#1312: metadata for read-only arrays

[mo-lottieturner]

.

[codecov[bot]]

Codecov Report

Attention: 4 lines in your changes are missing coverage. Please review.

Comparison is base (b393e04) 0.00% compared to head (fe4baf9) 99.84%.

Files	Patch %	Lines
...psyclone/domain/lfric/kernel/array_arg_metadata.py	94.00%	3 Missing :warning:
...ne/domain/lfric/kernel/common_meta_arg_metadata.py	92.30%	1 Missing :warning:

Additional details and impacted files

@@             Coverage Diff             @@
##           master    #2173       +/-   ##
===========================================
+ Coverage        0   99.84%   +99.84%     
===========================================
  Files           0      353      +353     
  Lines           0    47511    +47511     
===========================================
+ Hits            0    47438    +47438     
- Misses          0       73       +73     

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[mo-lottieturner]

(all tests pass except codecov)

[mo-lottieturner]

#1312

[mo-lottieturner]

tests all green except codecov

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 99.90%. Comparing base (b1f35f0) to head (3c37804).

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #2173   +/-   ##
=======================================
  Coverage   99.90%   99.90%           
=======================================
  Files         366      367    +1     
  Lines       51672    51755   +83     
=======================================
+ Hits        51622    51705   +83     
  Misses         50       50           

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

:rocket: New features to boost your workflow:

• :snowflake: Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[mo-lottieturner]

Just documentation left to do

[mo-lottieturner]

did a double check of pycodestyle and pylint, after noticing the above modifier line had slipped through - pycodestyle clear in src/psyclone, but pylint has quite a few 'Access to protected member' in the test files - should I ignore these?

tests/domain/lfric/kernel/array_arg_metadata_test.py:78:36: W0212: Access to a protected member _get_metadata of a client class (protected-access)

tests/domain/lfric/kernel/common_meta_arg_metadata_test.py:126:11: W0212: Access to a protected member _datatype of a client class (protected-access)
tests/domain/lfric/kernel/common_meta_arg_metadata_test.py:127:11: W0212: Access to a protected member _access of a client class (protected-access)
tests/domain/lfric/kernel/common_meta_arg_metadata_test.py:158:11: W0212: Access to a protected member _datatype of a client class (protected-access)
tests/domain/lfric/kernel/common_meta_arg_metadata_test.py:159:11: W0212: Access to a protected member _access of a client class (protected-access)
tests/domain/lfric/kernel/common_meta_arg_metadata_test.py:299:11: W0212: Access to a protected member _datatype of a client class (protected-access)
tests/domain/lfric/kernel/common_meta_arg_metadata_test.py:303:11: W0212: Access to a protected member _access of a client class (protected-access)

tests/domain/lfric/lfric_array_mdata_test.py:87:16: W0212: Access to a protected member _inits of a client class (protected-access)
tests/domain/lfric/lfric_array_mdata_test.py:89:8: W0212: Access to a protected member _init_array of a client class (protected-access)
tests/domain/lfric/lfric_array_mdata_test.py:160:17: W0212: Access to a protected member _inits of a client class (protected-access)
tests/domain/lfric/lfric_array_mdata_test.py:169:8: W0212: Access to a protected member _init_scalar of a client class (protected-access)

[arporter]

but pylint has quite a few 'Access to protected member' in the test files - should I ignore these?

For the tests, pylint should be run in the tests directory. That way it picks up some test-specific settings (one of which is to turn off such warnings).

[mo-lottieturner]

For posterity: If memory serves, we agreed to stick with GH_ARRAY, but that NRANKS*n can be switched to just n

[mo-lottieturner]

I think I've removed all of the NRANKS mentions, back to @arporter for review

[mo-lottieturner]

@arporter not all review comments have been addressed yet, but as I'm handing this over to Alistair, it would be helpful for you to mark the addressed comments as resolved, if you have a moment?

[arporter]

OK @mo-lottieturner and @mo-alistairp, I've gone through and resolved everything that has been done.

[arporter]

I can't find your comment @mo-alistairp but yes, "s/size/rank/" is my shorthand (sed notation) for replace "size" with "rank".

[mo-alistairp]

I've addrsesed all the open comments. It's possible I've been a little overzealous with the changes from array to ScalarArray, so just let me know. Otherwise, this is ready for review

[mo-alistairp]

I've added some documentation for the subroutine call. The design in #1312 seems to suggest that the ScalarArray is only variable needed in the subroutine definition, so hopefully this is sufficient. Let me know if you want anything else included in the documentation. Otherwise, I think this is ready for review again.