def summary_df(self) -> pd.DataFrame:
"""Generate a summary dataframe."""
summary_names = self.summarize_names()
summary_alts = self.summarize_alts(
) if self.summarize_alts is not None else {}
summary_defs = (self.summarize_definitions()
if self.summarize_definitions is not None else {})
return pd.DataFrame(
[(
prefix,
bioregistry.get_name(prefix),
bioregistry.get_homepage(prefix),
bioregistry.get_example(prefix),
bioregistry.get_link(prefix, bioregistry.get_example(prefix)),
names_count,
summary_alts.get(prefix, 0),
summary_defs.get(prefix, 0),
) for prefix, names_count in summary_names.items()],
columns=[
"prefix",
"name",
"homepage",
"example",
"link",
"names",
"alts",
"defs",
],
)
def summary_df(self) -> pd.DataFrame:
"""Generate a summary dataframe."""
summary_names = self.summarize_names()
summary_alts = self.summarize_alts(
) if self.summarize_alts is not None else {}
summary_defs = self.summarize_definitions(
) if self.summarize_definitions is not None else {}
return pd.DataFrame(
[(
prefix,
bioregistry.get_name(prefix),
bioregistry.get_homepage(prefix),
bioregistry.get_example(prefix),
bioregistry.get_link(prefix, bioregistry.get_example(prefix)),
names_count,
summary_alts.get(prefix, 0),
summary_defs.get(prefix, 0),
) for prefix, names_count in summary_names.items()],
columns=[
'prefix',
'name',
'homepage',
'example',
'link',
'names',
'alts',
'defs',
],
)
def metaresource(metaprefix: str):
"""Serve the a Bioregistry registry page."""
entry = bioregistry.get_registry(metaprefix)
if entry is None:
abort(404, f'Invalid metaprefix: {metaprefix}')
example_prefix = entry.get('example')
example_identifier = bioregistry.get_example(
example_prefix) if example_prefix else None
return render_template(
'metaresource.html',
metaprefix=metaprefix,
name=bioregistry.get_registry_name(metaprefix),
description=bioregistry.get_registry_description(metaprefix),
homepage=bioregistry.get_registry_homepage(metaprefix),
download=entry.get('download'),
formatter=entry.get('formatter'),
example_prefix=example_prefix,
example_prefix_url=bioregistry.get_registry_url(
metaprefix, example_prefix) if example_prefix else None,
example_identifier=example_identifier,
example_curie_url=(bioregistry.get_registry_resolve_url(
metaprefix, example_prefix, example_identifier)
if example_prefix and example_identifier else None),
entry=entry,
)
def resource(prefix: str):
"""Serve the a Bioregistry entry page."""
prefix = _normalize_prefix_or_404(prefix, '.' + resource.__name__)
if not isinstance(prefix, str):
return prefix
example = bioregistry.get_example(prefix)
return render_template(
'resource.html',
prefix=prefix,
name=bioregistry.get_name(prefix),
example=example,
mappings=_get_resource_mapping_rows(prefix),
synonyms=bioregistry.get_synonyms(prefix),
homepage=bioregistry.get_homepage(prefix),
pattern=bioregistry.get_pattern(prefix),
version=bioregistry.get_version(prefix),
has_terms=bioregistry.has_terms(prefix),
obo_download=bioregistry.get_obo_download(prefix),
owl_download=bioregistry.get_owl_download(prefix),
namespace_in_lui=bioregistry.namespace_in_lui(prefix),
deprecated=bioregistry.is_deprecated(prefix),
contact=bioregistry.get_email(prefix),
banana=bioregistry.get_banana(prefix),
description=bioregistry.get_description(prefix),
providers=None if example is None else _get_resource_providers(
prefix, example),
)
def test_url_auto(self):
"""Test formatting URLs."""
