semantic-benchmark 0.4.2__tar.gz → 0.4.4__tar.gz

{semantic_benchmark-0.4.2/src/semantic_benchmark.egg-info → semantic_benchmark-0.4.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: semantic-benchmark
-Version: 0.4.2
+Version: 0.4.4
 Summary: Reusable helpers for semantic benchmark metadata, RO-Crate aggregation, and RoHub provenance workflows.
 Author-email: Mahdi Jafarkhani <mahdi.jafarkhani@tik.uni-stuttgart.de>
 License-Expression: MIT
@@ -28,9 +28,9 @@ Provides-Extra: test
 Requires-Dist: pytest; extra == "test"
 Provides-Extra: rocrate
 Requires-Dist: rocrate; extra == "rocrate"
-Requires-Dist: roc-validator; extra == "rocrate"
+Requires-Dist: roc-validator==0.12.2; extra == "rocrate"
 Provides-Extra: all
-Requires-Dist: roc-validator; extra == "all"
+Requires-Dist: roc-validator==0.12.2; extra == "all"
 Requires-Dist: rocrate; extra == "all"
 Dynamic: license-file
 
@@ -96,11 +96,18 @@ It accepts `--provenance_folderpath` (the RO-Crate ZIP), `--benchmark-name`,
 `--use-production-rohub` arguments. The software URL and version are read from
 the SoftwareApplication node in the RO-Crate ZIP.
 
-`semantic_benchmark.rocrate.create_main_ro(...)` can validate the generated
-aggregate crate by passing `validation_profile`. The package writes the RO-Crate
+`semantic_benchmark.rocrate.create_main_ro(...)` always validates the generated
+aggregate crate against `workflow-run-crate-0.5`. The package writes the RO-Crate
 zip, unpacks it to a validation directory, and runs `validate_rocrate(...)`.
 Aggregate creation requires `subcrate_paths`, the exact reporter RO-Crate ZIPs
-from completed runs. The CLI accepts one `--subcrate-path` per run.
+from completed runs, and `workflow_path`, a YAML file used as the main workflow.
+The CLI accepts `--workflow-path` and one `--subcrate-path` per run. Each subcrate
+is the result of a `CreateAction` whose instrument is the YAML workflow; matched
+configuration values are action inputs and extracted metrics are action results.
+Pass `agent_name` and `agent_orcid` from the reporter configuration's
+`report-rocrate-agent-name` and `report-rocrate-agent-orcid` values. They identify
+the workflow author and creator in the aggregate crate. The CLI accepts
+`--agent-name` and `--agent-orcid`.
 
 Repository-specific projects can override the packaged RoHub defaults with:
 

{semantic_benchmark-0.4.2 → semantic_benchmark-0.4.4}/README.md

@@ -60,11 +60,18 @@ It accepts `--provenance_folderpath` (the RO-Crate ZIP), `--benchmark-name`,
 `--use-production-rohub` arguments. The software URL and version are read from
 the SoftwareApplication node in the RO-Crate ZIP.
 
-`semantic_benchmark.rocrate.create_main_ro(...)` can validate the generated
-aggregate crate by passing `validation_profile`. The package writes the RO-Crate
+`semantic_benchmark.rocrate.create_main_ro(...)` always validates the generated
+aggregate crate against `workflow-run-crate-0.5`. The package writes the RO-Crate
 zip, unpacks it to a validation directory, and runs `validate_rocrate(...)`.
 Aggregate creation requires `subcrate_paths`, the exact reporter RO-Crate ZIPs
-from completed runs. The CLI accepts one `--subcrate-path` per run.
+from completed runs, and `workflow_path`, a YAML file used as the main workflow.
+The CLI accepts `--workflow-path` and one `--subcrate-path` per run. Each subcrate
+is the result of a `CreateAction` whose instrument is the YAML workflow; matched
+configuration values are action inputs and extracted metrics are action results.
+Pass `agent_name` and `agent_orcid` from the reporter configuration's
+`report-rocrate-agent-name` and `report-rocrate-agent-orcid` values. They identify
+the workflow author and creator in the aggregate crate. The CLI accepts
+`--agent-name` and `--agent-orcid`.
 
