Using git daemon for smoke tests times out on Windows.
This CL adds support for local directory URLs to gclient so we can tests
using local directories as remotes instead.
Bug: 1024683
Change-Id: I6dcefa6eb0a93713cf46ea1e4c6b29311ad37565
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1915416
Reviewed-by: Anthony Polito <apolito@google.com>
Commit-Queue: Edward Lesmes <ehmaldonado@chromium.org>
Also add a better warning if 'gclient sync -D' fails due to a modified file.
I've tested this code via deleting a DEP and running gclient sync -D.
Bug: 981149
Change-Id: I97035ac238d163ccb1684c3ee423c223ed0f6299
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1891830
Reviewed-by: Edward Lesmes <ehmaldonado@chromium.org>
Commit-Queue: Anthony Polito <apolito@google.com>
This is a reland of 72a245e4c6
The "git checkout master --detach" operation has been modified to run in the solution
directory.
Original change's description:
> Fetch the input commit ref when both ref and id are specified.
>
> This causes --refs refs/head/master to be added to bot_update calls when
> the input build message contains a commit id and the refs/head/master
> ref. This is an effective no-op.
>
> When bot_update clones a git repository, it results in HEAD referring to
> master in contrast to the non-clone case which results in a detached
> HEAD. This prevents fetching refs/heads/master, so detach the HEAD after
> cloning.
>
> Bug: 922150
> Change-Id: I357a78305020218e9f5d350832a706b8f2b27967
> Recipe-Nontrivial-Roll: build
> Recipe-Nontrivial-Roll: build_limited_scripts_slave
> Recipe-Nontrivial-Roll: infra
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1900735
> Commit-Queue: Garrett Beaty <gbeaty@chromium.org>
> Reviewed-by: Edward Lesmes <ehmaldonado@chromium.org>
Bug: 922150
Change-Id: I7dc6ebed2dcbd880ff4d515f7e9fc6428ff24099
Recipe-Nontrivial-Roll: build
Recipe-Nontrivial-Roll: build_limited_scripts_slave
Recipe-Nontrivial-Roll: infra
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1910260
Reviewed-by: Edward Lesmes <ehmaldonado@chromium.org>
Commit-Queue: Garrett Beaty <gbeaty@chromium.org>
On Windows, scripts on the current directory take precedence over PATH, so
that when testing depot_tools on Windows, calling `vpython.bat` will
execute the copy of vpython of the depot_tools under test instead of the
one in the bot.
As a workaround, we run the tests from the parent directory instead.
Change-Id: I5187943b6cf7499d50b3f1a2ec906a5e8fd84835
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1910448
Reviewed-by: Anthony Polito <apolito@google.com>
Commit-Queue: Edward Lesmes <ehmaldonado@chromium.org>
git cl owners orders owners by score with alphabetization being the tie
breaker. This leads to some owners being suggested far more often than
others.
Adding a tiny amount of randomization to the scoring leads to an even
distribution of equally qualified reviewers. Less qualified reviewers
will still be sorted into distinct buckets - the randomness is too small
to do anything except break ties.
The tests were updated so that they can tolerate the randomness, but
only for breaking ties.
Bug: 1024083
Change-Id: If7d39d1b3bbd980b80b46ab3f62c65215309bdc8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1913642
Commit-Queue: Bruce Dawson <brucedawson@chromium.org>
Reviewed-by: Anthony Polito <apolito@google.com>
Reviewed-by: Edward Lesmes <ehmaldonado@chromium.org>
pythond and python3 are still downloaded from cipd, but python3 is no longer shim'd into
depot_tools on Linux, so is not available on PATH.
Modify PATH in vpython and vpython3 on Linux to point to the python3 download directory,
so that the version downloaded from cipd is used.
Bug: 1017812
Change-Id: I70bc8e7d460f60dcbfa5e141127edecfc2b44744
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1893982
Commit-Queue: Edward Lesmes <ehmaldonado@chromium.org>
Reviewed-by: Anthony Polito <apolito@google.com>
Reviewed-by: Mike Frysinger <vapier@chromium.org>
Reviewed-by: Dirk Pranke <dpranke@chromium.org>
The "CPU time" values printed by post_build_ninja_summary.py are not
actually measure of CPU time, they are measures of wall-clock time or
elapsed time. This just changes the labeling of the numbers - the actual
numbers printed are unchanged.
The "weighted time" number continues to be the more important number
since it is an approximation of how much the build step(s) are affecting
build time. See the comments in the script for details.
