Yuya Nishihara <yuya@tcha.org> [Sat, 13 Jan 2018 15:07:37 +0900] rev 35691
fileset: make it robust for bad function calls
Before, it could crash or show cryptic message.
Yuya Nishihara <yuya@tcha.org> [Thu, 04 Jan 2018 14:20:58 +0900] rev 35690
log: build follow-log filematcher at once
We no longer need to replay copy tracing to build filematcher as we can
walk (rev, fctxs) pairs.
Yuya Nishihara <yuya@tcha.org> [Thu, 04 Jan 2018 15:20:46 +0900] rev 35689
log: allow matchfn to be non-null even if both --patch/--stat are off
This makes the next patch slightly simpler.
Yuya Nishihara <yuya@tcha.org> [Thu, 04 Jan 2018 14:37:15 +0900] rev 35688
log: replace "not pats" with matcher attribute for consistency
We use match.always() in _makelogrevset(). match.always() is identical to
"not pats" here since scmutil.matchandpats() empties pats in that case.
Yuya Nishihara <yuya@tcha.org> [Wed, 03 Jan 2018 15:58:59 +0900] rev 35687
log: follow file history across copies even with -rREV (BC) (issue4959)
Still it falls back to changelog path if glob patterns or missing paths
are specified. This will be changed later.
.. bc::
``log --follow -rREV FILE..`` now follows file history across copies and
renames.
Yuya Nishihara <yuya@tcha.org> [Wed, 03 Jan 2018 15:46:15 +0900] rev 35686
log: resolve --follow thoroughly in getlogrevs()
This makes sense because --follow isn't really an option to filter revisions,
but an option to extend revisions to be filtered.
_fileancestors() is a minimal copy of revset._follow(). They are slightly
different in that which revision the matcher sees. _fileancestors() also
uses ctx.walk() instead of ctx.manifest().walk() to show a better warning
on bad match, which will be tested later.
Yuya Nishihara <yuya@tcha.org> [Wed, 03 Jan 2018 15:13:22 +0900] rev 35685
log: extract function that processes log file patterns
We'll need a matcher to compute revs followed from the given patterns.
Yuya Nishihara <yuya@tcha.org> [Sun, 10 Dec 2017 17:28:44 +0900] rev 35684
log: resolve --follow with -rREV in cmdutil.getlogrevs()
This also fixes alias expansion. Before, reverse() could be overridden by user
alias.
This isn't processed at _logrevs() as we'll need starting revisions to parse
file patterns. See the subsequent patches for details.
Yuya Nishihara <yuya@tcha.org> [Sun, 10 Dec 2017 18:22:06 +0900] rev 35683
log: remove dead code to follow descendants if ascending revisions specified
It's been disabled since 8b4b9ee6001a "log: make -fr show complete history
from the given revs" 2015-02-06, and we've got no bug report regarding this
exact behavior. For details what the heck "followdescendants" is, please see
the issue comment:
https://bz.mercurial-scm.org/show_bug.cgi?id=4959#c72
"'rev(%d)' % startrev" can be replaced with '.' since revs are 'reversed(:.)'.
Yuya Nishihara <yuya@tcha.org> [Sun, 10 Dec 2017 16:59:54 +0900] rev 35682
log: duplicate _logrevs() dedicated for walkchangerevs()
Prepares for refactoring getlogrevs() to fix the "log -frREV PATH" issue.
I initially thought I could get rid of walkchangerevs(), but it turned out
requiring non-trivial work because of a "prepare" callback and a scanning
window. This patch makes sure that walkchangerevs() will be unaffected by
subsequent changes.
Yuya Nishihara <yuya@tcha.org> [Tue, 16 Jan 2018 21:51:27 +0900] rev 35681
update: fix crash on bare update when directaccess is enabled
'rev' may be None. An empty 'rev' string shouldn't also be passed to
unhidehashlikerevs().
Yuya Nishihara <yuya@tcha.org> [Tue, 16 Jan 2018 21:42:42 +0900] rev 35680
templater: make sure expression is terminated by '}'
This can't be ensured by the tokenize() generator since the parser stop
consuming tokens once it reached an end of an operation.
Spotted by Matt Harbison.
Boris Feld <boris.feld@octobus.net> [Fri, 14 Jul 2017 19:59:10 +0200] rev 35679
bookmark: deprecate direct update of a bookmark value
This changeset was part of an older series preventing direct update of the
bookmarks store without the change being tracked by a transaction. For some
unclear reason, this was not sent to the list. So we are sending it now.
Initial series: 3325c7dcabaa+08bf0ebc6c8e+712a85b3677f.
We want to channel all bookmarks update through 'applychanges' so let's remove
more way to work around this function.
Paul Morelle <paul.morelle@octobus.net> [Mon, 16 Oct 2017 14:32:06 +0200] rev 35678
debugdeltachain: display how many chunks would be read in sparse-read mode
Alex Gaynor <agaynor@mozilla.com> [Wed, 17 Jan 2018 21:36:18 +0000] rev 35677
bdiff: handle the possibility of an integer overflow when allocating
Differential Revision: https://phab.mercurial-scm.org/D1877
Jordi Gutiérrez Hermoso <jordigh@octave.org> [Tue, 16 Jan 2018 08:40:14 -0500] rev 35676
svnsubrepo: check if subrepo is missing when checking dirty state (issue5657)
During an hg convert command, subrepos are not checked out. However, a
manifestmerge may occur during an hg convert if there is a merge
commit, which in particular will check the dirty state of subrepos.
Before this change, the lack of a checked out subrepo would error out.
This check for svn subrepos is modelled after 2fdea636f254 for git
subrepos.
Boris Feld <boris.feld@octobus.net> [Wed, 17 Jan 2018 13:02:42 +0100] rev 35675
win32text: use 'tiprev' when appropriate
This is cleaner than the current 'len(repo) - 1' form.
Boris Feld <boris.feld@octobus.net> [Wed, 17 Jan 2018 13:01:38 +0100] rev 35674
patchbomb: use 'tiprev' when appropriate
This is cleaner than the current 'len(repo) - 1' form.
Boris Feld <boris.feld@octobus.net> [Thu, 04 May 2017 02:28:19 +0200] rev 35673
revset: use 'tiprev' when appropriate
This is cleaner than the current 'len(repo) - 1' forms
Boris Feld <boris.feld@octobus.net> [Thu, 04 May 2017 02:24:04 +0200] rev 35672
changelog: use 'tiprev()' in 'tip()'
Since 'tip()' is just a node version of 'tiprev()' we should reuse the code.
Boris Feld <boris.feld@octobus.net> [Thu, 04 May 2017 02:23:21 +0200] rev 35671
changelog: introduce a 'tiprev' method
Accessing tiprev is a common need through the code base. It is usually done
using "len(changelog) -1". That form is tedious and error-prone. For example,
it will give wrong results on filtered changelog (if the unfiltered tip is
filtered).
As a result, we introduce a simple 'tiprev()' method to provide this exact
information in a nice way.
Augie Fackler <augie@google.com> [Wed, 17 Jan 2018 15:47:38 -0500] rev 35670
contrib: add some basic scaffolding for some fuzz test targets
I'd like to get the majority of our C code covered by automated fuzz
testing. I've started with bdiff because it was already decoupled from
libpython and therefore was fairly quick to produce a working
fuzzer. The code here is a little odd because I've been having trouble
convincing libfuzzer to define a main and I threw in the towel.
This code will also work with github.com/google/oss-fuzz, and once it
lands in our main repo I intend to enable automated fuzzing in
oss-fuzz with reports going to our security alias.
Differential Revision: https://phab.mercurial-scm.org/D1875
Augie Fackler <augie@google.com> [Wed, 17 Jan 2018 15:56:12 -0500] rev 35669
formatting: enforce clang-format on all .cc files as well as .c files
As part of my upcoming fuzzer patch we're going to grow some .cc
files. They're not part of the release build (they're only used for
fuzzing, which requires clang 6 which in turn implies a working C++
compiler), so I'm not terribly worried about growing some C++ files.
Differential Revision: https://phab.mercurial-scm.org/D1874
Augie Fackler <augie@google.com> [Wed, 17 Jan 2018 15:28:38 -0500] rev 35668
hghave: add test for clang 6 and later
clang 6 includes integrated libfuzzer support, which makes fuzzer
stuff a little easier to work with.