 Repository-specific projects can override the packaged RoHub defaults with:
 

{semantic_benchmark-0.4.2 → semantic_benchmark-0.4.4}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "semantic-benchmark"
-version = "0.4.2"
+version = "0.4.4"
 description = "Reusable helpers for semantic benchmark metadata, RO-Crate aggregation, and RoHub provenance workflows."
 readme = "README.md"
 requires-python = ">=3.10"
@@ -38,10 +38,10 @@ test = [
 ]
 rocrate = [
     "rocrate",
-    "roc-validator",
+    "roc-validator==0.12.2",
 ]
 all = [
-    "roc-validator",
+    "roc-validator==0.12.2",
     "rocrate",
 ]
 

{semantic_benchmark-0.4.2 → semantic_benchmark-0.4.4}/src/semantic_benchmark/rocrate/create.py

@@ -6,13 +6,13 @@ import argparse
 import json
 import logging
 import re
-import shutil
 import tempfile
 import uuid
 import zipfile
 from pathlib import Path
-from typing import Any, Literal, TypedDict, get_args
+from typing import Any, TypedDict
 
+from rocrate.model.computationalworkflow import ComputationalWorkflow
 from rocrate.model.softwareapplication import SoftwareApplication
 from rocrate.rocrate import ROCrate
 from semantic_benchmark import semantics
@@ -21,42 +21,33 @@ from semantic_benchmark.rocrate.validation import validate_rocrate
 LOG_FORMAT = "%(levelname)s:%(name)s:%(message)s"
 LOGGER = logging.getLogger(__name__)
 
-WorkflowLanguage = Literal["nextflow", "snakemake"]
-
 ROCRATE_CONFORMS_TO = [
     {"@id": "https://w3id.org/ro/crate/1.1"},
-    {"@id": "https://w3id.org/workflowhub/workflow-ro-crate/1.0"},
 ]
 ROOT_DATASET_CONFORMS_TO = [
-    {"@id": "https://w3id.org/ro/wfrun/process/0.4"},
-    {"@id": "https://w3id.org/ro/wfrun/workflow/0.4"},
-    {"@id": "https://w3id.org/ro/wfrun/provenance/0.4"},
-    {"@id": "https://w3id.org/workflowhub/workflow-ro-crate/1.0"},
+    {"@id": "https://w3id.org/ro/wfrun/process/0.6"},
+    {"@id": "https://w3id.org/ro/wfrun/workflow/0.6"},
+    {"@id": "https://w3id.org/workflowhub/workflow-ro-crate/1.1"},
 ]
+AGGREGATE_VALIDATION_PROFILE = "workflow-run-crate-0.5"
 PROFILE_CREATIVE_WORKS = (
     {
-        "@id": "https://w3id.org/ro/wfrun/process/0.4",
+        "@id": "https://w3id.org/ro/wfrun/process/0.6",
         "@type": "CreativeWork",
         "name": "Process Run Crate",
-        "version": "0.4",
+        "version": "0.6",
     },
     {
-        "@id": "https://w3id.org/ro/wfrun/workflow/0.4",
+        "@id": "https://w3id.org/ro/wfrun/workflow/0.6",
         "@type": "CreativeWork",
         "name": "Workflow Run Crate",
-        "version": "0.4",
-    },
-    {
-        "@id": "https://w3id.org/ro/wfrun/provenance/0.4",
-        "@type": "CreativeWork",
-        "name": "Provenance Run Crate",
-        "version": "0.4",
+        "version": "0.6",
     },
     {
-        "@id": "https://w3id.org/workflowhub/workflow-ro-crate/1.0",
+        "@id": "https://w3id.org/workflowhub/workflow-ro-crate/1.1",
         "@type": "CreativeWork",
         "name": "Workflow RO-Crate",
-        "version": "1.0",
+        "version": "1.1",
     },
 )
 
@@ -150,23 +141,6 @@ def _add_subcrates_to_main(
         )
 
