D483: tests: fixed a bytes/unicode confusion in the test runner

alex.gaynor created this revision. Herald added a subscriber: mercurial-devel. Herald added a reviewer: hg-reviewers. REPOSITORY rHG Mercurial BRANCH py3k-test (bookmark) on default (branch) REVISION DETAIL https://phab.mercurial-scm.org/D483 AFFECTED FILES tests/run-tests.py CHANGE

D482: bundle2: fixed usage of an attribute that was removed in py3k

indygreg accepted this revision. indygreg added a comment. This revision is now accepted and ready to land. The commit message should likely say something about how the existing use of `string.letters` was logically wrong, since it is locale dependent. `string.ascii_letters` is definitely

Re: D464: Use an unambigious path suffix for the commit editor file.

Jun Wu writes: > I think this patch is about changing the suffix. I guess some less powerful > editors may only support matching file types by suffixes. Yeah, it seems that way. > Please correct me if I'm wrong, but I failed to see how the > "experimental.editortmpinhg" setting

D451: revset: remove order information from tree

quark added inline comments. INLINE COMMENTS > quark wrote in revset.py:55 > Agree `anyorder` could be a surprise. I was trying to optimize aggressively. > Maybe `followorder` is a better default since `subset & x` is the old default. Actually, `defineorder` as default also makes sense and

[PATCH 2 of 7 v2] md5sum: assume hashlib exists now that we're 2.7 only

# HG changeset patch # User Augie Fackler # Date 1503428230 14400 # Tue Aug 22 14:57:10 2017 -0400 # Node ID 6a9a8a60bced4b42a4b019b2f98c97672ce7a42b # Parent 536b3969d2242224898f574919b134df4ba190a8 md5sum: assume hashlib exists now that we're 2.7 only diff --git

[PATCH 7 of 7 v2] python3: whitelist test-imports-checker.t, which now passes

# HG changeset patch # User Augie Fackler # Date 1503414046 14400 # Tue Aug 22 11:00:46 2017 -0400 # Node ID 2959ea035b9d1ddfe41dc48c2893a9ebd6c2f160 # Parent fb3748bd846a240924cbf6172162dbe6851514e5 python3: whitelist test-imports-checker.t, which now passes diff --git

[PATCH 6 of 7 v2] contrib: make import checker always think in terms of bytes

# HG changeset patch # User Augie Fackler # Date 1503421108 14400 # Tue Aug 22 12:58:28 2017 -0400 # Node ID fb3748bd846a240924cbf6172162dbe6851514e5 # Parent 8dd9a90a0dede651e23211a772732ed19e5a29c5 contrib: make import checker always think in terms of bytes The doctests

[PATCH 3 of 7 v2] tests: update test-archive to always use hashlib

# HG changeset patch # User Augie Fackler # Date 1503428272 14400 # Tue Aug 22 14:57:52 2017 -0400 # Node ID 97dba357fcdb6caf5a0f9f5f074a69306a5a8950 # Parent 6a9a8a60bced4b42a4b019b2f98c97672ce7a42b tests: update test-archive to always use hashlib We don't need the

[PATCH 1 of 7 v2] undumprevlog: update to valid Python 3 syntax

# HG changeset patch # User Augie Fackler # Date 1503421131 14400 # Tue Aug 22 12:58:51 2017 -0400 # Node ID 536b3969d2242224898f574919b134df4ba190a8 # Parent af20468eb0a499c094dbd6e27ffcacf54cf5a8e6 undumprevlog: update to valid Python 3 syntax I didn't do anything to

[PATCH 4 of 7 v2] tests: stop using old except syntax in test-bundle2-format

# HG changeset patch # User Augie Fackler # Date 1503428707 14400 # Tue Aug 22 15:05:07 2017 -0400 # Node ID ccb38cfe8b6d7833490e9db28eda18968f27c490 # Parent 97dba357fcdb6caf5a0f9f5f074a69306a5a8950 tests: stop using old except syntax in test-bundle2-format diff --git

