Update PRESUBMIT files for first round of ng trybots

Update PRESUBMIT files for first round of ng trybots

Clean up redundant PRESUBMIT files.

linux_redux has been 100% red since at least August 2014,
and is presumably not useful.

BUG=380709
R=sergiyb@chromium.org

Committed: https://chromium.googlesource.com/chromium/src/+/60f8228ca3cd73d071abfcf54f65c1a189fe5bb7

[Sergiy Byelozyorov, 2014-10-30 10:16:39]

sergiyb@chromium.org changed reviewers:
+ sergiyb@chromium.org

[Sergiy Byelozyorov, 2014-10-30 10:16:40]

https://codereview.chromium.org/685363002/diff/1/PRESUBMIT.py
File PRESUBMIT.py (right):

https://codereview.chromium.org/685363002/diff/1/PRESUBMIT.py#newcode1629
PRESUBMIT.py:1629: 'mac_chromium_rel': ['defaulttests'],
Why keep the old ones?

https://codereview.chromium.org/685363002/diff/1/base/PRESUBMIT.py
File base/PRESUBMIT.py (left):

https://codereview.chromium.org/685363002/diff/1/base/PRESUBMIT.py#oldcode51
base/PRESUBMIT.py:51: def GetPreferredTryMasters(project, change):
Why is this (and others) being removed? Shouldn't these be used to new *_ng bots
instead.

https://codereview.chromium.org/685363002/diff/1/ui/events/PRESUBMIT.py
File ui/events/PRESUBMIT.py (right):

https://codereview.chromium.org/685363002/diff/1/ui/events/PRESUBMIT.py#newco...
ui/events/PRESUBMIT.py:20: 'linux_chromium_chromeos_rel': tests,
Why ng bot is not added here?

[Paweł Hajdan Jr., 2014-10-30 10:19:11]

https://codereview.chromium.org/685363002/diff/1/PRESUBMIT.py
File PRESUBMIT.py (right):

https://codereview.chromium.org/685363002/diff/1/PRESUBMIT.py#newcode1629
PRESUBMIT.py:1629: 'mac_chromium_rel': ['defaulttests'],
On 2014/10/30 10:16:40, Sergiy Byelozyorov wrote:
> Why keep the old ones?

This helps avoid a possible presubmit crash with a key error here. The builders
will eventually be removed.

https://codereview.chromium.org/685363002/diff/1/base/PRESUBMIT.py
File base/PRESUBMIT.py (left):

https://codereview.chromium.org/685363002/diff/1/base/PRESUBMIT.py#oldcode51
base/PRESUBMIT.py:51: def GetPreferredTryMasters(project, change):
On 2014/10/30 10:16:40, Sergiy Byelozyorov wrote:
> Why is this (and others) being removed? Shouldn't these be used to new *_ng
bots
> instead.

This is redundant since top-level presubmit file already has these bots.

https://codereview.chromium.org/685363002/diff/1/ui/events/PRESUBMIT.py
File ui/events/PRESUBMIT.py (right):

https://codereview.chromium.org/685363002/diff/1/ui/events/PRESUBMIT.py#newco...
ui/events/PRESUBMIT.py:20: 'linux_chromium_chromeos_rel': tests,
On 2014/10/30 10:16:40, Sergiy Byelozyorov wrote:
> Why ng bot is not added here?

There is no ng bot for this one yet (and several other ones).

[Sergiy Byelozyorov, 2014-10-30 10:38:24]

https://codereview.chromium.org/685363002/diff/1/base/PRESUBMIT.py
File base/PRESUBMIT.py (left):

https://codereview.chromium.org/685363002/diff/1/base/PRESUBMIT.py#oldcode51
base/PRESUBMIT.py:51: def GetPreferredTryMasters(project, change):
On 2014/10/30 10:19:10, Paweł Hajdan Jr. wrote:
> On 2014/10/30 10:16:40, Sergiy Byelozyorov wrote:
> > Why is this (and others) being removed? Shouldn't these be used to new *_ng
> bots
> > instead.
> 
> This is redundant since top-level presubmit file already has these bots.

Is top-level PRESUBMIT executed when there is a more specific one?

https://codereview.chromium.org/685363002/diff/1/ui/events/PRESUBMIT.py
File ui/events/PRESUBMIT.py (right):

https://codereview.chromium.org/685363002/diff/1/ui/events/PRESUBMIT.py#newco...
ui/events/PRESUBMIT.py:20: 'linux_chromium_chromeos_rel': tests,
On 2014/10/30 10:19:10, Paweł Hajdan Jr. wrote:
> On 2014/10/30 10:16:40, Sergiy Byelozyorov wrote:
> > Why ng bot is not added here?
> 
> There is no ng bot for this one yet (and several other ones).

Sorry for the confusion. I meant why is linux_chromium_rel_ng and
win_chromium_rel_ng do not replace linux_chromium_rel and win_chromium_rel
respectively?

[Sergiy Byelozyorov, 2014-10-30 10:39:25]

https://codereview.chromium.org/685363002/diff/1/PRESUBMIT.py
File PRESUBMIT.py (right):

https://codereview.chromium.org/685363002/diff/1/PRESUBMIT.py#newcode1629
PRESUBMIT.py:1629: 'mac_chromium_rel': ['defaulttests'],
On 2014/10/30 10:19:10, Paweł Hajdan Jr. wrote:
> On 2014/10/30 10:16:40, Sergiy Byelozyorov wrote:
> > Why keep the old ones?
> 
> This helps avoid a possible presubmit crash with a key error here. The
builders
> will eventually be removed.

Can you elaborate please? Why would presubmit use old key?