 
-def _create_action_object_ids(input_path: Path, subcrates: list[Path]) -> dict[str, str]:
-    """Map run folder names to relative subcrate object ids.
-
-    Args:
-        input_path: Root simulation result directory.
-        subcrates: Reporter RO-Crate ZIPs for completed runs.
-
-    Returns:
-        Mapping from run folder name to the subcrate ZIP path relative to
-        ``input_path``.
-    """
-    object_ids: dict[str, str] = {}
-    for subcrate in subcrates:
-        object_ids.setdefault(subcrate.parent.name, str(subcrate.relative_to(input_path)))
-    return object_ids
-
-
 def _formal_parameter_key(part: semantics.ParameterEntry) -> tuple[Any, ...]:
     """Build a stable de-duplication key for a benchmark parameter.
 
@@ -205,15 +179,17 @@ def _formal_parameter_payload(
     }
 
     unit = getattr(part, "unit", None)
-    payload["additionalType"] = ""
+    payload["additionalType"] = (
+        "Float" if isinstance(part, (semantics.NumericalParameter, semantics.NumericalVariable)) else "Text"
+    )
 
     if unit is not None:
         payload["unitText"] = {"@id": unit}
 
     if isinstance(part, semantics.NumericalParameter):
-        payload["defaultValue"] = part.numerical_value
+        payload["value"] = part.numerical_value
     elif isinstance(part, semantics.TextParameter):
-        payload["defaultValue"] = part.string_value
+        payload["value"] = part.string_value
     elif (
         isinstance(part, semantics.NumericalVariable)
         and part.quantity_kind is not None
@@ -515,10 +491,10 @@ def _metric_result_payload(
         "@id": result_id,
         "@type": "PropertyValue",
         "name": metric.label,
-        "defaultValue": value,
+        "value": value,
     }
     if metric.unit:
-        payload["additionalType"] = metric.unit
+        payload["unitText"] = {"@id": metric.unit}
     return payload
 
 
@@ -597,12 +573,12 @@ def _configuration_entries_for_step(
 
 def _add_run_actions(
     crate: ROCrate,
-    subfolders: list[Path],
-    object_ids_by_run: dict[str, str],
+    subcrates: list[Path],
+    input_path: Path,
     processing_steps: list[semantics.ProcessingStep],
     configuration_entries: list[ConfigurationEntry],
     run_results_by_name: dict[str, list[dict[str, str]]],
-    software_id: str,
+    workflow_id: str,
 ) -> None:
     """Add run action nodes connecting runs, configurations, and results.
 
@@ -618,14 +594,11 @@ def _add_run_actions(
     Returns:
         None. The function mutates ``crate`` by adding ``CreateAction`` nodes.
     """
-    for run_folder in subfolders:
+    for subcrate in subcrates:
+        run_folder = subcrate.parent
         run_name = run_folder.name
-        run_object_id = object_ids_by_run.get(run_name)
-        if not run_object_id:
-            continue
-
         result_ids = run_results_by_name.get(run_name, [])
-
+        object_ids = []
         for processing_step in processing_steps:
             step_configuration_entries = _configuration_entries_for_step(
                 configuration_entries, processing_step
@@ -634,19 +607,20 @@ def _add_run_actions(
                 run_folder, step_configuration_entries
             )
 
-            step_name = processing_step.label or processing_step.id
-            run_action: dict[str, Any] = {
-                "@id": _new_jsonld_id(),
-                "@type": "CreateAction",
-                "name": f"{step_name} {run_name}",
-                "object": [{"@id": run_object_id}],
-                "instrument": {"@id": software_id},
-            }
             if config_id:
-                run_action["object"].append({"@id": config_id})
-            if result_ids:
-                run_action["result"] = result_ids
-            crate.add_jsonld(run_action)
+                object_ids.append({"@id": config_id})
+        run_action: dict[str, Any] = {
+            "@id": _new_jsonld_id(),
+            "@type": "CreateAction",
+            "name": f"Benchmark run {run_name}",
+            "instrument": {"@id": workflow_id},
+            "result": [{"@id": str(subcrate.relative_to(input_path))}, *result_ids],
+        }
+        if object_ids:
+            run_action["object"] = object_ids
+        action_id = run_action["@id"]
+        crate.add_jsonld(run_action)
+        crate.root_dataset.append_to("mentions", {"@id": action_id})
 
 
 def _configure_crate_metadata(
@@ -714,33 +688,48 @@ def _add_software_node(
     )
 