[PATCH 5 of 7 v2] contrib: work around some modules not existing on Py3 in import checker

# HG changeset patch # User Augie Fackler # Date 1503413909 14400 # Tue Aug 22 10:58:29 2017 -0400 # Node ID 8dd9a90a0dede651e23211a772732ed19e5a29c5 # Parent ccb38cfe8b6d7833490e9db28eda18968f27c490 contrib: work around some modules not existing on Py3 in import checker

D451: revset: remove order information from tree

quark added inline comments. INLINE COMMENTS > martinvonz wrote in test-revset.t:2893 > Does that mean you'll remove the other.sort() in fullreposet.__and__? In another way. With the new code (`anyorder` gets aggressively used), `fullrepo & xs` would be optimized to `xs & fullrepo` and the

[PATCH] obscache: use _readmarkers() from core with correct signature

# HG changeset patch # User Martin von Zweigbergk # Date 1503424759 25200 # Tue Aug 22 10:59:19 2017 -0700 # Node ID efc4b69aecfb3b407ca362cff31805e9672c # Parent 95470e817c00b03fcf99e486412cc7d7f0116681 obscache: use _readmarkers() from core with correct

D482: bundle2: fixed usage of an attribute that was removed in py3k

alex.gaynor created this revision. Herald added a subscriber: mercurial-devel. Herald added a reviewer: hg-reviewers. REPOSITORY rHG Mercurial BRANCH py3k-stuff (bookmark) on default (branch) REVISION DETAIL https://phab.mercurial-scm.org/D482 AFFECTED FILES mercurial/bundle2.py

[PATCH] copies: fix misaligned lines

# HG changeset patch # User Gábor Stefanik # Date 1503411399 -7200 # Tue Aug 22 16:16:39 2017 +0200 # Node ID 3b8c9025f9b966a3e58334024c3f7b3136f500fe # Parent e02a08d63ad20097419f935fdd8a91c42ecd88e3 copies: fix misaligned lines diff -r e02a08d63ad2 -r 3b8c9025f9b9

Re: [PATCH 7 of 9 V3] template: compute dates in obsfatedate

On Mon, 21 Aug 2017 10:43:59 +0200, Boris Feld wrote: > # HG changeset patch > # User Boris Feld > # Date 1499088850 -7200 > # Mon Jul 03 15:34:10 2017 +0200 > # Node ID faad6d683f7a30996007116d296df9ebf853c44d > # Parent 2fee3b06f7b09f35dcfa312b645cf94090c11fb9 > #

Re: [PATCH 5 of 9 V3] template: compute verb in obsfateverb

On Mon, 21 Aug 2017 10:43:57 +0200, Boris Feld wrote: > # HG changeset patch > # User Boris Feld > # Date 1499088807 -7200 > # Mon Jul 03 15:33:27 2017 +0200 > # Node ID 66a2b0407c8cffedfae4897bd7a2f0e2d6a22363 > # Parent 84a35f753f9ab500902f5c06ce3845ce85e5b5a9 > #

Re: [PATCH 4 of 9 V3] template: add minimal obsfate template function

On Mon, 21 Aug 2017 10:43:56 +0200, Boris Feld wrote: > # HG changeset patch > # User Boris Feld > # Date 1502987171 -7200 > # Thu Aug 17 18:26:11 2017 +0200 > # Node ID 84a35f753f9ab500902f5c06ce3845ce85e5b5a9 > # Parent 89a659e236b906e58a48dfed8a4e3b89372d7184 > #

Re: [PATCH 6 of 9 V3] template: compute user in obsfateusers

On Mon, 21 Aug 2017 10:43:58 +0200, Boris Feld wrote: > # HG changeset patch > # User Boris Feld > # Date 1499088840 -7200 > # Mon Jul 03 15:34:00 2017 +0200 > # Node ID 2fee3b06f7b09f35dcfa312b645cf94090c11fb9 > # Parent 66a2b0407c8cffedfae4897bd7a2f0e2d6a22363 > #

D476: util: add an mmapread method

mbthomas created this revision. Herald added a subscriber: mercurial-devel. Herald added a reviewer: hg-reviewers. REVISION SUMMARY This is useful for large files that are only partly touched. TEST PLAN Will be used and tested in a later patch. REPOSITORY rHG Mercurial BRANCH default

mercurial@33866: new changeset

New changeset in mercurial: https://www.mercurial-scm.org/repo/hg/rev/4e8a46c25fac changeset: 33866:4e8a46c25fac bookmark:@ tag: tip user:Matthieu Laneuville date:Tue Aug 22 11:00:00 2017 +0200 summary: run-tests: pre instantiate

D451: revset: remove order information from tree

yuja added a comment. In https://phab.mercurial-scm.org/D451#7257, @yuja wrote: > So, `anyorder` for `not x` would be my mistake because `x and not y` could be > theoretically flipped. FWIW, this should be okay since `not x` follows the order of the input subset. The order

D477: revlog: add option to mmap revlog index

mbthomas created this revision. Herald added a reviewer: indygreg. Herald added a subscriber: mercurial-devel. Herald added a reviewer: hg-reviewers. REVISION SUMMARY Following on from Jun Wu's patch last October[1], we have found that using mmap for the revlog index in repos with large

D451: revset: remove order information from tree

yuja added a comment. In https://phab.mercurial-scm.org/D451#7281, @quark wrote: > I think it's more correct if all core revsets support `defineorder` explicitly. The current code depends on `revset.makematcher` using ascending fullreposet as default to make `defineorder` implicitly

D451: revset: remove order information from tree

quark added inline comments. INLINE COMMENTS > yuja wrote in revset.py:55 > Perhaps the default `order=defineorder` would be safer > at this point. Agree `anyorder` could be a surprise. I was trying to optimize aggressively. Maybe `followorder` is a better default since `subset & x` is the old

[PATCH] copies: fix typo in comment

# HG changeset patch # User Gábor Stefanik # Date 1503410911 -7200 # Tue Aug 22 16:08:31 2017 +0200 # Node ID e02a08d63ad20097419f935fdd8a91c42ecd88e3 # Parent af20468eb0a499c094dbd6e27ffcacf54cf5a8e6 copies: fix typo in comment "will not be limited" was meant to be

Re: [PATCH] run-tests: pre instantiate pygments objects

On Tue, 22 Aug 2017 11:29:09 +0200, mlaneuvi...@gmail.com wrote: > # HG changeset patch > # User Matthieu Laneuville > # Date 1503392400 -7200 > # Tue Aug 22 11:00:00 2017 +0200 > # Node ID 8f0443200c166b82d42078362d6908f73c7fb58a > # Parent

D451: revset: remove order information from tree

quark added a comment. In https://phab.mercurial-scm.org/D451#7456, @yuja wrote: > Perhaps `_flipand(1:0, _flipand(1:0, 0::1))` would return `[1, 0]` if the input set > were reversed. IMHO, that's correct under the original design. I see. The old code allows "weak define" that

Re: [PATCH 4 of 6] i18n: use saved object to get actual function information if available

On Tue, 22 Aug 2017 04:18:17 +0900, FUJIWARA Katsunori wrote: > At Mon, 21 Aug 2017 21:46:35 +0900, > Yuya Nishihara wrote: > > > > On Wed, 16 Aug 2017 01:19:04 +0900, FUJIWARA Katsunori wrote: > > > # HG changeset patch > > > # User FUJIWARA Katsunori > > > # Date

[PATCH 4 of 7] tests: stop using old except syntax in test-bundle2-format

# HG changeset patch # User Augie Fackler # Date 1503428707 14400 # Tue Aug 22 15:05:07 2017 -0400 # Node ID ccb38cfe8b6d7833490e9db28eda18968f27c490 # Parent 97dba357fcdb6caf5a0f9f5f074a69306a5a8950 tests: stop using old except syntax in test-bundle2-format diff --git