Differential Revision: https://phab.mercurial-scm.org/D1873
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 14 Jan 2018 12:06:27 -0800] rev 35667
py3: cast error message to localstr in blackbox.py
According to the exceptions reporting mechanism, this is the #1 crasher
in the test harness for Python 3.
Since this exception is in an except block, we'll likely get a
new #1 crasher after this change. But at least we won't die
during error handling!
Differential Revision: https://phab.mercurial-scm.org/D1478
Jun Wu <quark@fb.com> [Wed, 10 Jan 2018 21:27:05 -0800] rev 35666
lfs: remove internal url in test
`test-lfs-test-server.t` refers to a FB internal domain and requires certain
implementation (ex. set error code to 404) at that endpoint. Without any
workaround, It should in theory error out like "Domain cannot be resolved".
I don't know how Matt Harbison ran the test.
This patch changes the test to only depend on `lfs-test-server`.
Unfortunately the logic has to be changed since `lfs-test-server` does not
set error code to 404 but just removes "download" from "actions".
Differential Revision: https://phab.mercurial-scm.org/D1849
Matt Harbison <matt_harbison@yahoo.com> [Sun, 14 Jan 2018 18:12:51 -0500] rev 35665
lfs: control tracked file selection via a tracked file
Since the lfs tracking policy can dramatically affect the repository, it makes
more sense to have the policy file checked in, than to rely on all developers
configuring their .hgrc properly. The inspiration for this is the .hgeol file.
The configuration lives under '[track]', so that other things can be added in
the future. Eventually, the config option should be limited to `convert` only.
If the file can't be parsed for any reason (including unrecognized elements of
the minifileset language), the commit will abort until the problem is corrected.
This seems more useful than the warning that hgeol emits, and has no effect on
reading the data, so there's no compatibility concerns.
My initial thought was to read the file and change each "key = value" line into
"((key) & (value))", so that each line could be ORed together, and make a single
pass at compiling. Unfortunately, that prevents exclusions if there's a
catchall rule. Consider what happens to a large *.c file here:
[track]
**.c = none()
** = size('>1MB')
# ((**.c) & (none())) | ((**) & (size('>1MB'))) => anything > 1MB
I also thought about having separate [include] and [exclude] sections. But that
just seems to open things up to user mistakes. Consider:
[include]
**.zip = all()
**.php = size('>10MB')
[exclude]
**.zip = all() # Who wins?
**.php = none() # Effectively 'all()' (i.e. nothing excluded), or >10MB ?
Therefore, it just compiles each key and value separately, and walks until the
key matches something. I'm not sure how to enforce just file patterns on LHS
without leaking knowledge about the minifileset here. That means this will
allow odd looking lines like this:
[track]
**.c | **.txt = none()
But that's also fewer lines to compile, so slightly more efficient? Some things
like 'none()' won't work as expected on LHS though, because that won't match, so
that line is skipped. For now, these quirks are not mentioned in the
documentation.
Jun previously expressed concern about efficiency when scaling to large repos,
so I tried avoiding 'repo[None]'. (localrepo.commit() gets repo[None] already,
but doesn't tie it to the workingcommitctx used here.) Therefore, I looked at
the passed context for 'AMR' status. But that doesn't help with the normal case
where the policy file is tracked, but clean. That requires looking up p1() to
read the file. I don't see any way to get the content of one file without first
creating the full parent context.
Matt Harbison <matt_harbison@yahoo.com> [Sun, 14 Jan 2018 01:04:45 -0500] rev 35664
lfs: move the tracked file function creation to a method
Once a commitable file format for tracked config is agreed upon, I can't see any
reason to have a config based way to control this. (Other than convert. That
will be necessary to override the file when converting to normal files. Also,
converting to lfs needs this if not splicing the file in at the beginning. So
maybe the existing config option should be `convert` specific.) Looking to
hgeol for precedent, it looks like policy that affects how items are stored are
handled only by the tracked file, while policy that affects the checkout can be
handled by either a user config or the tracked file (but the latter takes
precedence).
