C++ porting of the initial python script for conversational speech generation.

C++ porting of the initial python script for conversational speech generation.

This CL removes the Python script and adds its C++ porting.
The former was in its early stage and it has permanently been removed.

BUG=webrtc:7218
NOTRY=True

Review-Url: https://codereview.webrtc.org/2740063004
Cr-Commit-Position: refs/heads/master@{#17254}
Committed: https://chromium.googlesource.com/external/webrtc/+/0cf3aa6d0d170365a2dedb57960fbface736f851

[AleBzk, 2017-03-09 09:37:41]

Description was changed from

==========
C++ porting of the initial python script for conversational speech generation.

This CL removes the Python script and adds its C++ porting.
The former was in its early stage and it has permanently been removed.

BUG=webrtc:7218
==========

to

==========
C++ porting of the initial python script for conversational speech generation.

This CL removes the Python script and adds its C++ porting.
The former was in its early stage and it has permanently been removed.

BUG=webrtc:7218
==========

[AleBzk, 2017-03-09 09:37:41]

alessiob@webrtc.org changed reviewers:
+ henrik.lundin@webrtc.org

[AleBzk, 2017-03-09 09:38:53]

Hi Henrik,
As agreed, I switched to C++ for the conversation speech tool.

Alessio

[hlundin-webrtc, 2017-03-13 12:28:20]

Good start! See some comments inline.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
File
webrtc/modules/audio_processing/test/conversational_speech/convspeech_gen.cc
(right):

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/convspeech_gen.cc:41:
if (!settings.ValidateAndReport()) {
I've been increasingly radical about error handling in test code. I suggest you
simply
RTC_CHECK(settings.ValidateAndReport());

But...

I suggest you check the file names here instead. Use the flag validator
functionality in gflags: https://gflags.github.io/gflags/#validate.

static bool ValidateFile(const char* flagname, const std::string& value) {
  // Test the file name with FileExists from
webrtc/test/testsupport/fileutils.h.
}
DEFINE_string(t, "", "Path to the timing text file");
DEFINE_validator(t, &ValidateFile);

Similar for directories.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
File webrtc/modules/audio_processing/test/conversational_speech/settings.cc
(right):

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.cc:30: bool
bln_valid_audiotracks_path = ValidateDirectoryPath(&audiotracks_path_);
We typically don't add bln_ to qualify the variable type.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
File webrtc/modules/audio_processing/test/conversational_speech/settings.h
(right):

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:19:
namespace convspeechgen {
I think you can drop this namespace, and maybe give the Settings class a
slightly more specific name.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:21: class
Settings {
If all of the error handling can be done by gflags instead, then this class
reduces to a simple data container, which warrants using a struct instead of a
class. Then you can get rid of the getters too, since a struct as allowed (by
the style guide) to have public member variables. But you may have more
elaborate plans for this going forward, in which case it is fine to keep it as a
class.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:23:
Settings(std::string audiotracks_path, std::string timing_filepath,
Make the input parameters const references instead (const std::string&).

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:28: //
Getters.
Superfluous comment.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:29: const
std::string AudiotracksPath(void);
Don't add the void parameter.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:29: const
std::string AudiotracksPath(void);
const std::string makes little sense. Either drop the const, or make it return a
reference or pointer, which should then be const.

The method itself should be const-declared, though, just before the closing ;

For accessors (getters) and mutators, the style guide allows (but does not
require) that the name of the method matches the underlying variable.
https://google.github.io/styleguide/cppguide.html#Function_Names

Summing up, I'd like to see
std::string audiotracks_path() const;  // Note that I omitted the trailing
underscore.
-- or --
const std::string& audiotracks_path() const;

Same for all getters.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:34: bool
ValidateAndReport(void);
Drop void.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:34: bool
ValidateAndReport(void);
const method?

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:37:
std::string audiotracks_path_;
Declare methods before member variables.
https://google.github.io/styleguide/cppguide.html#Declaration_Order

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:37:
std::string audiotracks_path_;
You can make these variables const.
const std::string audiotracks_path_;
const ...

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:42: static
bool ValidateFilePath(std::string* path);
You can use FileExists in webrtc/test/testsupport/fileutils.h.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:45: static
bool ValidateDirectoryPath(std::string* path);
Consider extending webrtc/test/testsupport/fileutils.h with a DirExists method
instead.

[AleBzk, 2017-03-13 14:14:42]

Hi Henrik,

Thanks a lot for your comments.
Everything has been addressed.

