350 utils

[sophie22]

Summary

in line with Python best practice, remove all commands possible from the init.py into a new main.py script, make required changes to entry point and subsequently update the READme with the relevant commands
EDIT: upon further reading of documentation, looks like there is no real benefit to having the main function in a separate script from init.

combine utility functions from init, tools and shapes.py into a new utils.py, make required changes to all task scripts that use utility functions and their applicable test scripts

resolves #347

Outstanding TODO:

• update original test_tools and test_hazenlib scripts and correctly group unit tests in line with new utils.py and main.py

[github-actions]

Coverage Report

File	Stmts	Miss	Cover	Missing
hazenlib
HazenTask.py	26	3	88%	32–34
__init__.py	53	19	64%	109–119, 150, 162, 164–166, 168–174, 179–181, 189
exceptions.py	21	4	81%	17–21
relaxometry.py	311	90	71%	248–266, 640, 699–701, 755, 803–825, 843–858, 1186–1189, 1198–1201, 1213–1226, 1229–1234, 1245–1274
utils.py	189	43	77%	61, 65, 75, 80, 117, 124–129, 140, 143–150, 170–172, 190–192, 211–213, 222, 227, 233, 284, 287, 295–300, 303, 346, 355, 371
hazenlib/tasks
acr_geometric_accuracy.py	146	56	62%	38–75, 179–195, 209–233
acr_ghosting.py	117	43	63%	33–55, 93–95, 125–127, 163–196
acr_slice_position.py	155	54	65%	53–76, 154, 215–260
acr_slice_thickness.py	154	64	58%	40–62, 188–243
acr_snr.py	139	60	57%	34–75, 100, 169–179, 212–225, 258–271
acr_spatial_resolution.py	244	74	70%	66–90, 169, 212, 225–234, 323–378
acr_uniformity.py	90	33	63%	34–56, 123–140
ghosting.py	152	53	65%	18–35, 50, 112–113, 117, 127–128, 154–156, 173–175, 221–259
relaxometry.py	7	7	0%	1–11
slice_position.py	117	23	80%	28, 37–38, 49, 103–104, 130, 210, 217–234
slice_width.py	360	53	85%	34–37, 41, 109, 168–188, 453, 458–459, 465, 470, 532–533, 782–823
snr.py	164	67	59%	48, 65–70, 164–182, 197–206, 224–234, 261–271, 276–286, 317–330, 335–343, 372–385
snr_map.py	104	1	99%	291
spatial_resolution.py	248	45	82%	36–39, 43, 64, 149, 208, 334–370
uniformity.py	79	20	75%	42–45, 51, 93–94, 101, 135–149
TOTAL	2892	812	72%

Tests	Skipped	Failures	Errors	Time
219	0 💤	0 ❌	0 🔥	3m 12s ⏱️

[sophie22]

https://realpython.com/python-main-function/
main function is not supposed to have a return value, but could call other functions that do --> will have to update the test_hazenlib unit tests that rely on evaluating the stdout to check output is as expected

[tomaroberts]

Changes look great.

Just need to resolve the GitHub Actions issue where the CLI tests fail.

[sophie22]

as discussed last week, GitHub tests fail either the CLI test when tasks complete with exit code 1 due to a string being returned from the main function, or the development tests fail: 2 tests within test_utils rely on having the string returned by main to be compared against expected values.

So to resolve these, I plan to change the main function so it doesn't return anything and printing of the result is handled before the end of main and then modifying the test_hazenlib functions to parse sys.stdout to capture the result printed.
https://pytest.org/en/7.4.x/how-to/capture-stdout-stderr.html

@tomaroberts let me know if this sounds sensible and I'll get on it

[tomaroberts]

As discussed on Teams, give it a go :)

[sophie22]

made the agreed modifications such that hazenlib.main() does not return anything (completes with exit code 0)
and modified the tests for the SNR and relaxometry tasks such that they do not rely on hazenlib.main() which now doesn't return anything, but instead call their respective run() function with equivalent arguments as passed to CLI parser.

[sophie22]

in the future, I recommend to separate hazenlib.main() into the following sub-functions:

• parse_CLI_args: purely responsible for presenting a useful help and parsing the passed arguments
  • it allows for extendability by adding new tasks or new optional arguments to existing tasks easily
• prepare: all sorts of checks on the inputs provided: includes checking that an available task was requested with applicable optional arguments, and that suitable DICOMs exist within the folder specified. Prepare dicom files to be passed to a task and create report output folder, if applicable.
• run_task: make an instance of the selected task class and call its run() function with arguments as identified above, this is what would return the result dictionary that is printed within the main() function, which returns nothing
  • this is what could be tested in the unit tests that task with known input returns calculated values asa expected

[tomaroberts]

Great! :)

Couple of lines to remove, unless you think useful to keep.

Go ahead and merge main into this branch and resolve any conflicts.