Add ContainerImage

[simleo]

Adds support for converting "cwlprov:image" entries to "ContainerImage".

[codecov]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Comparison is base (d052f77) 97.61% compared to head (8a56a22) 97.62%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #66      +/-   ##
==========================================
+ Coverage   97.61%   97.62%   +0.01%     
==========================================
  Files          12       13       +1     
  Lines        1928     1983      +55     
==========================================
+ Hits         1882     1936      +54     
- Misses         46       47       +1     

Files	Coverage Δ
src/runcrate/convert.py	97.89% <100.00%> (+0.04%)	⬆️
tests/test_cwlprov_crate_builder.py	99.89% <100.00%> (+<0.01%)	⬆️
tests/test_utils.py	100.00% <100.00%> (ø)
src/runcrate/utils.py	96.77% <96.29%> (-3.23%)	⬇️

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

[simleo]

@mr-c I had to pin cwl-utils to 0.29 since 0.30 makes most tests fail: https://github.com/ResearchObject/runcrate/actions/runs/6746561680/job/18340868647

[simleo]

I've tried to update the code by changing the occurrences of type to type_ where appropriate, but two tests (test_revsort and test_secondary_files) still fail with:

data_dir = PosixPath('/tmp/runcrate/tests/data'), tmpdir = PosixPath('/tmp/pytest-of-simleo/pytest-16/test_revsort0')

    def test_revsort(data_dir, tmpdir):
        root = data_dir / "revsort-run-1"
        output = tmpdir / "revsort-run-1-crate"
        license = "Apache-2.0"
        readme = data_dir / "README.md"
        inputs_file = root / "workflow" / "primary-job.json"
        workflow_name = "RevSort"
>       builder = ProvCrateBuilder(root, workflow_name=workflow_name, license=license, readme=readme)

tests/test_cwlprov_crate_builder.py:39: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
src/runcrate/convert.py:203: in __init__
    self.cwl_defs = get_workflow(self.wf_path)
src/runcrate/convert.py:182: in get_workflow
    defs = load_document_by_yaml(json_wf, wf_path.absolute().as_uri(), load_all=True)
/home/simleo/git/runcrate/venv/lib/python3.11/site-packages/cwl_utils/parser/__init__.py:227: in load_document_by_yaml
    result = cwl_v1_0.load_document_by_yaml(
/home/simleo/git/runcrate/venv/lib/python3.11/site-packages/cwl_utils/parser/cwl_v1_0.py:20290: in load_document_by_yaml
    result, metadata = _document_load(
/home/simleo/git/runcrate/venv/lib/python3.11/site-packages/cwl_utils/parser/cwl_v1_0.py:876: in _document_load
    loader.load(doc["$graph"], baseuri, loadingOptions),
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

self = CommandLineTool | ExpressionTool | Workflow | array<CommandLineTool | ExpressionTool | Workflow>
doc = [{'class': 'Workflow', 'doc': 'Reverse the lines in a document, then sort those lines.', 'hints': [{'class': 'DockerRe...ommand': 'sort', 'class': 'CommandLineTool', 'doc': 'Sort lines using the `sort` command', 'id': '#sorttool.cwl', ...}]
baseuri = 'file:///tmp/runcrate/tests/data/revsort-run-1/workflow/packed.cwl'
loadingOptions = <cwl_utils.parser.cwl_v1_0.LoadingOptions object at 0x7f345e586990>, docRoot = None, lc = []

    def load(self, doc, baseuri, loadingOptions, docRoot=None, lc=None):
        # type: (Any, str, LoadingOptions, Optional[str], Optional[List[Any]]) -> Any
        errors = []
    
        if lc is None:
            lc = []
    
        for t in self.alternates:
            try:
                return t.load(doc, baseuri, loadingOptions, docRoot=docRoot, lc=lc)
            except ValidationException as e:
                if isinstance(t, _ArrayLoader) and len(self.alternates) > 1:
                    continue
                if isinstance(doc, (CommentedMap, dict)):
                    if "class" in doc:
                        if str(doc.get("class")) == str(t):
                            errors.append(
                                ValidationException(
                                    f"Object `{baseuri.split('/')[-1]}` is not valid because:",
                                    SourceLine(doc, next(iter(doc)), str),
                                    [e],
                                )
                            )
                    else:
                        if "array" in str(t):
                            continue
                        else:
                            if "id" in doc:
                                id = baseuri.split("/")[-1] + "#" + str(doc.get("id"))
                                if "id" in lc:
                                    errors.append(
                                        ValidationException(
                                            f"checking object `{id}`",
                                            SourceLine(lc, "id", str),
                                            [e],
                                        )
                                    )
                                else:
                                    errors.append(
                                        ValidationException(
                                            f"checking object `{id}`",
                                            SourceLine(lc, doc.get("id"), str),
                                            [e],
                                        )
                                    )
                            else:
                                if not isinstance(
                                    t, (_PrimitiveLoader)
                                ):  # avoids 'tried <class "NoneType"> was {x}' errors
                                    errors.append(
                                        ValidationException(f"tried `{t}` but", None, [e])
                                    )
                else:
                    # avoids "tried <class "CWLType"> but x" and instead returns the values for parsing
                    errors.append(ValidationException("", None, [e]))
    
        if isinstance(doc, (CommentedMap, dict)) and "class" in doc:
            if str(doc.get("class")) not in str(self.alternates):
                errors.append(
                    ValidationException(
                        "Field `class` contains undefined reference to "
                        + "`"
                        + "/".join(baseuri.split("/")[0:-1])
                        + "/"
                        + str(doc.get("class"))
                        + "`",
                        SourceLine(doc, "class", str),
                        [],
                    )
                )
>       raise ValidationException("", None, errors, "*")
E       schema_salad.exceptions.ValidationException: Value is a array, but valid type for this field is an object.
E       Value is a array, but valid type for this field is an object.
E       Value is a array, but valid type for this field is an object.

[mr-c]

@simleo yep, I ran into the same thing; I've filled an issue and assigned it to the author of the change common-workflow-language/schema_salad#766