Change Base image purpose code

[josecelano]

Issue: #143

[github-actions]

MegaLinter status: ✅ SUCCESS

Descriptor	Linter	Files	Fixed	Errors	Elapsed time
✅ ACTION	actionlint	2		0	0.05s
✅ CREDENTIALS	secretlint	yes		no	1.29s
✅ GIT	git_diff	yes		no	0.01s
✅ JSON	eslint-plugin-jsonc	8		0	1.48s
✅ JSON	jsonlint	2		0	0.67s
✅ JSON	prettier	2		0	1.74s
✅ JSON	v8r	2		0	2.86s
✅ MARKDOWN	markdownlint	16		0	0.6s
✅ MARKDOWN	markdown-link-check	16		0	13.89s
✅ MARKDOWN	markdown-table-formatter	16		0	0.37s
✅ SPELL	cspell	47		0	3.79s
✅ SPELL	misspell	47		0	0.09s
✅ YAML	prettier	3		0	1.49s
✅ YAML	v8r	3		0	4.25s
✅ YAML	yamllint	3		0	0.19s

See errors details in artifact MegaLinter reports on CI Job page

You could have the same capabilities but better runtime performances if you use a MegaLinter flavor:

• megalinter/megalinter/flavors/documentation@v5 (39 linters)- megalinter/megalinter/flavors/dart@v5 (40 linters)- megalinter/megalinter/flavors/ruby@v5 (40 linters)- megalinter/megalinter/flavors/rust@v5 (40 linters)- megalinter/megalinter/flavors/scala@v5 (40 linters)- megalinter/megalinter/flavors/swift@v5 (40 linters)- megalinter/megalinter/flavors/go@v5 (41 linters)- megalinter/megalinter/flavors/java@v5 (41 linters)- megalinter/megalinter/flavors/salesforce@v5 (42 linters)- megalinter/megalinter/flavors/php@v5 (44 linters)- megalinter/megalinter/flavors/terraform@v5 (45 linters)- megalinter/megalinter/flavors/dotnet@v5 (46 linters)- megalinter/megalinter/flavors/javascript@v5 (48 linters)- megalinter/megalinter/flavors/python@v5 (48 linters)

[josecelano]

The Gold Drawings Processing workflow fails because we are using the new Base image purpose code (52):

I'm working on the Librarian to fix it.

[josecelano]

I've fixed the previous error but now I have a new one: https://github.com/Nautilus-Cyberneering/chinese-ideographs/runs/5375335931?check_suite_focus=true#step:8:56

We are pulling from DVC remote storage only new or modified images but we validate the image size also for the renamed images. In this case, I've only renamed the images and it fails because the image is not found.

We have to modified this function:

def extract_list_of_media_file_changes_from_dvc_diff_output(
    dvc_diff, only_basename=True
):
    return extract_added_and_modified_and_renamed_files_from_dvc_diff(
        dvc_diff, only_basename
    )

To not include the renamed ones.

cc @yeraydavidrodriguez

EDIT: I've created a new issue on the Librarian repo.

[josecelano]

There was one more bug. Filepath validation for renamed files was failing because renamed files in dvc diff are not a plain path (string) but a dict with the old and new name. Only the new name has to be validated.

[josecelano]

There is another bug: nautilus-cyberneering/nautilus-librarian#101. The check_image_changes action does not work with renamed files either :-(

I've added a new task to issue nautilus-cyberneering/nautilus-librarian#100 to refactor that code. I think this different format for renamed files is causing a lot of problems.

And another thing we should consider is using more or full typing. We are using types only in some cases and I think we should enforce it like in Typescript.

@da2ce7 @yeraydavidrodriguez do you agree?

I wonder if you can enforce full typing with mypy.

[josecelano]

It seems the workflow is finally working. We should definitively improve the command output: