The purpose is to allow mixed usage of Rietveld/Gerrit in the same repo.
There, a branch can use Gerrit, even though repo's default is Rietveld.
The only remaining repo-wide check GetIsGerrit is to determine codereview
for a branch which has no uploads yet.
This also simplifies future test expectations.
R=andybons@chromium.org,bauerb@chromium.org
BUG=598681
Review URL: https://codereview.chromium.org/1851663002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299603 0039d316-1c4b-4281-b951-d872f2087c98
This adds pluggable codereview-specific implementations into
Changelist class. The specific implementation is chosen at
Changelist automatically, with Rietveld being default for
backwards compatibility.
Gerrit implementation for Gerrit is incomplete, and will be
added in later CLs. However, it is sufficient to ensure
current functionality of this tool is not diminished.
Sadly, the base class isn't completely free from Rietveld
assumptions because of presubmit_support. Apparently, PRESUBMIT
scripts can make use of Rietveld instance for RPCs directly.
This use doesn't make sense for Gerrit, which substitutes
rietveld instance with a dummy object, which raises exception
on any attribute access with a diagnostic message.
This also includes refactoring of some related code which
(ab)used ChangeList. Overall, this CL adds a few extra call to
git config in order to determine which codereview to use, but
but it shouldn't have any performance impact.
These is a reland of these 4 CLs + with several fixes.
patch from issue 1827523003 at patchset 20001 (http://crrev.com/1827523003#ps20001)
patch from issue 1830703004 at patchset 1 (http://crrev.com/1830703004#ps1)
patch from issue 1830923002 at patchset 60001 (http://crrev.com/1830923002#ps60001)
patch from issue 1805193002 at patchset 380001 (http://crrev.com/1805193002#ps380001)
This CL without a fix was also committed and reverted as
patch from issue 1830973003 at patchset 40001 (http://crrev.com/1830973003#ps40001)
R=machenbach@chromium.org,sergiyb@chromium.org,andybons@chromium.org
BUG=579160,597638
Review URL: https://codereview.chromium.org/1838143002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299530 0039d316-1c4b-4281-b951-d872f2087c98
Reason for revert:
argh, CQ is way too quick.
Original issue's description:
> Reland of git cl: Rework Changelist class for Rietveld/Gerrit use. (patchset #1 id:1 of https://codereview.chromium.org/1840833002/ )
>
> Reason for revert:
> with a fix.
>
> Original issue's description:
> > Revert of git cl: Rework Changelist class for Rietveld/Gerrit use. (patchset #3 id:40001 of https://codereview.chromium.org/1830973003/ )
> >
> > Reason for revert:
> > Speculative revert, see crbug.com/598428.
> >
> > Original issue's description:
> > > git cl: Rework Changelist class for Rietveld/Gerrit use.
> > >
> > > This adds pluggable codereview-specific implementations into
> > > Changelist class. The specific implementation is chosen at
> > > Changelist automatically, with Rietveld being default for
> > > backwards compatibility.
> > >
> > > Gerrit implementation for Gerrit is incomplete, and will be
> > > added in later CLs. However, it is sufficient to ensure
> > > current functionality of this tool is not diminished.
> > >
> > > Sadly, the base class isn't completely free from Rietveld
> > > assumptions because of presubmit_support. Apparently, PRESUBMIT
> > > scripts can make use of Rietveld instance for RPCs directly.
> > > This use doesn't make sense for Gerrit, which substitutes
> > > rietveld instance with a dummy object, which raises exception
> > > on any attribute access with a diagnostic message.
> > >
> > > This also includes refactoring of some related code which
> > > (ab)used ChangeList. Overall, this CL adds a few extra call to
> > > git config in order to determine which codereview to use, but
> > > but it shouldn't have any performance impact.
> > >
> > >
> > >
> > > These is a reland of these 4 CLs + a fix.