Alessio

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
File
webrtc/modules/audio_processing/test/conversational_speech/convspeech_gen.cc
(right):

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/convspeech_gen.cc:41:
if (!settings.ValidateAndReport()) {
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> I've been increasingly radical about error handling in test code. I suggest
you
> simply
> RTC_CHECK(settings.ValidateAndReport());
> 
> But...
> 
> I suggest you check the file names here instead. Use the flag validator
> functionality in gflags: https://gflags.github.io/gflags/#validate.
> 
> static bool ValidateFile(const char* flagname, const std::string& value) {
>   // Test the file name with FileExists from
> webrtc/test/testsupport/fileutils.h.
> }
> DEFINE_string(t, "", "Path to the timing text file");
> DEFINE_validator(t, &ValidateFile);
> 
> Similar for directories.

Done.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
File webrtc/modules/audio_processing/test/conversational_speech/settings.cc
(right):

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.cc:30: bool
bln_valid_audiotracks_path = ValidateDirectoryPath(&audiotracks_path_);
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> We typically don't add bln_ to qualify the variable type.

Acknowledged.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
File webrtc/modules/audio_processing/test/conversational_speech/settings.h
(right):

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:19:
namespace convspeechgen {
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> I think you can drop this namespace, and maybe give the Settings class a
> slightly more specific name.

Done.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:21: class
Settings {
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> If all of the error handling can be done by gflags instead, then this class
> reduces to a simple data container, which warrants using a struct instead of a
> class. Then you can get rid of the getters too, since a struct as allowed (by
> the style guide) to have public member variables. But you may have more
> elaborate plans for this going forward, in which case it is fine to keep it as
a
> class.

Done.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:23:
Settings(std::string audiotracks_path, std::string timing_filepath,
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> Make the input parameters const references instead (const std::string&).

Done.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:28: //
Getters.
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> Superfluous comment.

Done.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:29: const
std::string AudiotracksPath(void);
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> Don't add the void parameter.

Acknowledged.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:29: const
std::string AudiotracksPath(void);
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> const std::string makes little sense. Either drop the const, or make it return
a
> reference or pointer, which should then be const.
> 
> The method itself should be const-declared, though, just before the closing ;
> 
> For accessors (getters) and mutators, the style guide allows (but does not
> require) that the name of the method matches the underlying variable.
> https://google.github.io/styleguide/cppguide.html#Function_Names
> 
> Summing up, I'd like to see
> std::string audiotracks_path() const;  // Note that I omitted the trailing
> underscore.
> -- or --
> const std::string& audiotracks_path() const;
> 
> Same for all getters.

Acknowledged.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:34: bool
ValidateAndReport(void);
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> Drop void.

Acknowledged.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:37:
std::string audiotracks_path_;
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> You can make these variables const.
> const std::string audiotracks_path_;
> const ...

Done.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:42: static
bool ValidateFilePath(std::string* path);
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> You can use FileExists in webrtc/test/testsupport/fileutils.h.

Done.

https://codereview.webrtc.org/2740063004/diff/1/webrtc/modules/audio_processi...
webrtc/modules/audio_processing/test/conversational_speech/settings.h:45: static
bool ValidateDirectoryPath(std::string* path);
On 2017/03/13 12:28:19, hlundin-webrtc wrote:
> Consider extending webrtc/test/testsupport/fileutils.h with a DirExists method
> instead.

Done.

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/modules/audio_proc...
File
webrtc/modules/audio_processing/test/conversational_speech/convspeech_gen.cc
(right):

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/modules/audio_proc...
webrtc/modules/audio_processing/test/conversational_speech/convspeech_gen.cc:28:
};
See below DEFINE_validator. DirExists is used twice and the macro doesn't
directly accept DirExists. This is why I'm using lambdas.

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/test/testsupport/f...
File webrtc/test/testsupport/fileutils_unittest.cc (right):

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/test/testsupport/f...
webrtc/test/testsupport/fileutils_unittest.cc:91: << "Existing file recognized
as existing directory";
I added this test here, but we could add a new test.

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/test/testsupport/f...
webrtc/test/testsupport/fileutils_unittest.cc:112: << "Removed directory found";
I added these tests here, but we could add a new test.

[hlundin-webrtc, 2017-03-13 23:12:27]

Very nice now!

LGTM for webrtc/modules/audio_processing/

You will need an owner for the testsupport changes.