 
+def _add_yaml_language_node(crate: ROCrate) -> dict[str, str]:
+    """Add the YAML language node and return its JSON-LD reference."""
+    yaml_id = "https://www.wikidata.org/entity/Q281876"
+    crate.add_jsonld(
+        {
+            "@id": yaml_id,
+            "@type": "ComputerLanguage",
+            "identifier": yaml_id,
+            "name": "YAML",
+            "url": "https://yaml.org/",
+        }
+    )
+    return {"@id": yaml_id}
+
+
 def _add_workflow_node(
     crate: ROCrate,
     workflow_path: Path,
     software_id: str,
-    workflow_id: str,
-    lang: WorkflowLanguage,
+    agent_name: str,
+    agent_orcid: str,
 ) -> None:
     """Add the main workflow file to the aggregate crate.
 
     Args:
         crate: Aggregate RO-Crate being built.
         workflow_path: Location of the workflow file to add.
-        software_id: JSON-LD id of the software application linked as ``hasPart``.
-        workflow_id: JSON-LD id of the workflow file.
-        lang: Workflow language used by the workflow file.
+        software_id: JSON-LD id of the software application requirement.
+        agent_name: Name of the workflow author and creator.
+        agent_orcid: ORCID URL identifying the author and creator.
 
     Returns:
         None. The function mutates ``crate``.
     """
-    crate.add_workflow(
-        source=str(workflow_path),
-        lang=lang,
-        properties={
-            "hasPart": {"@id": software_id},
-            "mainEntity": {"@id": workflow_id},
-        },
-    )
+    workflow = crate.add(ComputationalWorkflow(crate, source=str(workflow_path)))
+    workflow["encodingFormat"] = "application/yaml"
+    workflow["softwareRequirements"] = {"@id": software_id}
+    workflow["programmingLanguage"] = _add_yaml_language_node(crate)
+    person = {"@id": agent_orcid}
+    workflow["author"] = person
+    workflow["creator"] = person
+    crate.add_jsonld({"@id": agent_orcid, "@type": "Person", "name": agent_name})
 
 
 def create_main_ro(
@@ -752,12 +741,11 @@ def create_main_ro(
     crate_name: str,
     crate_description: str,
     subcrate_paths: list[str | Path],
-    workflow_path: str | Path | None = None,
-    lang: WorkflowLanguage = "snakemake",
-    validation_profile: str | None = None,
-    validation_dir: str | Path | None = None,
+    workflow_path: str | Path = "",
     software_url: str | None = None,
     software_version: str | None = None,
+    agent_name: str | None = None,
+    agent_orcid: str | None = None,
 ) -> None:
     """Create and write an aggregate RO-Crate for a benchmark result directory.
 
@@ -770,16 +758,12 @@ def create_main_ro(
         crate_license: License URL stored on the aggregate RO-Crate.
         crate_name: Human-readable name stored on the aggregate RO-Crate.
         crate_description: Description stored on the aggregate RO-Crate.
-        workflow_path: Location of the workflow file to add to the crate. Defaults
-            to the workflow found alongside the first subcrate.
-        lang: Workflow language. Defaults to ``snakemake``.
-        validation_profile: Optional RO-Crate profile identifier. When provided,
-            the written crate is unpacked and validated against this profile.
-        validation_dir: Optional directory used for unpacked validation content.
-            Defaults to ``unpacked_rocrate`` next to ``rocrate_path``.
+        workflow_path: Required YAML workflow file to add as the main entity.
         software_url: Stable software identifier (for example, a zbMATH URL).
             If omitted, a local identifier is generated.
         software_version: Software version to record, if known.
+        agent_name: Name supplied by the reporter configuration.
+        agent_orcid: ORCID URL supplied by the reporter configuration.
         subcrate_paths: Exact reporter RO-Crate ZIPs to include.
 
