add an optional test to make sure full validate produce is run when l…

.github/workflows/make-tests.yaml

@@ -27,4 +27,5 @@ jobs:
           pip install .
           pip install pytest
       - name: run make on rules
+        # reminder: please run `make travis_test_full` locally to include slow running tests not suitable for CI
         run: make travis_test

.gitignore

@@ -25,8 +25,6 @@ var/
 *.egg
 # keep poetry files out of the repo for now, until we decide if we are moving to a .toml file specification
 # for requirements in the future.
-pyproject.toml
-poetry.lock
 
 # PyInstaller
 #  Usually these files are written by a python script from a template

Makefile

@@ -8,22 +8,47 @@ PACKAGES = ontobio prefixcommons
 subpackage_tests: $(patsubst %,test-%,$(PACKAGES))
 
 test:
-	pytest tests/*.py tests/unit/
+	pytest -m "not slow" tests/*.py tests/unit/
 
 debug_test:
-	pytest -s -vvvv tests/*.py
+	pytest -m "not slow" -s -vvvv tests/*.py
 
 t-%:
-	pytest tests/test_$*.py
+	pytest -m "not slow" tests/test_$*.py
 
 tv-%:
-	pytest -s tests/test_$*.py
+	pytest -m "not slow" -s tests/test_$*.py
 
 foo:
 	which pytest
 
 # only run local tests
 travis_test:
+	@if [ -d ".venv" ] && [ -f "pyproject.toml" ]; then \
+		echo "Running tests in Poetry environment..."; \
+		poetry run pytest -m "not slow" tests/test_*local*.py tests/test_*parse*.py tests/test*writer*.py tests/test_qc.py \
+		tests/test_rdfgen.py tests/test_phenosim_engine.py tests/test_ontol.py \
+		tests/test_validation_rules.py tests/unit/test_annotation_scorer.py \
+		tests/test_goassociation_model.py tests/test_relations.py \
+		tests/unit/test_golr_search_query.py tests/unit/test_owlsim2_api.py \
+		tests/test_collections.py \
+		tests/test_gocamgen.py \
+		tests/test_gpi_isoform_replacement.py \
+		tests/test_validate_cli.py; \
+	else \
+		pytest -m "not slow" tests/test_*local*.py tests/test_*parse*.py tests/test*writer*.py tests/test_qc.py \
+		tests/test_rdfgen.py tests/test_phenosim_engine.py tests/test_ontol.py \
+		tests/test_validation_rules.py tests/unit/test_annotation_scorer.py \
+		tests/test_goassociation_model.py tests/test_relations.py \
+		tests/unit/test_golr_search_query.py tests/unit/test_owlsim2_api.py \
+		tests/test_collections.py \
+		tests/test_gocamgen.py \
+		tests/test_gpi_isoform_replacement.py \
+		tests/test_validate_cli.py; \
+	fi
+
+
+travis_test_full:
 	@if [ -d ".venv" ] && [ -f "pyproject.toml" ]; then \
 		echo "Running tests in Poetry environment..."; \
 		poetry run pytest tests/test_*local*.py tests/test_*parse*.py tests/test*writer*.py tests/test_qc.py \
@@ -33,7 +58,8 @@ travis_test:
 		tests/unit/test_golr_search_query.py tests/unit/test_owlsim2_api.py \
 		tests/test_collections.py \
 		tests/test_gocamgen.py \
-		tests/test_gpi_isoform_replacement.py; \
+		tests/test_gpi_isoform_replacement.py \
+		tests/test_validate_cli.py; \
 	else \
 		pytest tests/test_*local*.py tests/test_*parse*.py tests/test*writer*.py tests/test_qc.py \
 		tests/test_rdfgen.py tests/test_phenosim_engine.py tests/test_ontol.py \
@@ -42,9 +68,11 @@ travis_test:
 		tests/unit/test_golr_search_query.py tests/unit/test_owlsim2_api.py \
 		tests/test_collections.py \
 		tests/test_gocamgen.py \
-		tests/test_gpi_isoform_replacement.py; \
+		tests/test_gpi_isoform_replacement.py \
+		tests/test_validate_cli.py; \
 	fi
 
+
 cleandist:
 	rm dist/* || true
 

bin/README.md

@@ -1,16 +1,26 @@
 See [command line docs](http://ontobio.readthedocs.io/en/latest/commandline.html#commandline) on ReadTheDocs
 
 To test validate.py "validate" command, the command that produces the final GPADs in the pipeline via the "mega make" 
-(aka: "produces GAFs, GPADs, ttl" stage), on a particular source:
+(aka: "produces GAFs, GPADs, ttl" stage), on a particular source, run:
+```bash
+make test_travis_full
+```
+
+This makefile target will run the full validate.produce command using goa_cow, mgi, zfin, and goa_chicken sources,
+producing GPAD, GAF files in the groups subdirectory and then do a check of the content of these products.  These
+tests only run manually, not via CI because they take minutes to run.
 