[AleBzk, 2017-03-14 07:54:55]

alessiob@webrtc.org changed reviewers:
+ kjellander@webrtc.org

[AleBzk, 2017-03-14 07:57:56]

I added H. Kjellander since he is an owner of //webrtc/test. I need his approval
for the DirExists function and the unit tests that I changed.

[kjellander_webrtc, 2017-03-14 13:29:06]

lgtm but please run the linux_internal trybot, since I suspect we might run into
trouble due to the fileutils.h addition.

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/modules/audio_proc...
File webrtc/modules/audio_processing/test/conversational_speech/BUILD.gn
(right):

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/modules/audio_proc...
webrtc/modules/audio_processing/test/conversational_speech/BUILD.gn:16: }  #
conversational_speech
I don't think the  # conversational_speech is needed here.

It was mostly a thing we had for GYP since GYP files were so hard to read. GN is
much better in this aspect.

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/modules/audio_proc...
File
webrtc/modules/audio_processing/test/conversational_speech/convspeech_gen.cc
(right):

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/modules/audio_proc...
webrtc/modules/audio_processing/test/conversational_speech/convspeech_gen.cc:54:
// TODO(alessiob): remove line below once debugged.
Should this still be here?

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/test/testsupport/f...
File webrtc/test/testsupport/fileutils_unittest.cc (right):

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/test/testsupport/f...
webrtc/test/testsupport/fileutils_unittest.cc:91: << "Existing file recognized
as existing directory";
On 2017/03/13 14:14:42, AleBzk wrote:
> I added this test here, but we could add a new test.

Yes, please test only one thing in each test case, i.e. add new tests
specifically for DirExists.

[AleBzk, 2017-03-14 16:19:21]

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/modules/audio_proc...
File webrtc/modules/audio_processing/test/conversational_speech/BUILD.gn
(right):

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/modules/audio_proc...
webrtc/modules/audio_processing/test/conversational_speech/BUILD.gn:16: }  #
conversational_speech
On 2017/03/14 13:29:06, kjellander_webrtc wrote:
> I don't think the  # conversational_speech is needed here.
> 
> It was mostly a thing we had for GYP since GYP files were so hard to read. GN
is
> much better in this aspect.

Done.

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/modules/audio_proc...
File
webrtc/modules/audio_processing/test/conversational_speech/convspeech_gen.cc
(right):

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/modules/audio_proc...
webrtc/modules/audio_processing/test/conversational_speech/convspeech_gen.cc:54:
// TODO(alessiob): remove line below once debugged.
On 2017/03/14 13:29:06, kjellander_webrtc wrote:
> Should this still be here?

Yes, it will be removed in future CLs.

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/test/testsupport/f...
File webrtc/test/testsupport/fileutils_unittest.cc (right):

https://codereview.webrtc.org/2740063004/diff/20001/webrtc/test/testsupport/f...
webrtc/test/testsupport/fileutils_unittest.cc:91: << "Existing file recognized
as existing directory";
On 2017/03/14 13:29:06, kjellander_webrtc wrote:
> On 2017/03/13 14:14:42, AleBzk wrote:
> > I added this test here, but we could add a new test.
> 
> Yes, please test only one thing in each test case, i.e. add new tests
> specifically for DirExists.

Done.

[AleBzk, 2017-03-14 16:22:11]

The CQ bit was checked by alessiob@webrtc.org to run a CQ dry run

[commit-bot: I haz the power, 2017-03-14 16:22:19]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-03-14 16:24:50]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-14 16:24:51]

Dry run: Try jobs failed on following builders:
  ios32_sim_ios9_dbg on master.tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/ios32_sim_ios9_dbg/buil...)
  ios64_sim_ios10_dbg on master.tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/ios64_sim_ios10_dbg/bui...)
  ios_arm64_dbg on master.tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/ios_arm64_dbg/builds/18558)
  ios_dbg on master.tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/ios_dbg/builds/23840)

[AleBzk, 2017-03-14 17:30:50]

The CQ bit was checked by alessiob@webrtc.org to run a CQ dry run

[commit-bot: I haz the power, 2017-03-14 17:31:03]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-03-14 17:33:40]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-14 17:33:41]

Dry run: Try jobs failed on following builders:
  ios32_sim_ios9_dbg on master.tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/ios32_sim_ios9_dbg/buil...)
  ios_dbg on master.tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/ios_dbg/builds/23843)

[AleBzk, 2017-03-14 17:50:58]

Patchset #4 (id:60001) has been deleted

[AleBzk, 2017-03-15 14:38:49]

Patchset #3 (id:40001) has been deleted

[AleBzk, 2017-03-15 14:39:33]

Patchset #2 (id:20001) has been deleted

[AleBzk, 2017-03-15 14:42:03]

Patchset #2 (id:80001) has been deleted

[AleBzk, 2017-03-15 14:50:06]

The CQ bit was checked by alessiob@webrtc.org to run a CQ dry run

[commit-bot: I haz the power, 2017-03-15 14:50:13]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[AleBzk, 2017-03-15 14:52:32]

For simplicity, I split this CL adding a new one.