     Returns:
@@ -788,20 +772,23 @@ def create_main_ro(
 
     Raises:
         NotADirectoryError: If ``path`` is not a directory.
-        FileNotFoundError: If the explicit or inferred workflow path is not a file.
-        ValueError: If ``lang`` is unsupported or no matching subcrate zip files
-            are found.
+        FileNotFoundError: If the workflow path is not a file.
+        ValueError: If no matching subcrate zip files are found.
     """
     crate = ROCrate(version="1.1")
     input_path = Path(path)
 
     if not input_path.is_dir():
         raise NotADirectoryError(f"{path} is not a valid directory")
-    if lang not in get_args(WorkflowLanguage):
-        raise ValueError(
-            f"Unsupported workflow language {lang!r}; expected one of "
-            f"{', '.join(get_args(WorkflowLanguage))}"
-        )
+    if not workflow_path:
+        raise ValueError("workflow_path is required")
+    if not agent_name or not agent_orcid:
+        raise ValueError("agent_name and agent_orcid are required")
+    workflow_file = Path(workflow_path).resolve()
+    if not workflow_file.is_file():
+        raise FileNotFoundError(f"{workflow_file} is not a valid workflow file")
+    if workflow_file.suffix.lower() not in {".yaml", ".yml"}:
+        raise ValueError("workflow_path must be a YAML file")
     LOGGER.info(
         "Creating aggregate RO-Crate from simulation results in %s...",
         input_path,
@@ -825,35 +812,23 @@ def create_main_ro(
 
         _add_subcrates_to_main(crate, subcrates, input_path)
 
-        object_ids_by_run = _create_action_object_ids(input_path, subcrates)
         configuration_entries = _add_configuration_nodes(crate, benchmark_object)
         run_results = _add_evaluates_nodes(
             crate, benchmark_object, extracted_subfolders
         )
         run_results_by_name = _run_results_by_name(run_results)
 
-        workflow_id = get_workflow_id(
-            subcrates[0],
-            fallback=Path(workflow_path).name if workflow_path is not None else "Snakefile",
-        )
-        workflow_file = (
-            Path(workflow_path)
-            if workflow_path is not None
-            else extraction_path / subcrates[0].parent.relative_to(input_path) / workflow_id
-        )
-        if not workflow_file.is_file():
-            raise FileNotFoundError(f"{workflow_file} is not a valid workflow file")
-
+        workflow_id = workflow_file.name
         software_id = software_url or _new_jsonld_id()
 
         _add_run_actions(
             crate=crate,
-            subfolders=subfolders,
-            object_ids_by_run=object_ids_by_run,
+            subcrates=subcrates,
+            input_path=input_path,
             processing_steps=benchmark_object.processing_steps,
             configuration_entries=configuration_entries,
             run_results_by_name=run_results_by_name,
-            software_id=software_id,
+            workflow_id=workflow_id,
         )
         _configure_crate_metadata(
             crate,
@@ -864,28 +839,18 @@ def create_main_ro(
         )
         _add_software_node(crate, software_id, software_name, software_version)
         _add_profile_creative_works(crate)
-        _add_workflow_node(crate, workflow_file, software_id, workflow_id, lang)
+        _add_workflow_node(crate, workflow_file, software_id, agent_name, agent_orcid)
 
         crate.write_zip(rocrate_path)
 
-    if validation_profile:
-        validation_path = Path(validation_dir) if validation_dir else (
-            Path(rocrate_path).parent / "unpacked_rocrate"
-        )
-        if validation_path.exists():
-            shutil.rmtree(validation_path)
-        validation_path.mkdir(parents=True, exist_ok=True)
-
-        try:
-            with zipfile.ZipFile(rocrate_path, "r") as zip_ref:
-                zip_ref.extractall(validation_path)
+    with tempfile.TemporaryDirectory(prefix="benchmark-validation-") as validation_dir:
+        with zipfile.ZipFile(rocrate_path, "r") as zip_ref:
+            zip_ref.extractall(validation_dir)
 