Change-Id: Ibdb4efdd327ece34492ab10337c234a826514197
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1900019
Commit-Queue: Bruce Dawson <brucedawson@chromium.org>
Reviewed-by: Dirk Pranke <dpranke@chromium.org>
This changes the behaviour when applying a patch to not fetch tags
along with the patch being applied. The fetch to apply a patch only
needs the commits that will be cherry picked, it does not need tags.
Fetching tags creates slow bottlenecks if a repo such as chromium has
10k+ tags. Using --no-tags prevents the server from sending the list
of all tags and the client from processing them.
Bug: 1019824
Change-Id: I756beb8c2b0c7ec3c0fc48b7431d0cf8c0bbbbd3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1894393
Reviewed-by: Edward Lesmes <ehmaldonado@chromium.org>
Reviewed-by: Erik Chen <erikchen@chromium.org>
Commit-Queue: danakj <danakj@chromium.org>
This reverts commit 72a245e4c6.
Reason for revert: crbug.com/1022617
Original change's description:
> Fetch the input commit ref when both ref and id are specified.
>
> This causes --refs refs/head/master to be added to bot_update calls when
> the input build message contains a commit id and the refs/head/master
> ref. This is an effective no-op.
>
> When bot_update clones a git repository, it results in HEAD referring to
> master in contrast to the non-clone case which results in a detached
> HEAD. This prevents fetching refs/heads/master, so detach the HEAD after
> cloning.
>
> Bug: 922150
> Change-Id: I357a78305020218e9f5d350832a706b8f2b27967
> Recipe-Nontrivial-Roll: build
> Recipe-Nontrivial-Roll: build_limited_scripts_slave
> Recipe-Nontrivial-Roll: infra
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1900735
> Commit-Queue: Garrett Beaty <gbeaty@chromium.org>
> Reviewed-by: Edward Lesmes <ehmaldonado@chromium.org>
TBR=ehmaldonado@chromium.org,gbeaty@chromium.org
Change-Id: Ibd6c6ed272b0c98f82be50cefa521aaee0444f11
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 922150, 1022617
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1904099
Reviewed-by: Stephen Martinis <martiniss@chromium.org>
Commit-Queue: Stephen Martinis <martiniss@chromium.org>
This causes --refs refs/head/master to be added to bot_update calls when
the input build message contains a commit id and the refs/head/master
ref. This is an effective no-op.
When bot_update clones a git repository, it results in HEAD referring to
master in contrast to the non-clone case which results in a detached
HEAD. This prevents fetching refs/heads/master, so detach the HEAD after
cloning.
Bug: 922150
Change-Id: I357a78305020218e9f5d350832a706b8f2b27967
Recipe-Nontrivial-Roll: build
Recipe-Nontrivial-Roll: build_limited_scripts_slave
Recipe-Nontrivial-Roll: infra
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1900735
Commit-Queue: Garrett Beaty <gbeaty@chromium.org>
Reviewed-by: Edward Lesmes <ehmaldonado@chromium.org>
There are 3 layers modified, starting from the bottom up:
1. git_cache.py populate
Now takes a --no-fetch-tags option. If specified, the cache will not
fetch updated tags from the server by passing --no-tags to git fetch.
This prevents the server from sending all the tag refs. In chromium
this prevents significant time bottlenecks dealing with 10k+ tags.
2. bot_update.py
bot_update has to deal with multiple git repos, it has the root repo
that is checked out through git-cache, as well as repos synched via
DEPS and gclient.
The script now takes a --no_fetch_tags option. If specified,
the git-cache populate call will include --no-fetch-tags. Otherwise, it
won't. So this controls (for chromium) if fetches to the src.git server
are passed --no-tags.
3. bot_update/api.py
This is the entry point for recipes and also has to deal with multiple
git repos. The behaviour at this point is not changed from the default.
A |no_fetch_tags| parameter is added to ensure_checkout() that defaults
to False.
This CL is a refactor with no intended behavior change.
The next step will be to change the chromium build repo to pass True
for ensure_checkout(no_fetch_tags) on chromium trybots.
This represents solution #2 in
https://docs.google.com/document/d/1hDIunJjjfpmr50y3YnZ4o3aq1NZF4gJa1TS0p7AIL64/edit#
Bug: 1019824
Change-Id: I935430603299daa9e301a95a5184c0ce486fd912
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1894352
Reviewed-by: Aaron Gable <agable@chromium.org>
Reviewed-by: Erik Chen <erikchen@chromium.org>
Commit-Queue: danakj <danakj@chromium.org>
Auto-Submit: danakj <danakj@chromium.org>
This reverts commit c87ed60623.