+alternatively, you can run the following commands to test the validate.produce command on a particular source, locally:
+
+```bash
 Note: snapshot below in the URL can be changed to any pipeline branch; its listed here for ease of cp/paste.
 ```bash
 poetry install
 poetry run validate produce -m ../go-site/metadata --gpad -t . -o go-basic.json --base-download-url "http://skyhook.berkeleybop.org/snapshot/" --only-dataset mgi MGI --gpad-gpi-output-version 2.0
 poetry run validate produce -m ../go-site/metadata --gpad -t . -o go-basic.json --base-download-url "http://skyhook.berkeleybop.org/snapshot/" --only-dataset goa_chicken goa --gpad-gpi-output-version 2.0
+poetry run validate produce -m ../go-site/metadata --gpad -t . -o go-basic.json --base-download-url "http://skyhook.berkeleybop.org/snapshot/" --only-dataset zfin ZFIN --gpad-gpi-output-version 2.0
 ```
 
-
 To test whether a GAF file is valid (passes all the GORules):
 ```bash
 poetry install

bin/validate.py

@@ -116,14 +116,20 @@ def download_a_dataset_source(group, dataset_metadata, target_dir, source_url, b
     return path
 
 
-def download_source_gafs(group_metadata, target_dir, exclusions=[], base_download_url=None, replace_existing_files=True,
+def download_source_gafs(group_metadata,
+                         target_dir,
+                         exclusions=[],
+                         base_download_url=None,
+                         replace_existing_files=True,
                          only_dataset=None):
     """
     This looks at a group metadata dictionary and downloads each GAF source that is not in the exclusions list.
     For each downloaded file, keep track of the path of the file. If the file is zipped, it will unzip it here.
     This function returns a list of tuples of the dataset dictionary mapped to the downloaded source path.
     """
-    # Grab all datasets in a group, excluding non-gaf, datasets that are explicitely excluded from an option, and excluding datasets with the `exclude key` set to true
+    # Grab all datasets in a group, excluding non-gaf, datasets that are explicitly excluded
+    # from an option, and excluding datasets with the `exclude key` set to true
+
     gaf_urls = []
     if only_dataset is None:
         gaf_urls = [(data, data["source"]) for data in group_metadata["datasets"] if
@@ -132,7 +138,7 @@ def download_source_gafs(group_metadata, target_dir, exclusions=[], base_downloa
         gaf_urls = [(data, data["source"]) for data in group_metadata["datasets"] if data["dataset"] == only_dataset]
     # List of dataset metadata to gaf download url
 
-    click.echo("Found {}".format(", ".join([kv[0]["dataset"] for kv in gaf_urls])))
+    logger.info("Found gaf_urls {}".format(", ".join([kv[0]["dataset"] for kv in gaf_urls])))
     downloaded_paths = []
     for dataset_metadata, gaf_url in gaf_urls:
         dataset = dataset_metadata["dataset"]
@@ -148,7 +154,7 @@ def download_source_gafs(group_metadata, target_dir, exclusions=[], base_downloa
             # otherwise file is coming in uncompressed. But we want to make sure
             # to zip up the original source also
             tools.zipup(path)
-
+        logger.info("Downloaded {}".format(path))
         downloaded_paths.append((dataset_metadata, path))
 
     return downloaded_paths
@@ -645,6 +651,8 @@ def produce(ctx, group, metadata_dir, gpad, gpad_gpi_output_version, ttl, target
     click.echo("Products will go in {}".format(absolute_target))
     absolute_metadata = os.path.abspath(metadata_dir)
 
+    print("group", group)
+    print("dataset", )
     group_metadata = metadata.dataset_metadata_file(absolute_metadata, group)
     click.echo("Loading ontology: {}...".format(ontology))
     ontology_graph = OntologyFactory().create(ontology, ignore_cache=True)
@@ -654,6 +662,7 @@ def produce(ctx, group, metadata_dir, gpad, gpad_gpi_output_version, ttl, target
                                                   replace_existing_files=not skip_existing_files,
                                                   only_dataset=only_dataset)
 
+    click.echo("Downloaded GAF sources: {}".format(downloaded_gaf_sources))
     # extract the titles for the go rules, this is a dictionary comprehension
     rule_metadata = metadata.yamldown_lookup(os.path.join(absolute_metadata, "rules"))
     goref_metadata = metadata.yamldown_lookup(os.path.join(absolute_metadata, "gorefs"))