-            validate_rocrate(
-                rocrate_path=str(validation_path),
-                profile=validation_profile,
-            )
-        finally:
-            shutil.rmtree(validation_path)
+        validate_rocrate(
+            rocrate_path=validation_dir,
+            profile=AGGREGATE_VALIDATION_PROFILE,
+        )
 
 
 def parse_args(argv: list[str] | None = None) -> argparse.Namespace:
@@ -948,30 +913,20 @@ def parse_args(argv: list[str] | None = None) -> argparse.Namespace:
         required=True,
         help="Description recorded in the generated aggregate RO-Crate",
     )
-    parser.add_argument(
-        "--lang",
-        dest="workflow_lang",
-        default="snakemake",
-        choices=get_args(WorkflowLanguage),
-        help="Language of the workflow file (default: snakemake)",
-    )
     parser.add_argument(
         "--workflow-path",
-        default=None,
-        help=(
-            "Path to the workflow file added to the generated RO-Crate "
-            "(default: workflow alongside the first subcrate)"
-        ),
+        required=True,
+        help="Path to the YAML workflow file used as the main entity",
     )
     parser.add_argument(
-        "--validation-profile",
-        default=None,
-        help="Optional RO-Crate profile identifier used to validate the generated crate",
+        "--agent-name",
+        required=True,
+        help="Workflow author and creator name from report-rocrate-agent-name",
     )
     parser.add_argument(
-        "--validation-dir",
-        default=None,
-        help="Optional directory for unpacked validation content",
+        "--agent-orcid",
+        required=True,
+        help="Workflow author and creator ORCID URL from report-rocrate-agent-orcid",
     )
 
     return parser.parse_args(argv)
@@ -999,22 +954,10 @@ def main() -> None:
         crate_description=args.crate_description,
         subcrate_paths=args.subcrate_paths,
         workflow_path=args.workflow_path,
-        lang=args.workflow_lang,
-        validation_profile=args.validation_profile,
-        validation_dir=args.validation_dir,
+        agent_name=args.agent_name,
+        agent_orcid=args.agent_orcid,
     )
 
 
-def get_workflow_id(subcrate: str | Path, fallback: str = "Snakefile") -> str:
-    """Return the computational workflow id stored in a subcrate."""
-    crate = ROCrate(subcrate)
-
-    for e in crate.get_entities():
-        if e.type == ["ComputationalWorkflow"]:
-            return e.id
-
-    return fallback
-
-
 if __name__ == "__main__":
     main()

{semantic_benchmark-0.4.2 → semantic_benchmark-0.4.4}/src/semantic_benchmark/rohub/provenance.py

@@ -194,14 +194,14 @@ def _value_block(
     {parent} {relation} ?{node_prefix}_{safe_name} .
     ?{node_prefix}_{safe_name} a {_node_type(node_prefix)} ;
         {name_predicate} "{escaped_name}" ;
-        schema:defaultValue ?{safe_name} .
+        (schema:value|schema:defaultValue) ?{safe_name} .
     """.strip()
 
 
 def _parameter_block(
     name: str,
     var_map: dict[str, str],
-    name_predicate: str = "schema:name",
+    name_predicate: str = FOAF_NAME,
 ) -> str:
     """Build a graph pattern that extracts one configuration parameter."""
     return _value_block(
@@ -217,7 +217,7 @@ def _parameter_block(
 def _metric_block(
     name: str,
     var_map: dict[str, str],
-    name_predicate: str = "schema:name",
+    name_predicate: str = FOAF_NAME,
 ) -> str:
     """Build an optional graph pattern that extracts one run metric."""
     value_block = _value_block(
@@ -300,9 +300,9 @@ def build_dynamic_query(
     inner_query = _join_blocks(
         "\n".join(_create_action_links(include_tool=True)),
         "\n".join(
-            _parameter_block(name, var_map, FOAF_NAME) for name in parameters
+            _parameter_block(name, var_map) for name in parameters
         ),
-        "\n".join(_metric_block(name, var_map, FOAF_NAME) for name in metrics),
+        "\n".join(_metric_block(name, var_map) for name in metrics),
     )
 
     order_name = parameters[0] if parameters else None

{semantic_benchmark-0.4.2 → semantic_benchmark-0.4.4}/src/semantic_benchmark/runner.py

