(closes #3250 #279) Complete code coverage #3258

sergisiso · 2025-12-11T13:07:48Z

No description provided.

codecov · 2025-12-11T13:18:28Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.96%. Comparing base (97dd02d) to head (66129b0).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3258      +/-   ##
==========================================
+ Coverage   99.91%   99.96%   +0.04%     
==========================================
  Files         376      376              
  Lines       53529    53482      -47     
==========================================
- Hits        53484    53461      -23     
+ Misses         45       21      -24

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

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

sergisiso · 2025-12-17T11:23:30Z

@arporter This is ready for review. It brings the codecov misses from 45 to 27 and adds no-cover pragmas to the remaining (maybe you know how to do better in the remaining?). I also added a more strict reporting for deleted lines (indirect changes).

arporter

Thanks very much for tackling this Sergi.
There are a couple of cases where I think we can add tests rather than no-cover pragmas. Also, you've removed some validation code from ACCEnterDataTrans and I don't understand why.

arporter · 2025-12-17T16:44:31Z

src/psyclone/domain/lfric/lfric_kern.py

            # mixed-precision kernel.
            alg_precision = api_config.precision_map[alg_arg.precision]
-            if alg_precision != actual_precision:
+            if alg_precision != actual_precision:  # pragma: no-cover


I think adding the following test to lfric_kern_test.py will cover this line. I'll let you write a suitable docstring :-)

def test_validate_kernel_code_arg_against_alg_arg(monkeypatch): ''' ''' psy, invoke = get_invoke("26.8_mixed_precision_args.f90", TEST_API, name="invoke_0", dist_mem=False) sched = invoke.schedule # Second kernel has args of precision r_solver. kernel = sched.walk(LFRicKern, stop_type=LFRicKern)[1] read_access = ArgumentInterface(ArgumentInterface.Access.READ) lfric_real_scalar_symbol = LFRicTypes("LFRicRealScalarDataSymbol")( "scalar", interface=read_access) with pytest.raises(GenerationError) as err: kernel._validate_kernel_code_arg(lfric_real_scalar_symbol, lfric_real_scalar_symbol, kernel.args[0]) assert ("Precision (4 bytes) of algorithm-layer argument " "'scalar_r_solver' does not match that (8 bytes) of the " "corresponding kernel subroutine argument 'scalar' for kernel " "'mixed_code'" in str(err.value))

arporter · 2025-12-17T16:53:58Z

src/psyclone/gocean1p0.py

        # if the kernel requires a grid-property argument.
        return None

-    @property


I see you've also update lfric.py (thanks) so I've added #279 to the issues this PR will close (as I've grepped and those are the only mentions).

arporter · 2025-12-17T16:57:52Z

src/psyclone/psyGen.py

        return [self._field]

-    def check_vector_halos_differ(self, node):
+    def check_vector_halos_differ(self, node):  # pragma: no-cover


psyGen_test does have some tests of this method so it can't all be uncovered? In fact, I've grepped and it seems this method is completely unused?

I delete it and there was no issues

arporter · 2025-12-17T17:10:38Z

src/psyclone/psyir/nodes/node.py

+# no cover: start
 try:
    # pylint disable=import-outside-toplevel
    from termcolor import colored


Can't we monkeypatch the packages that appear to be available from termcolor? I had a google and found https://stackoverflow.com/questions/8658043/how-to-mock-an-import

(I think it's quite important to get coverage for this because if we somehow broke it it would break a lot of stuff!)

I have reverted this for now, as I see we have the same problem with graphviz and I have not yet figure out how to solve it

arporter · 2025-12-17T17:13:26Z

src/psyclone/tests/gocean1p0_test.py



+def test_invalid_field_accesses_for_iteration_space():
+    ''' Tests that we when retrieving the iteration_space_arg we validate


s/we when/when/

arporter · 2025-12-17T17:16:01Z

src/psyclone/tests/psyir/backend/fortran_unsupported_declns_test.py

        ''' UnsupportedType is abstract so sub-class it for this test '''
        def __str__(self):
-            return "OtherType"
+            ...


I didn't know you could do that!

arporter · 2025-12-17T17:16:50Z

src/psyclone/tests/psyir/frontend/fparser2_program_handler_test.py



+def test_empty_program_handler(parser):
+    '''Test that an empty program handler still returns a FileContainer


"the program handler still returns a FileContainer for an empty program" perhaps?

Yes, that's better

arporter · 2025-12-17T17:22:17Z

src/psyclone/transformations.py

        # extract async. Default to False.
        async_queue = options.get('async_queue', False)

-        # check


How come you've removed this? I see we still support the specification of an async_queue?

I don't understand why we checked that all directives in the same schedule had the same async_queue? Wouldn't that mean that we cannot do ACCEnterData asynchronously at all?

arporter · 2025-12-17T17:23:23Z

src/psyclone/transformations.py

        try:
            kernels = node.get_callees()
-        except SymbolError as err:
+        except SymbolError as err:  # pragma: no-cover


This should be easy to cover as we can monkeypatch node.get_callees() with a function that just raises SymbolError.

sergisiso · 2025-12-19T08:58:03Z

@arporter This is ready for another look.

sergisiso added 4 commits December 11, 2025 12:27

#3250 Fix some test coverage issues

44a59e5

Merge remote-tracking branch 'origin/master' into 3250_complete_coverage

e7ee8c3

#3250 Fix more test coverage

404db07

#3250 Fix more test coverage

a7a40d0

sergisiso added 6 commits December 11, 2025 14:37

#3250 Removes pragmas no-cover by adding appropriate tests

f61544a

#3250 Removes pragmas no-cover by adding appropriate tests

fdc0d18

#3250 Add no-cover start/stop comments

6a1aa7c

#3250 Add codecov.yml

ba11ebf

Merge remote-tracking branch 'origin/master' into 3250_complete_coverage

5a7f27f

#3250 Remove no-cover pragmas

1620554

sergisiso requested a review from arporter December 17, 2025 11:19

sergisiso self-assigned this Dec 17, 2025

sergisiso added the ready for review label Dec 17, 2025

arporter added under review and removed ready for review labels Dec 17, 2025

arporter changed the title ~~(closes #3250) Complete code coverage~~ (closes #3250 #279) Complete code coverage Dec 17, 2025

arporter requested changes Dec 17, 2025

View reviewed changes

arporter added reviewed with actions and removed under review labels Dec 17, 2025

sergisiso added 5 commits December 18, 2025 13:18

#3250 Remove old LFRic code

9f94bb6

#3250 Fix flake8 issue

60b01cd

#3250 Revert unintended change

9636734

#3250 Improve test coverage

b7771c1

#3250 Remove some no cover comments

66129b0

sergisiso temporarily deployed to integration December 18, 2025 14:51 — with GitHub Actions Inactive

sergisiso added ready for review and removed reviewed with actions labels Dec 19, 2025



		def test_invalid_field_accesses_for_iteration_space():
		''' Tests that we when retrieving the iteration_space_arg we validate



		def test_empty_program_handler(parser):
		'''Test that an empty program handler still returns a FileContainer

(closes #3250 #279) Complete code coverage #3258

Are you sure you want to change the base?

(closes #3250 #279) Complete code coverage #3258

Uh oh!

Conversation

sergisiso commented Dec 11, 2025

Uh oh!

codecov bot commented Dec 11, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

sergisiso commented Dec 17, 2025

Uh oh!

arporter left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

sergisiso commented Dec 19, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

codecov bot commented Dec 11, 2025 •

edited

Loading