We probably need a transition period, so this transition policy can be
controlled by the function. Additionally, it provides a place for convert to
wrap to override the file based config.
Matt Harbison <matt_harbison@yahoo.com> [Tue, 16 Jan 2018 20:15:07 -0500] rev 35663
lfs: allow the pointer file to be viewed with `hg cat -T '{rawdata}'`
The only other interface to this data is `hg debugdata`, which requires
knowledge of the filelog revision that corresponds to the changeset. Since the
data is uninterpreted, this is an important debugging capability, and needs to
be simpler to use than that.
For non-LFS files, this displays the regular data.
Alternately, we could forego the messy function extraction in the last patch if
this template keyword can just be added unconditionally.
Matt Harbison <matt_harbison@yahoo.com> [Tue, 16 Jan 2018 19:56:00 -0500] rev 35662
cat: factor out a function that populates the formatter
This will allow extensions to add data to the templater.
Jordi Gutiérrez Hermoso <jordigh@octave.org> [Sun, 14 Jan 2018 12:07:06 -0500] rev 35661
svnsubrepo: add new method _svnmissing
This is modelled after _gitmissing from dea6efdd7ec4 and also
necessary for svn. I'm not sure exactly how this hasn't been more of a
problem for svn until now, actually.
Jordi Gutiérrez Hermoso <jordigh@octave.org> [Sun, 14 Jan 2018 12:05:28 -0500] rev 35660
svnsubrepo: decorate dirty method with annotatesubrepoerror
This function invokes svn commands which can error out in any number
of ways, so it's helpful to know in which subrepo this error happens.
Yuya Nishihara <yuya@tcha.org> [Tue, 16 Jan 2018 22:14:33 +0900] rev 35659
match: do not weirdly include explicit files excluded by -X option
Actually, this was the original behavior. Before a83a7d27911e, "log" and
"files" showed nothing if "FILE -X FILE" was specified, whereas "debugwalk"
got confused by an explicit FILE pattern. Under the hood, "log" and "files"
use m() and ctx.matches(m) respectively, and "debugwalk" uses ctx.walk(m).
I suspect dirstate.walk() goes wrong in _walkexplicit(), which seems to
blindly trust m.files().
I reckon the original "log"/"files" behavior is correct, and drop the hack
from the differencematcher.
Matt Harbison <matt_harbison@yahoo.com> [Sun, 14 Jan 2018 15:56:22 -0500] rev 35658
lfs: add the '{oid}' template keyword to '{lfs_files}'
The 'sha256:' prefix is skipped because this seems like the most convenient way
to consume it. Maybe we should also add a '{oid_type}' keyword? Then again,
that can be added in the future if a different algorithm is supported.
Matt Harbison <matt_harbison@yahoo.com> [Sun, 14 Jan 2018 15:39:10 -0500] rev 35657
lfs: convert '{lfs_files}' keyword to a hybrid list
This will allow more attributes about the file to be queried.
Boris Feld <boris.feld@octobus.net> [Sun, 10 Dec 2017 21:42:33 +0100] rev 35656
showstack: add an extension docstring
Now, running `hg help showstack` will give details on how to use the extension.
Boris Feld <boris.feld@octobus.net> [Mon, 15 Jan 2018 10:44:49 +0000] rev 35655
bookmark: run 'pushkey' hooks after bookmark move, not 'prepushkey'
This was a silly copy paste mistake.
Spotted by Mitchell Plamann from Jane Street.
Anton Shestakov <av6@dwimlabs.net> [Mon, 15 Jan 2018 19:44:18 +0800] rev 35654
hgweb: remove unused second argument of nextPageVarGet()
nextPageVarGet is a function that's used in ajaxScrollInit() to produce URL of
the next page. Before f84b01257e06, its second argument previousVal was a
number on /graph pages, and the code was simply adding 60 to it and returning
the resulting value. Now previousVal can only be a string containing changeset
hash, which can't be used the same way (and in fact isn't used in any way).
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 14 Jan 2018 20:06:56 -0800] rev 35653
dispatch: handle IOError when writing to stderr
Previously, attempts to write to stderr in dispatch.run() may lead to
an exception being thrown. This would likely be handled by Python's
default exception handler, which would print the exception and exit
1.