@@ -203,7 +203,9 @@ def create_aggregate_rocrate(
     crate_name: str,
     crate_description: str,
     subcrate_paths: list[str | Path],
-    validation_dir: Path | None = None,
+    workflow_path: str | Path,
+    agent_name: str,
+    agent_orcid: str,
     software_url: str | None = None,
     software_version: str | None = None,
 ) -> None:
@@ -212,9 +214,6 @@ def create_aggregate_rocrate(
     # the core runner helpers remain available without the ``rocrate`` extra.
     from . import rocrate
 
-    options = {}
-    if validation_dir is not None:
-        options["validation_dir"] = validation_dir
     rocrate.create_main_ro(
         path=str(results_dir),
         benchmark_object=benchmark,
@@ -226,6 +225,7 @@ def create_aggregate_rocrate(
         crate_name=crate_name,
         crate_description=crate_description,
         subcrate_paths=subcrate_paths,
-        validation_profile="provenance-run-crate-0.5",
-        **options,
+        workflow_path=workflow_path,
+        agent_name=agent_name,
+        agent_orcid=agent_orcid,
     )

{semantic_benchmark-0.4.2 → semantic_benchmark-0.4.4/src/semantic_benchmark.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: semantic-benchmark
-Version: 0.4.2
+Version: 0.4.4
 Summary: Reusable helpers for semantic benchmark metadata, RO-Crate aggregation, and RoHub provenance workflows.
 Author-email: Mahdi Jafarkhani <mahdi.jafarkhani@tik.uni-stuttgart.de>
 License-Expression: MIT
@@ -28,9 +28,9 @@ Provides-Extra: test
 Requires-Dist: pytest; extra == "test"
 Provides-Extra: rocrate
 Requires-Dist: rocrate; extra == "rocrate"
-Requires-Dist: roc-validator; extra == "rocrate"
+Requires-Dist: roc-validator==0.12.2; extra == "rocrate"
 Provides-Extra: all
-Requires-Dist: roc-validator; extra == "all"
+Requires-Dist: roc-validator==0.12.2; extra == "all"
 Requires-Dist: rocrate; extra == "all"
 Dynamic: license-file
 
@@ -96,11 +96,18 @@ It accepts `--provenance_folderpath` (the RO-Crate ZIP), `--benchmark-name`,
 `--use-production-rohub` arguments. The software URL and version are read from
 the SoftwareApplication node in the RO-Crate ZIP.
 
-`semantic_benchmark.rocrate.create_main_ro(...)` can validate the generated
-aggregate crate by passing `validation_profile`. The package writes the RO-Crate
+`semantic_benchmark.rocrate.create_main_ro(...)` always validates the generated
+aggregate crate against `workflow-run-crate-0.5`. The package writes the RO-Crate
 zip, unpacks it to a validation directory, and runs `validate_rocrate(...)`.
 Aggregate creation requires `subcrate_paths`, the exact reporter RO-Crate ZIPs
-from completed runs. The CLI accepts one `--subcrate-path` per run.
+from completed runs, and `workflow_path`, a YAML file used as the main workflow.
+The CLI accepts `--workflow-path` and one `--subcrate-path` per run. Each subcrate
+is the result of a `CreateAction` whose instrument is the YAML workflow; matched
+configuration values are action inputs and extracted metrics are action results.
+Pass `agent_name` and `agent_orcid` from the reporter configuration's
+`report-rocrate-agent-name` and `report-rocrate-agent-orcid` values. They identify
+the workflow author and creator in the aggregate crate. The CLI accepts
+`--agent-name` and `--agent-orcid`.
 
 Repository-specific projects can override the packaged RoHub defaults with:
 

{semantic_benchmark-0.4.2 → semantic_benchmark-0.4.4}/src/semantic_benchmark.egg-info/requires.txt

@@ -5,12 +5,12 @@ rocrate
 rohub
 
 [all]
-roc-validator
+roc-validator==0.12.2
 rocrate
 
 [rocrate]
 rocrate
-roc-validator
+roc-validator==0.12.2
 
 [test]
 pytest