[PATCH 3 of 7] tests: update test-archive to always use hashlib

# HG changeset patch # User Augie Fackler # Date 1503428272 14400 # Tue Aug 22 14:57:52 2017 -0400 # Node ID 97dba357fcdb6caf5a0f9f5f074a69306a5a8950 # Parent 6a9a8a60bced4b42a4b019b2f98c97672ce7a42b tests: update test-archive to always use hashlib We don't need the

[PATCH 1 of 7] undumprevlog: update to valid Python 3 syntax

# HG changeset patch # User Augie Fackler # Date 1503421131 14400 # Tue Aug 22 12:58:51 2017 -0400 # Node ID 536b3969d2242224898f574919b134df4ba190a8 # Parent af20468eb0a499c094dbd6e27ffcacf54cf5a8e6 undumprevlog: update to valid Python 3 syntax I didn't do anything to

[PATCH 6 of 7] contrib: make import checker always think in terms of bytes

# HG changeset patch # User Augie Fackler # Date 1503421108 14400 # Tue Aug 22 12:58:28 2017 -0400 # Node ID b093ed60c94fe3f7c4edaead1b0b0a5b7b8cddda # Parent d31f856bd5d31a1adc377f2679da764b85700bf0 contrib: make import checker always think in terms of bytes The doctests

[PATCH 5 of 7] contrib: work around some modules not existing on Py3 in import checker

# HG changeset patch # User Augie Fackler # Date 1503413909 14400 # Tue Aug 22 10:58:29 2017 -0400 # Node ID d31f856bd5d31a1adc377f2679da764b85700bf0 # Parent ccb38cfe8b6d7833490e9db28eda18968f27c490 contrib: work around some modules not existing on Py3 in import checker

[PATCH 7 of 7] python3: whitelist test-imports-checker.t, which now passes

# HG changeset patch # User Augie Fackler # Date 1503414046 14400 # Tue Aug 22 11:00:46 2017 -0400 # Node ID 3045a38b4c069b3d3b2f306a5bd7a2be624bedaf # Parent b093ed60c94fe3f7c4edaead1b0b0a5b7b8cddda python3: whitelist test-imports-checker.t, which now passes diff --git

[PATCH 2 of 7] md5sum: assume hashlib exists now that we're 2.7 only

# HG changeset patch # User Augie Fackler # Date 1503428230 14400 # Tue Aug 22 14:57:10 2017 -0400 # Node ID 6a9a8a60bced4b42a4b019b2f98c97672ce7a42b # Parent 536b3969d2242224898f574919b134df4ba190a8 md5sum: assume hashlib exists now that we're 2.7 only diff --git

D478: run-tests: remove unused pygments token type

martinvonz created this revision. Herald added a subscriber: mercurial-devel. Herald added a reviewer: hg-reviewers. REPOSITORY rHG Mercurial REVISION DETAIL https://phab.mercurial-scm.org/D478 AFFECTED FILES tests/run-tests.py CHANGE DETAILS diff --git a/tests/run-tests.py

D451: revset: remove order information from tree

martinvonz added a comment. I had also wanted to remove the need to pass subset, so I'd be happy so that change. INLINE COMMENTS > quark wrote in test-revset.t:2893 > This is caused by `fullreposet` having a default order. If we remove that, it > would be optimized to `` here. Does that

[Bug 5666] New: Config Phabricator In-bound emails

https://bz.mercurial-scm.org/show_bug.cgi?id=5666 Bug ID: 5666 Summary: Config Phabricator In-bound emails Product: Mercurial Version: 4.3-rc Hardware: All OS: All Status: UNCONFIRMED Severity: feature

Re: [PATCH 5 of 7] contrib: work around some modules not existing on Py3 in import checker

This patch flunks test-check-code.t, please disregard this series in favor of a forthcoming v2. > On Aug 22, 2017, at 3:08 PM, Augie Fackler wrote: > > # HG changeset patch > # User Augie Fackler > # Date 1503413909 14400 > # Tue Aug 22 10:58:29 2017

D451: revset: remove order information from tree

quark added a comment. Interestingly, I checked some well known 3rd-party extensions: hgsubversion: - svnrev() uses "x for x in myset if x in subset" therefore enforces "defineorder" and wrong. - fromsvn() is also wrong similarly. hg-git: - fromgit(), gitnode(): happened

Re: D464: Use an unambigious path suffix for the commit editor file.