> > > patch from issue 1827523003 at patchset 20001 (http://crrev.com/1827523003#ps20001)
> > > patch from issue 1830703004 at patchset 1 (http://crrev.com/1830703004#ps1)
> > > patch from issue 1830923002 at patchset 60001 (http://crrev.com/1830923002#ps60001)
> > > patch from issue 1805193002 at patchset 380001 (http://crrev.com/1805193002#ps380001)
> > >
> > >
> > >
> > > R=machenbach@chromium.org,sergiyb@chromium.org,andybons@chromium.org
> > > BUG=579160,597638
> > >
> > > Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=299506
> >
> > TBR=andybons@chromium.org,machenbach@chromium.org,sergiyb@chromium.org,tandrii@chromium.org
> > # Skipping CQ checks because original CL landed less than 1 days ago.
> > NOPRESUBMIT=true
> > NOTREECHECKS=true
> > NOTRY=true
> > BUG=579160,597638
> >
> > Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=299515
>
> TBR=andybons@chromium.org,machenbach@chromium.org,sergiyb@chromium.org,dnj@chromium.org
> # Skipping CQ checks because original CL landed less than 1 days ago.
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=579160,597638
>
> Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=299528
TBR=andybons@chromium.org,machenbach@chromium.org,sergiyb@chromium.org,dnj@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=579160,597638
Review URL: https://codereview.chromium.org/1839973002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299529 0039d316-1c4b-4281-b951-d872f2087c98
Reason for revert:
with a fix.
Original issue's description:
> Revert of git cl: Rework Changelist class for Rietveld/Gerrit use. (patchset #3 id:40001 of https://codereview.chromium.org/1830973003/ )
>
> Reason for revert:
> Speculative revert, see crbug.com/598428.
>
> Original issue's description:
> > git cl: Rework Changelist class for Rietveld/Gerrit use.
> >
> > This adds pluggable codereview-specific implementations into
> > Changelist class. The specific implementation is chosen at
> > Changelist automatically, with Rietveld being default for
> > backwards compatibility.
> >
> > Gerrit implementation for Gerrit is incomplete, and will be
> > added in later CLs. However, it is sufficient to ensure
> > current functionality of this tool is not diminished.
> >
> > Sadly, the base class isn't completely free from Rietveld
> > assumptions because of presubmit_support. Apparently, PRESUBMIT
> > scripts can make use of Rietveld instance for RPCs directly.
> > This use doesn't make sense for Gerrit, which substitutes
> > rietveld instance with a dummy object, which raises exception
> > on any attribute access with a diagnostic message.
> >
> > This also includes refactoring of some related code which
> > (ab)used ChangeList. Overall, this CL adds a few extra call to
> > git config in order to determine which codereview to use, but
> > but it shouldn't have any performance impact.
> >
> >
> >
> > These is a reland of these 4 CLs + a fix.
> > patch from issue 1827523003 at patchset 20001 (http://crrev.com/1827523003#ps20001)
> > patch from issue 1830703004 at patchset 1 (http://crrev.com/1830703004#ps1)
> > patch from issue 1830923002 at patchset 60001 (http://crrev.com/1830923002#ps60001)
> > patch from issue 1805193002 at patchset 380001 (http://crrev.com/1805193002#ps380001)
> >
> >
> >
> > R=machenbach@chromium.org,sergiyb@chromium.org,andybons@chromium.org
> > BUG=579160,597638
> >
> > Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=299506
>
> TBR=andybons@chromium.org,machenbach@chromium.org,sergiyb@chromium.org,tandrii@chromium.org
> # Skipping CQ checks because original CL landed less than 1 days ago.
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=579160,597638
>
> Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=299515
TBR=andybons@chromium.org,machenbach@chromium.org,sergiyb@chromium.org,dnj@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=579160,597638
Review URL: https://codereview.chromium.org/1844523002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299528 0039d316-1c4b-4281-b951-d872f2087c98
Reason for revert:
Speculative revert, see crbug.com/598428.
