Unit test base network

[finozzifa]

Closes # (if applicable).

Changes proposed in this Pull Request

The pull request proposes the following unit tests:

test/test_base_network.py:

• test_get_country
• test_get_linetypes_config
• test_get_linetype_by_voltage
• test_get_oid
• test_load_buses
• test_load_converters_from_eg
• test_load_converters_from_osm
• test_load_lines
• test_load_links_from_eg
• test_load_links_from_osm
• test_load_transformers
• test_reconnect_crimea
• test_set_electrical_parameters_converters
• test_set_electrical_parameters_lines_eg
• test_set_electrical_parameters_lines_osm
• test_set_electrical_parameters_links_osm
• test_set_electrical_parameters_transformers

Checklist

• I tested my contribution locally and it works as intended.
• Code and workflow changes are sufficiently documented.
• Changed dependencies are added to envs/environment.yaml.
• Changes in configuration options are added in config/config.default.yaml.
• Changes in configuration options are documented in doc/configtables/*.csv.
• Sources of newly added data are documented in doc/data_sources.rst.
• A release note doc/release_notes.rst is added.

[lkstrp]

Thanks for continuing the unit testing journey @finozzifa!
I was mostly just skimming through things, but looks good to me.
If you haven't already done so locally, we could think about integrating codecov like in the other package repos. See here. Not as a check (yet), but just to get an overview to focus on the most useful tests.

[lkstrp]

Do you have any best practices for choosing scopes? I think the function scope works well for DataFrames because they initialise quickly. But in PyPSA I've been playing around with the scope of test networks because they are sometimes modified and sometimes not in the tests, which adds up if function is used as scope.

[lkstrp]

And same goes for the two network fixtures you created here already

[finozzifa]

Hey @lkstrp,

indeed great question!

I usually used function being the default one and (most likely) the safest one.

However, now that I think about it, the config fixture can be initialized only once per test session. Changing the scope for this fixture actually leads to a speed up of the test execution (almost 30%). You can find the results below. The scope session and package behave equally well.

This is of course not very relevant now, but it will become once we add more and more tests.

@pytest.fixture(scope="function")
def config():

hyperfine pytest test

Time (mean ± σ):      2.223 s ±  0.108 s    [User: 1.949 s, System: 0.251 s]
Range (min … max):    2.106 s …  2.459 s    10 runs

@pytest.fixture(scope="session")
def config():

hyperfine pytest test

Time (mean ± σ):      1.568 s ±  0.004 s    [User: 1.333 s, System: 0.228 s]
Range (min … max):    1.563 s …  1.573 s    10 runs

@pytest.fixture(scope="package")
def config():

hyperfine pytest test

Time (mean ± σ):      1.570 s ±  0.003 s    [User: 1.334 s, System: 0.229 s]
Range (min … max):    1.567 s …  1.574 s    10 runs