[Paweł Hajdan Jr., 2014-10-30 10:40:50]

https://codereview.chromium.org/685363002/diff/1/base/PRESUBMIT.py
File base/PRESUBMIT.py (left):

https://codereview.chromium.org/685363002/diff/1/base/PRESUBMIT.py#oldcode51
base/PRESUBMIT.py:51: def GetPreferredTryMasters(project, change):
On 2014/10/30 10:38:24, Sergiy Byelozyorov wrote:
> Is top-level PRESUBMIT executed when there is a more specific one?

All of them get executed, please see
http://dev.chromium.org/developers/how-tos/depottools/presubmit-scripts

In other words, the top-level one is always executed.

https://codereview.chromium.org/685363002/diff/1/ui/events/PRESUBMIT.py
File ui/events/PRESUBMIT.py (right):

https://codereview.chromium.org/685363002/diff/1/ui/events/PRESUBMIT.py#newco...
ui/events/PRESUBMIT.py:20: 'linux_chromium_chromeos_rel': tests,
On 2014/10/30 10:38:24, Sergiy Byelozyorov wrote:
> On 2014/10/30 10:19:10, Paweł Hajdan Jr. wrote:
> > On 2014/10/30 10:16:40, Sergiy Byelozyorov wrote:
> > > Why ng bot is not added here?
> > 
> > There is no ng bot for this one yet (and several other ones).
> 
> Sorry for the confusion. I meant why is linux_chromium_rel_ng and
> win_chromium_rel_ng do not replace linux_chromium_rel and win_chromium_rel
> respectively?

Same reason as other removals in this CL: ng bots are in the top-level
PRESUBMIT, I don't see a need to duplicate that here.

[Paweł Hajdan Jr., 2014-10-30 10:48:59]

https://codereview.chromium.org/685363002/diff/1/PRESUBMIT.py
File PRESUBMIT.py (right):

https://codereview.chromium.org/685363002/diff/1/PRESUBMIT.py#newcode1629
PRESUBMIT.py:1629: 'mac_chromium_rel': ['defaulttests'],
On 2014/10/30 10:39:24, Sergiy Byelozyorov wrote:
> Can you elaborate please? Why would presubmit use old key?

Please see "dereference" on line 1645 and trace back. Crashes like this have
happened in the past. I don't see a downside of keeping *still existing* builder
names in this dict. Feel free to elaborate more about possible problem you see
here.

[Sergiy Byelozyorov, 2014-10-30 10:52:57]

lgtm

https://codereview.chromium.org/685363002/diff/1/PRESUBMIT.py
File PRESUBMIT.py (right):

https://codereview.chromium.org/685363002/diff/1/PRESUBMIT.py#newcode1629
PRESUBMIT.py:1629: 'mac_chromium_rel': ['defaulttests'],
On 2014/10/30 10:48:59, Paweł Hajdan Jr. wrote:
> On 2014/10/30 10:39:24, Sergiy Byelozyorov wrote:
> > Can you elaborate please? Why would presubmit use old key?
> 
> Please see "dereference" on line 1645 and trace back. Crashes like this have
> happened in the past. I don't see a downside of keeping *still existing*
builder
> names in this dict. Feel free to elaborate more about possible problem you see
> here.

The downside is keeping dead code. But I see now how can this result in a crash.

https://codereview.chromium.org/685363002/diff/1/base/PRESUBMIT.py
File base/PRESUBMIT.py (left):

https://codereview.chromium.org/685363002/diff/1/base/PRESUBMIT.py#oldcode51
base/PRESUBMIT.py:51: def GetPreferredTryMasters(project, change):
On 2014/10/30 10:40:49, Paweł Hajdan Jr. wrote:
> On 2014/10/30 10:38:24, Sergiy Byelozyorov wrote:
> > Is top-level PRESUBMIT executed when there is a more specific one?
> 
> All of them get executed, please see
> http://dev.chromium.org/developers/how-tos/depottools/presubmit-scripts
> 
> In other words, the top-level one is always executed.

Acknowledged.

https://codereview.chromium.org/685363002/diff/1/ui/events/PRESUBMIT.py
File ui/events/PRESUBMIT.py (right):

https://codereview.chromium.org/685363002/diff/1/ui/events/PRESUBMIT.py#newco...
ui/events/PRESUBMIT.py:20: 'linux_chromium_chromeos_rel': tests,
On 2014/10/30 10:40:49, Paweł Hajdan Jr. wrote:
> On 2014/10/30 10:38:24, Sergiy Byelozyorov wrote:
> > On 2014/10/30 10:19:10, Paweł Hajdan Jr. wrote:
> > > On 2014/10/30 10:16:40, Sergiy Byelozyorov wrote:
> > > > Why ng bot is not added here?
> > > 
> > > There is no ng bot for this one yet (and several other ones).
> > 
> > Sorry for the confusion. I meant why is linux_chromium_rel_ng and
> > win_chromium_rel_ng do not replace linux_chromium_rel and win_chromium_rel
> > respectively?
> 
> Same reason as other removals in this CL: ng bots are in the top-level
> PRESUBMIT, I don't see a need to duplicate that here.

Acknowledged.

[commit-bot: I haz the power, 2014-10-30 10:56:42]

Patchset 1 (id:??) landed as
https://crrev.com/60f8228ca3cd73d071abfcf54f65c1a189fe5bb7
Cr-Commit-Position: refs/heads/master@{#302067}

[Paweł Hajdan Jr., 2014-10-30 10:56:47]

Committed patchset #1 (id:1) manually as
60f8228ca3cd73d071abfcf54f65c1a189fe5bb7.