Original issue's description:
> git cl: Rework Changelist class for Rietveld/Gerrit use.
>
> This adds pluggable codereview-specific implementations into
> Changelist class. The specific implementation is chosen at
> Changelist automatically, with Rietveld being default for
> backwards compatibility.
>
> Gerrit implementation for Gerrit is incomplete, and will be
> added in later CLs. However, it is sufficient to ensure
> current functionality of this tool is not diminished.
>
> Sadly, the base class isn't completely free from Rietveld
> assumptions because of presubmit_support. Apparently, PRESUBMIT
> scripts can make use of Rietveld instance for RPCs directly.
> This use doesn't make sense for Gerrit, which substitutes
> rietveld instance with a dummy object, which raises exception
> on any attribute access with a diagnostic message.
>
> This also includes refactoring of some related code which
> (ab)used ChangeList. Overall, this CL adds a few extra call to
> git config in order to determine which codereview to use, but
> but it shouldn't have any performance impact.
>
>
>
> These is a reland of these 4 CLs + a fix.
> patch from issue 1827523003 at patchset 20001 (http://crrev.com/1827523003#ps20001)
> patch from issue 1830703004 at patchset 1 (http://crrev.com/1830703004#ps1)
> patch from issue 1830923002 at patchset 60001 (http://crrev.com/1830923002#ps60001)
> patch from issue 1805193002 at patchset 380001 (http://crrev.com/1805193002#ps380001)
>
>
>
> R=machenbach@chromium.org,sergiyb@chromium.org,andybons@chromium.org
> BUG=579160,597638
>
> Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=299506
TBR=andybons@chromium.org,machenbach@chromium.org,sergiyb@chromium.org,tandrii@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=579160,597638
Review URL: https://codereview.chromium.org/1840833002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299515 0039d316-1c4b-4281-b951-d872f2087c98
This adds pluggable codereview-specific implementations into
Changelist class. The specific implementation is chosen at
Changelist automatically, with Rietveld being default for
backwards compatibility.
Gerrit implementation for Gerrit is incomplete, and will be
added in later CLs. However, it is sufficient to ensure
current functionality of this tool is not diminished.
Sadly, the base class isn't completely free from Rietveld
assumptions because of presubmit_support. Apparently, PRESUBMIT
scripts can make use of Rietveld instance for RPCs directly.
This use doesn't make sense for Gerrit, which substitutes
rietveld instance with a dummy object, which raises exception
on any attribute access with a diagnostic message.
This also includes refactoring of some related code which
(ab)used ChangeList. Overall, this CL adds a few extra call to
git config in order to determine which codereview to use, but
but it shouldn't have any performance impact.
These is a reland of these 4 CLs + a fix.
patch from issue 1827523003 at patchset 20001 (http://crrev.com/1827523003#ps20001)
patch from issue 1830703004 at patchset 1 (http://crrev.com/1830703004#ps1)
patch from issue 1830923002 at patchset 60001 (http://crrev.com/1830923002#ps60001)
patch from issue 1805193002 at patchset 380001 (http://crrev.com/1805193002#ps380001)
R=machenbach@chromium.org,sergiyb@chromium.org,andybons@chromium.org
BUG=579160,597638
Review URL: https://codereview.chromium.org/1830973003
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299506 0039d316-1c4b-4281-b951-d872f2087c98
Reason for revert:
Broke one project presubmit.
Original issue's description:
> git cl: Rework Changelist class for Rietveld/Gerrit use.
>
> This adds pluggable codereview-specific implementations into
> Changelist class. The specific implementation is chosen at
> Changelist automatically, with Rietveld being default for
> backwards compatibility.
>
> Gerrit implementation for Gerrit is incomplete, and will be
> added in later CLs. However, it is sufficient to ensure
> current functionality of this tool is not diminished.
>
> Sadly, the base class isn't completely free from Rietveld
> assumptions because of presubmit_support. Apparently, PRESUBMIT
> scripts can make use of Rietveld instance for RPCs directly.
