands.py

cache: rebuild branch cache from scratch when inconsistencies are detected This should recover automatically from some corruptions that for unknown reasons are seen in the wild.

cache: safer handling of failing seek when writing revision branch cache If the seek for some reason fails (perhaps because the file is too short to search to the requested position), make sure we seek to the start and rewrite everything. It is unknown if this fixes a real problem that ever happened.

cache: remove branch revision file before rewriting the branch name file New branch names are usually appended to the branch name file. If that fails or the file has been modified by another process, it is rewritten. That left a small opportunity that there could be references to non-existent entries in the file while it was rewritten. To avoid that, remove the revision branch cache file with the references to the branch name file before rewriting the branch name file. Worst case, when interrupted at the wrong time, the cache will be lost and rebuilt next time. It is unknown if this fixes a real problem that ever happened.

chg: do not redirect stdout to /dev/null Redirecting stdout to /dev/null has unwanted side effects, namely ui.write will stop working. This patch removes the redirection code and helps chg to pass test-bad-extension.t.

pager: skip uisetup if chg is detected chg has its own pager implementation that it wants to skip pager's uisetup. It is currently done by redirecting stdout to /dev/null, which has unintended side effects. This patch makes pager aware of chg and skip uisetup directly from pager. We may want to merge chg and pager's pager implementation to make this unnecessary in the future.

chgserver: add an explicit gc to trigger __del__ SocketServer.ForkingMixIn uses os._exit which will skip all cleanup handlers. We want to run __del__ to make things like transactions, {ssh,http}peer, atomictempfile, dirstateguard, etc. work. This patch adds a "gc.collect()" to trigger __del__. It is helpful for chg to pass some test cases in test-devel-warnings.t.

blackbox: do not assume self._bb{vfs,repo,fp} are set in blackboxui.__init__ It's possible for the blackboxui code to do a "del self._bbvfs", then ui.copy() or similar attempt will fail. It will also fail when constructing a blackboxui from a non-blackbox ui. This patch fixes the issue by not assuming any _bb* attr is set.

chg: downgrade "failed to read channel" from abortmsg to debugmsg If the server has an uncaught exception, it will exit without being able to write the channel information. In this case, the client is likely to complain about "failed to read channel", which looks inconsistent with original hg. This patch silences the error message and makes uncaught exception behavior more like original hg. It will help chg to pass test-fileset.t.

histedit: do not close stdin Closing stdin is unexpected by chgserver and is not a good idea generally. This patch refactors related code a bit and make sure stdin is not closed. It will make chg much happier on test-histedit*.t.

tests: reorder hg serve commands chg currently does not support hg serve -d. It has a quick path testing if the command is hg serve -d and fallbacks to hg if so. But the test only works if "serve" is the first argument since the test wants to avoid false positives (for example, "-r serve" is different). This patch reorders "hg server" commands in tests, making them chg friendly.

templater: add debugtemplate command This is useful for debugging template parsing. Several tests are ported to this command.

templater: expand list of parsed templates to template node This patch eliminates a nested data structure other than the parsed tree. ('template', [(op, data), ..]) -> ('template', (op, data), ..) New expanded tree can be processed by common parser functions. This change will help implementing template aliases. Because a (template ..) node should have at least one child node, an empty template (template []) is mapped to (string ''). Also a trivial string (template [(string ..)]) node is unwrapped to (string ..) at parsing phase, instead of compiling phase.

templater: relax type of mapped template Now compiled template fragments are packed into a generic type, (func, data), a string can be a valid template. This change allows us to unwrap a trivial string node. See the next patch for details.

templater: lift parsed and compiled templates to generic data types Before this patch, parsed and compiled templates were kept as lists. That was inconvenient for applying transformation such as alias expansion. This patch changes the types of the outermost objects as follows: stage old new -------- -------------- ------------------------------ parsed [(op, ..)] ('template', [(op, ..)]) compiled [(func, data)] (runtemplate, [(func, data)]) New templater.parse() function has the same signature as revset.parse() and fileset.parse().