I think this patch is about changing the suffix. I guess some less powerful editors may only support matching file types by suffixes. Please correct me if I'm wrong, but I failed to see how the "experimental.editortmpinhg" setting could be used as-is to change the suffix. Therefore I think the

D464: commit: use an unambiguous path suffix for the commit editor file

quark requested changes to this revision. quark added a comment. There are some discussion in the list (https://www.mercurial-scm.org/pipermail/mercurial-devel/2017-August/103382.html). Just mentioning it here since Phabricator In-Bound email handling is not configured properly. REPOSITORY

D472: extensions: make wrapfunction() return a context manager

quark added subscribers: mjpieters, quark. quark accepted this revision. quark added a comment. cc python expert @mjpieters to see if he has additional comments. But I think this is good enough. REPOSITORY rHG Mercurial REVISION DETAIL https://phab.mercurial-scm.org/D472 To:

Re: [PATCH 7 of 7 v2] python3: whitelist test-imports-checker.t, which now passes

This series looks good to me. :) On Wed, Aug 23, 2017 at 2:57 AM, Augie Fackler wrote: > # HG changeset patch > # User Augie Fackler > # Date 1503414046 14400 > # Tue Aug 22 11:00:46 2017 -0400 > # Node ID 2959ea035b9d1ddfe41dc48c2893a9ebd6c2f160 > #

[PATCH] run-tests: pre instantiate pygments objects

# HG changeset patch # User Matthieu Laneuville # Date 1503392400 -7200 # Tue Aug 22 11:00:00 2017 +0200 # Node ID 8f0443200c166b82d42078362d6908f73c7fb58a # Parent 3cfc9070245fbaa8c4c010327f24d579a416370f run-tests: pre instantiate pygments objects Pre

D464: commit: use an unambiguous path suffix for the commit editor file

ryanmce requested changes to this revision. ryanmce added a comment. This revision now requires changes to proceed. But you should include a test to prevent regressions. REPOSITORY rHG Mercurial REVISION DETAIL https://phab.mercurial-scm.org/D464 To: mbolin, quark, durin42,

D464: commit: use an unambiguous path suffix for the commit editor file

ryanmce accepted this revision. ryanmce added a comment. Should be safe and effective. REPOSITORY rHG Mercurial REVISION DETAIL https://phab.mercurial-scm.org/D464 To: mbolin, quark, durin42, #hg-reviewers, ryanmce Cc: ryanmce, mercurial-devel

D472: extensions: make wrapfunction() return a context manager

martinvonz created this revision. Herald added a subscriber: mercurial-devel. Herald added a reviewer: hg-reviewers. REVISION SUMMARY Several extensions exist that temporarily want to wrap a function (at least narrowhg, any many of the extensions in hg-experimental). That's why we have the

mercurial@33862: 2 new changesets (2 on stable)

2 new changesets (2 on stable) in mercurial: https://www.mercurial-scm.org/repo/hg/rev/0e15d5ae52cf changeset: 33861:0e15d5ae52cf branch: stable parent: 33755:cde4cfeb6e3e user:Martin von Zweigbergk date:Fri Aug 18 12:50:26 2017 -0700 summary:

mercurial@33865: 3 new changesets

3 new changesets in mercurial: https://www.mercurial-scm.org/repo/hg/rev/3160876c6e4e changeset: 33863:3160876c6e4e parent: 33860:3cfc9070245f user:Jun Wu date:Thu Aug 10 22:17:15 2017 -0700 summary: rebase: choose merge base without unwanted revisions