> This use doesn't make sense for Gerrit, which substitutes
> rietveld instance with a dummy object, which raises exception
> on any attribute access with a diagnostic message.
>
> This also includes refactoring of some related code which
> (ab)used ChangeList. Overall, this CL adds a few extra call to
> git config in order to determine which codereview to use, but
> but it shouldn't have any performance impact.
>
>
> BUG=579160
>
> Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=299462
TBR=phajdan.jr@chromium.org,andybons@chromium.org,machenbach@chromium.org,sergiyb@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=579160,597638
Review URL: https://codereview.chromium.org/1831813003
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299476 0039d316-1c4b-4281-b951-d872f2087c98
This adds pluggable codereview-specific implementations into
Changelist class. The specific implementation is chosen at
Changelist automatically, with Rietveld being default for
backwards compatibility.
Gerrit implementation for Gerrit is incomplete, and will be
added in later CLs. However, it is sufficient to ensure
current functionality of this tool is not diminished.
Sadly, the base class isn't completely free from Rietveld
assumptions because of presubmit_support. Apparently, PRESUBMIT
scripts can make use of Rietveld instance for RPCs directly.
This use doesn't make sense for Gerrit, which substitutes
rietveld instance with a dummy object, which raises exception
on any attribute access with a diagnostic message.
This also includes refactoring of some related code which
(ab)used ChangeList. Overall, this CL adds a few extra call to
git config in order to determine which codereview to use, but
but it shouldn't have any performance impact.
BUG=579160
Review URL: https://codereview.chromium.org/1805193002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299462 0039d316-1c4b-4281-b951-d872f2087c98
Special case when the change description has no footers, but looks
like a footer.
R=machenbach@chromium.org,andybons@chromium.org
BUG=579176
Review URL: https://codereview.chromium.org/1812803002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299343 0039d316-1c4b-4281-b951-d872f2087c98
This allows to generate new IDs on the fly without installing
commit-msg hook and works just fine on Windows.
BUG=579183
Review URL: https://codereview.chromium.org/1757133002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299094 0039d316-1c4b-4281-b951-d872f2087c98
This should insert the message according to Gerrit's own commit-msg
hook implementation.
BUG=579183
Review URL: https://codereview.chromium.org/1758943002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@299093 0039d316-1c4b-4281-b951-d872f2087c98
Added --ignore-file argument, so you can specify ignored commits in a
file rather than as raw command-line arguments. Also, automatically
searches for a file called .git-blame-ignore-revs, which is
automatically used as an ignore list by default.
Also, specifying an unknown revision (either on the command line or in a
file) now generates a warning, not an error.
Notes on some decisions:
- The file is called .git-blame-ignore-revs (not mentioning hyper-blame)
because we may use the same list in tools other than hyper-blame in
the future.
- We look at the *currently checked out* version of
.git-blame-ignore-revs (not the version at the specified revision) for
consistency with .git-ignore. Because we only expect revisions to be
added (not deleted), it should be fine to use an ignore list from a
newer version than the revision being blamed.
- We considered using git notes for the ignore list so that you could
add a revision to the ignore list without needing a follow-up CL.
However, there are some problems with this approach. git notes is not
automatically synced with git clone/pull. Also the Chromium infra
tools (Reitveld, CQ) are not set up to allow modification of git
notes, nor are changes to git notes subject to OWNERS checks. Using a
regular file ensures all users synced to a particular revision are
using the same ignore list.
BUG=574290
Review URL: https://codereview.chromium.org/1697423004
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@298897 0039d316-1c4b-4281-b951-d872f2087c98
\n doesn't work on Windows, and %B is shorter anyway.