DirExists change is now in https://codereview.webrtc.org/2749163003/ (comments
from kjellander only affect the new CL).

Upstream of this CL changed to 2749163003.

[commit-bot: I haz the power, 2017-03-15 14:52:59]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-15 14:53:00]

Dry run: Try jobs failed on following builders:
  ios32_sim_ios9_dbg on master.tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/ios32_sim_ios9_dbg/buil...)

[AleBzk, 2017-03-15 14:53:04]

The CQ bit was unchecked by alessiob@webrtc.org

[AleBzk, 2017-03-15 14:53:30]

Description was changed from

==========
C++ porting of the initial python script for conversational speech generation.

This CL removes the Python script and adds its C++ porting.
The former was in its early stage and it has permanently been removed.

BUG=webrtc:7218
==========

to

==========
C++ porting of the initial python script for conversational speech generation.

This CL removes the Python script and adds its C++ porting.
The former was in its early stage and it has permanently been removed.

BUG=webrtc:7218
NOTRY=True
==========

[AleBzk, 2017-03-15 14:53:59]

The CQ bit was checked by alessiob@webrtc.org

[AleBzk, 2017-03-15 14:54:00]

The patchset sent to the CQ was uploaded after l-g-t-m from
henrik.lundin@webrtc.org, kjellander@webrtc.org
Link to the patchset: https://codereview.webrtc.org/2740063004/#ps100001 (title:
"rebase")

[commit-bot: I haz the power, 2017-03-15 14:54:11]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-03-15 14:56:27]

CQ is committing da patch.

Bot data: {"patchset_id": 100001, "attempt_start_ts": 1489589639974190,
"parent_rev": "2549ad4fef419a055671c5be39fcfa27ba573c3b", "commit_rev":
"0cf3aa6d0d170365a2dedb57960fbface736f851"}

[commit-bot: I haz the power, 2017-03-15 14:56:29]

Description was changed from

==========
C++ porting of the initial python script for conversational speech generation.

This CL removes the Python script and adds its C++ porting.
The former was in its early stage and it has permanently been removed.

BUG=webrtc:7218
NOTRY=True
==========

to

==========
C++ porting of the initial python script for conversational speech generation.

This CL removes the Python script and adds its C++ porting.
The former was in its early stage and it has permanently been removed.

BUG=webrtc:7218
NOTRY=True

Review-Url: https://codereview.webrtc.org/2740063004
Cr-Commit-Position: refs/heads/master@{#17254}
Committed:
https://chromium.googlesource.com/external/webrtc/+/0cf3aa6d0d170365a2dedb579...
==========

[commit-bot: I haz the power, 2017-03-15 14:56:31]

Committed patchset #2 (id:100001) as
https://chromium.googlesource.com/external/webrtc/+/0cf3aa6d0d170365a2dedb579...

[AleBzk, 2017-03-15 15:19:16]

A revert of this CL (patchset #2 id:100001) has been created in
https://codereview.webrtc.org/2753843002/ by alessiob@webrtc.org.

The reason for reverting is: Even if the conversational speech tool is external
and not a core part of webrtc, there are too many trybots failing..

[kjellander_webrtc, 2017-03-16 15:49:58]

On 2017/03/15 15:19:16, AleBzk wrote:
> A revert of this CL (patchset #2 id:100001) has been created in
> https://codereview.webrtc.org/2753843002/ by mailto:alessiob@webrtc.org.
> 
> The reason for reverting is: Even if the conversational speech tool is
external
> and not a core part of webrtc, there are too many trybots failing..

Please never use NOTRY=True unless you've manually run all relevant trybots or
you're absolutely sure your test doesn't need to run trybots. Our trybot
resources are not as scarce as they use to be, so feel free to use the CQ for
most CLs (but if you change like one line in an OWNERS file NOTRY=True is
certainly a saving). Better spend a few extra resources than breaking the tree
though.