Code in this function is already catching IOError for stdout failures
and converting to exit code 255 (-1 & 255 == 255). Why we weren't
doing the same for stderr for the sake of consistency, I don't know.
I do know that chg and hg diverged in behavior here (as the changed
test-basic.t shows).
After this commit, we catch I/O failure on stderr and change the
exit code to 255. chg and hg now behave consistently. As a bonus,
Rust hg also now passes this test.
I'm skeptical at changing the exit code due to failures this late
in the process. I think we should consider preserving the current
exit code - assuming it is non-0. And, we may want to preserve the
exit code completely if the I/O error is EPIPE (and potentially
other special error classes). There's definitely room to tweak
behavior. But for now, let's at least prevent the uncaught exception.
Differential Revision: https://phab.mercurial-scm.org/D1860
Gregory Szorc <gregory.szorc@gmail.com> [Sun, 14 Jan 2018 19:30:48 -0800] rev 35652
commandserver: restore cwd in case of exception
The order of the statements was also changed a bit. But it shouldn't
matter.
Differential Revision: https://phab.mercurial-scm.org/D1859
Gregory Szorc <gregory.szorc@gmail.com> [Wed, 10 Jan 2018 20:02:35 -0800] rev 35651
tests: make hg frame optional
When `hg` is a Rust binary, the `hg` frame doesn't exist because an
`hg` Python script doesn't exist.
This commit updates expected test output to make the `hg` frame
optional.
There /might/ be a way to do this more accurately with the
"(feature !)" syntax in .t files. However, I poked at it for a
few minutes and couldn't get it to work. Worst case with using
(?) is we drop the frame from output for Python `hg`. The `hg`
frame isn't terribly important. So the worst case doesn't feel that
bad. If someone wants to enlighten me on how to use "(feature !)"
for optional output based on hghave features, I'd be more than
willing to update this.
Differential Revision: https://phab.mercurial-scm.org/D1858
Matt Harbison <matt_harbison@yahoo.com> [Sat, 13 Jan 2018 22:40:33 -0500] rev 35650
test-lfs: add tests to show that hashes remain unchanged by conversions
This is a very cool feature that we should document, but I'll punt that to the
freeze. From what I can tell, git doesn't have this capability.
Matt Harbison <matt_harbison@yahoo.com> [Sat, 13 Jan 2018 22:29:18 -0500] rev 35649
test-lfs: drop an unresolved issue note, now that lfs.track=none() is a thing
Matt Harbison <matt_harbison@yahoo.com> [Sat, 13 Jan 2018 20:07:14 -0500] rev 35648
lfs: always exclude '.hg*' text files
I can't think of any problematic scenarios (though things might get interesting
with .hgtags, since every head is consulted). The eol extension explicitly
disables handling these files, and that seems reasonable here too.
Pulkit Goyal <7895pulkit@gmail.com> [Mon, 15 Jan 2018 00:16:11 +0530] rev 35647
bookmarks: calculate visibility exceptions only once
In the loop "for mark in names", the rev is same in each iteration, so it does
not makes sense to call unhidehashlikerevs multiple times. Thanks to Yuya for
spotting this.
Yuya Nishihara <yuya@tcha.org> [Sun, 10 Dec 2017 18:25:33 +0900] rev 35646
log: rewrite --follow-first -rREV like --follow for consistency (BC)
This helps fixing the "--follow -rREV PATH" issue.
.. bc::
``log --follow-first -rREV``, which is deprecated, now follows the first
parent of merge revisions from the specified ``REV`` just like
``log --follow -rREV``.
Yuya Nishihara <yuya@tcha.org> [Tue, 02 Jan 2018 17:37:01 +0900] rev 35645
log: use revsetlang.formatspec() thoroughly
This patch replaces %(val)s and %(val)r with %r (expression) and %s (string)
respectively. _matchfiles() is the exception as it takes a list of string
parameters.
"--prune REV" could take a revset expression if it were "ancestors(%(val)s)",
but this patch doesn't change the existing behavior.