BUG=586344
Review URL: https://codereview.chromium.org/1705193003
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@298845 0039d316-1c4b-4281-b951-d872f2087c98
This will hopefully make Rietveld._send retry 500s like it promises to
BUG=585632
Review URL: https://codereview.chromium.org/1681333005
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@298726 0039d316-1c4b-4281-b951-d872f2087c98
This will hopefully make Rietveld._send retry 500s like it promises to
BUG=
Review URL: https://codereview.chromium.org/1683603002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@298688 0039d316-1c4b-4281-b951-d872f2087c98
Previously, when a commit was skipped, it would be blamed on the line
number the line had *after* the skipped commit. This could mean a
totally unrelated commit gets blamed. Now, a heuristic analyses the diff
of the skipped commit to discover approximately what line number the
line had *before* the skipped commit, so it can hopefully be blamed on
the right commit.
BUG=574290
Review URL: https://codereview.chromium.org/1629253002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@298609 0039d316-1c4b-4281-b951-d872f2087c98
Currently, the script requires you to pass the unwanted commits on the
command line, but eventually, you could configure it with a file
(checked into the repo) that provides a fixed set of commits to always
skip (such as commits that do a huge amount of renaming and nothing
else).
BUG=574290
Review URL: https://codereview.chromium.org/1559943003
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@298544 0039d316-1c4b-4281-b951-d872f2087c98
This fixes the test being dependent on the system time, and undefined
behaviour resulting from negative timestamps in positive-offset
timezones.
BUG=581895
Review URL: https://codereview.chromium.org/1640973002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@298438 0039d316-1c4b-4281-b951-d872f2087c98
This is a good prototypical example of how to do it :-).
TBR for OWNERS change for recipes.cfg.
BUG=564920
R=iannucci@chromium.org,martiniss@chromium.org
TBR=phajdan.jr@chromium.org
Review URL: https://codereview.chromium.org/1494103004
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@297851 0039d316-1c4b-4281-b951-d872f2087c98
Fix a bug that left branches tracking a dead branch if both their parent
and grand-parent were left with no changes after a "rebase-update" step.
Given the following initial state:
$ git map-branches -v
origin/master
a
b
c * [ ahead 1 ]
without this patch, a "git rebase-update" on this tree state would
leave the branch "c" as tracking a non-existing branch "a":
$ git recursive-rebase
a up-to-date
b up-to-date
c up-to-date
Reparented c to track a (was tracking b)
Deleted branch b (was 448d1da).
Deleted branch a (was 448d1da).
$ git map-branches -v
{a:GONE}
c *
with the patch, we record that the branch "c" is tracking must be
updated twice and we end up in a state were "c" is correctly tracking
"origin/master":
$ git recursive-rebase
a up-to-date
b up-to-date
c up-to-date
Reparented c to track origin/master (was tracking b)
Deleted branch b (was 448d1da).
Deleted branch a (was 448d1da).
$ git map-branches -v
origin/master
c * [ ahead 1 ]
BUG=456806
Review URL: https://codereview.chromium.org/1482753002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@297792 0039d316-1c4b-4281-b951-d872f2087c98
Manually generating param strings shows that the 1/2 values do not work.
Review URL: https://codereview.chromium.org/1482153002
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@297732 0039d316-1c4b-4281-b951-d872f2087c98
Reason for revert:
breaks git-bash on windows
Original issue's description:
> git_cl/gclient: use python2
>
> Newer distros are defaulting /usr/bin/python to python3, so use python2
> explicitly so we continue working.
>
> Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=297535
TBR=stip@chromium.org,sergeyberezin@chromium.org,vapier@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/1442583004
git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/depot_tools@297544 0039d316-1c4b-4281-b951-d872f2087c98
tandrii@chromium.org
dnj@chromium.org
mgiuca@chromium.org
bauerb@chromium.org
hinoka@google.com
hinoka@chromium.org
dsansome@chromium.org
thakis@chromium.org
andybons@chromium.org
danakj@chromium.org
luqui@chromium.org
iannucci@chromium.org
sdefresne@chromium.org
cbentzel@chromium.org
thestig@chromium.org
phajdan.jr@chromium.org