Reason for revert:
Causes a regressions for CL descriptions on Windows
Original change's description:
> git-cl: Fix some python3 compatibility errors.
>
> Also, fix bug in git cl status where the 'updated' field was used to compare messages, even though
> it doesn't exist (see [1]). This CL modifies it to use 'date', which does exist.
>
> [1] https://gerrit-review.googlesource.com/Documentation/rest-api-changes.html#change-message-info
>
> Bug: 1002209
> Change-Id: I5a5e1193b8502c3ad35d94808ea178cad7f44ac6
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1891259
> Commit-Queue: Edward Lesmes <ehmaldonado@chromium.org>
> Reviewed-by: Anthony Polito <apolito@google.com>
TBR=ehmaldonado@chromium.org,apolito@google.com
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: 1002209
Change-Id: I004f202b12c6b99cb6b24cb12a14fba7898569e2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1898547
Reviewed-by: Edward Lesmes <ehmaldonado@chromium.org>
Commit-Queue: Edward Lesmes <ehmaldonado@chromium.org>
Sorting owners by score is non-deterministic, so sort by
name too.
Bug: 1009814
Change-Id: I93a9370a5852634b4bbb2768b9ebbb5b866348e6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1891665
Commit-Queue: Edward Lesmes <ehmaldonado@chromium.org>
Reviewed-by: Anthony Polito <apolito@google.com>
This makes no-rebase-patch-ref correctly populate `file_list` and work with
`git diff --cached` (i.e. they will show only the files affected by the
patchset).
Previously gclient would `reset --soft` to master. Because we didn't rebase
on top of master with `no-rebase-patch-ref`, it has roughly nothing to do
with the patchref we just checked out.
R=gbeaty@chromium.org, ltina@google.com, tandrii@chromium.org
Context:
Tricium recipes sometimes need to use no-rebase-patch-ref in order to get
accurate linenumbers in the diff (i.e. so that the tricium comments will
actually line up correctly with the patchset the user uploaded); Tricium
recipes also need to accurately get the diff file list in this mode.
Change-Id: I5f3c95cd4958cf407a83b96c238b8c55c452ac81
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1895041
Auto-Submit: Robbie Iannucci <iannucci@chromium.org>
Reviewed-by: Andrii Shyshkalov <tandrii@google.com>
Commit-Queue: Andrii Shyshkalov <tandrii@google.com>
The finder.owners_queue contains a set of owners but they appear in an
arbitrary order on each invocation of the test. The current method of
assertEqual() with a list fails randomly because of the ordering being
different.
Example: https://ci.chromium.org/p/infra/builders/try/Depot%20Tools%20Presubmit%20(win)/b8897976148721436512
This changes comparisons of owners_queue to use an assertEqualSets()
which does a comparison that is order-independent.
Change-Id: Ia827795af283dadeb56f76ae0cf14fa4ad7e9eeb
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1895699
Reviewed-by: Aaron Gable <agable@chromium.org>
Reviewed-by: Garrett Beaty <gbeaty@chromium.org>
Commit-Queue: danakj <danakj@chromium.org>
The problems on crbug.com/1017367 should have been fixed by crrev.com/c/1880013
Bug: 1009814
Change-Id: Ia98304ca983b4e7e293fedb9df1fee589de58ba9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1887714
Reviewed-by: Anthony Polito <apolito@google.com>
Commit-Queue: Edward Lesmes <ehmaldonado@chromium.org>
It's not used and uses basestring, which is not supported in Python 3.
Bug: 1002209
Change-Id: I9df900076978636c4a5c75e84432a8cc02d46909
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1891654
Reviewed-by: Anthony Polito <apolito@google.com>
Commit-Queue: Edward Lesmes <ehmaldonado@chromium.org>
basestring doesn't exist in Python 3. scm.Capture status doesn't need to support strings either.
Bug: 1002209
Change-Id: Ie9edfdfd09b789bfaace038a6ad8f3a7da344bc4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/tools/depot_tools/+/1890925
Reviewed-by: Anthony Polito <apolito@google.com>
Commit-Queue: Edward Lesmes <ehmaldonado@chromium.org>
John Budorick
recipe-roller
Edward Lemur
Anthony Polito
Garrett Beaty
Bruce Dawson
Yang Guo
danakj
Stephen Martinis
Jasper Chapman-Black
Vadim Shtayura
Edward Lesmes
Robert Iannucci