Yuya Nishihara <yuya@tcha.org> [Tue, 02 Jan 2018 17:13:18 +0900] rev 35644
log: use revsetlang.formatspec() to concatenate list expression
This rewrites 'not ancestors(x) and not ...' as 'not (ancestors(x) or ...)'
so we can use '%lr'. 'isinstance(val, list)' is replaced with 'listop' to
make sure 'listop' is applied.
Yuya Nishihara <yuya@tcha.org> [Tue, 02 Jan 2018 17:00:48 +0900] rev 35643
log: simplify 'x or ancestors(x)' expression
'ancestors(x)' includes 'x'.
Yuya Nishihara <yuya@tcha.org> [Tue, 02 Jan 2018 16:58:37 +0900] rev 35642
log: make opt2revset table a module constant
Just makes it clear that the table isn't updated in _makelogrevset().
Paul Morelle <paul.morelle@octobus.net> [Sun, 14 Jan 2018 13:04:26 -0800] rev 35641
revlog: group revision info into a dedicated structure
Paul Morelle <paul.morelle@octobus.net> [Sun, 14 Jan 2018 13:01:35 -0800] rev 35640
revlog: rename 'rev' to 'base', as it is the base revision
Paul Morelle <paul.morelle@octobus.net> [Sun, 14 Jan 2018 12:59:46 -0800] rev 35639
revlog: separate diff computation from the collection of other info
Paul Morelle <paul.morelle@octobus.net> [Fri, 12 Jan 2018 18:58:44 +0100] rev 35638
revlog: introduce 'deltainfo' to distinguish from 'delta'
A 'delta' is a binary diff between two revisions, as returned by revdiff.
A 'deltainfo' is an object storing information about a delta, including
the 'delta' itself.
Formerly, it was stored in a 7-position tuple, which was less readable.
Paul Morelle <paul.morelle@octobus.net> [Fri, 12 Jan 2018 18:10:03 +0100] rev 35637
revlog: extract 'builddelta' closure function from _addrevision
Paul Morelle <paul.morelle@octobus.net> [Fri, 12 Jan 2018 15:55:25 +0100] rev 35636
revlog: extract 'buildtext' closure function from _addrevision
Paul Morelle <paul.morelle@octobus.net> [Sun, 14 Jan 2018 12:49:24 -0800] rev 35635
revlog: choose between ifh and dfh once for all
Paul Morelle <paul.morelle@octobus.net> [Sun, 14 Jan 2018 12:46:03 -0800] rev 35634
revlog: refactor out the selection of candidate revisions
The new function will be useful to retrieve all the revisions which will be
needed to determine the best delta, and parallelize the computation of the
necessary diffs.
Pulkit Goyal <7895pulkit@gmail.com> [Sat, 30 Dec 2017 00:13:56 +0530] rev 35633
py3: use email.parser module to parse email messages
Before this patch we use email.Parser.Parser() from the email module which is
not available on Python 3.
On Python 2:
>>> import email
>>> import email.parser as emailparser
>>> email.Parser.Parser is emailparser.Parser
True
Yuya Nishihara <yuya@tcha.org> [Fri, 12 Jan 2018 22:18:42 +0900] rev 35632
rust: convert Unix path to CString transparently
On Unix, path is just a sequence of bytes. We shouldn't convert it to UTF-8
string.
Yuya Nishihara <yuya@tcha.org> [Fri, 12 Jan 2018 22:09:34 +0900] rev 35631
rust: extract function to convert Path to platform CString
It can be better on Unix.
Boris Feld <boris.feld@octobus.net> [Mon, 09 Oct 2017 11:46:27 +0200] rev 35630
pylint: add a check for multiple statement on a single line
Now that they are all cleaned up, we can install a check so that no new
occurrence appears.
Differential Revision: https://phab.mercurial-scm.org/D1834
Boris Feld <boris.feld@octobus.net> [Mon, 09 Oct 2017 11:45:21 +0200] rev 35629
style: remove multiple statement on a single line in zeroconf
Differential Revision: https://phab.mercurial-scm.org/D1833
Boris Feld <boris.feld@octobus.net> [Mon, 09 Oct 2017 11:44:02 +0200] rev 35628
style: remove multiple statement on a single line
This is similar to the change made in 5326e4ef1dab.
Differential Revision: https://phab.mercurial-scm.org/D1832