for prefix, entry in bioregistry.read_registry().items():
if prefix in IDOT_BROKEN:
continue
identifier = bioregistry.get_example(prefix)
if identifier is None:
continue
if ('example' not in entry and 'banana' not in entry
and 'pattern' not in entry):
continue
url = get_identifiers_org_url(prefix, identifier)
if url is None:
continue
print(prefix)
with self.subTest(prefix=prefix, identifier=identifier):
# The following tests don't work because the CURIE generation often throws away the prefix.
# miriam_prefix = bioregistry.get_identifiers_org_prefix(prefix)
# self.assertIsNotNone(miriam_prefix)
# self.assertTrue(
# url.startswith(f'https://identifiers.org/{miriam_prefix}:'),
# msg=f"bad prefix for {prefix}. Expected {miriam_prefix} in {url}",
# )
res = self.session.get(url, allow_redirects=False)
self.assertEqual(302,
res.status_code,
msg='\n' + dedent(f'''\
Prefix: {prefix}
Identifier: {identifier}
URL: {url}
Text: ''') + fill(res.text, 70, subsequent_indent=' '))
def test_examples(self):
"""Test that all entries have examples."""
for prefix, entry in self.registry.items():
if 'pattern' not in entry: # TODO remove this later
continue
with self.subTest(prefix=prefix):
msg = f'{prefix} is missing an example local identifier'
if 'ols' in entry:
msg += f'\nSee: https://www.ebi.ac.uk/ols/ontologies/{entry["ols"]["prefix"]}/terms'
self.assertIsNotNone(bioregistry.get_example(prefix), msg=msg)
def resources():
"""Serve the Bioregistry page."""
rows = [
dict(
prefix=prefix,
name=bioregistry.get_name(prefix),
example=bioregistry.get_example(prefix),
homepage=bioregistry.get_homepage(prefix),
pattern=bioregistry.get_pattern(prefix),
namespace_in_lui=bioregistry.namespace_in_lui(prefix),
banana=bioregistry.get_banana(prefix),
description=bioregistry.get_description(prefix),
) for prefix in bioregistry.read_registry()
]
return render_template('resources.html', rows=rows)
def main(registry):
"""Add examples to the bioregistry from Inspector Javert's Xref Database."""
missing = {key for key in registry if get_example(key) is None}
with gzip.open(MAPPINGS_DB_TSV_CACHE, 'rt') as file:
for line in tqdm(file):
source_ns, source_id, target_ns, target_id, _ = line.strip().split(
'\t')
if source_ns in missing:
registry[source_ns]['example'] = source_id
tqdm.write(f'added example {source_ns} {source_id}')
missing.remove(source_ns)
if target_ns in missing:
registry[target_ns]['example'] = target_id
tqdm.write(f'added example {target_ns} {target_id}')
missing.remove(target_ns)
return registry
def test_examples_pass_patterns(self):
"""Test that all examples pass the patterns."""
for prefix, entry in self.registry.items():
pattern = bioregistry.get_pattern_re(prefix)
example = bioregistry.get_example(prefix)
if pattern is None or example is None:
continue
if bioregistry.namespace_in_lui(prefix):
miriam_prefix = get_identifiers_org_prefix(prefix)
if entry.get('namespace.capitalized') or 'obofoundry' in entry:
miriam_prefix = miriam_prefix.upper()
example = f'{miriam_prefix}:{example}'
if bioregistry.validate(prefix, example):
continue
with self.subTest(prefix=prefix):
self.assertRegex(example, pattern)
def curation():
"""Make curation list."""
missing_wikidata_database = _g(lambda prefix, entry: entry.get(
'wikidata', {}).get('database') is None)
missing_pattern = _g(
lambda prefix, entry: bioregistry.get_pattern(prefix) is None)
missing_format_url = _g(
lambda prefix, entry: bioregistry.get_format(prefix) is None)
missing_example = _g(
lambda prefix, entry: bioregistry.get_example(prefix) is None)
with open(os.path.join(DOCS_DATA, 'curation.yml'), 'w') as file:
yaml.safe_dump(
{
'wikidata': missing_wikidata_database,
'pattern': missing_pattern,
'formatter': missing_format_url,
'example': missing_example,
},
file,
)
def test_banana(self):
"""Test that entries curated with a new banana are resolved properly."""
for prefix, entry in bioregistry.read_registry().items():
banana = entry.get('banana')
if banana is None:
continue
if prefix in IDOT_BROKEN:
continue # identifiers.org is broken for these prefixes
with self.subTest(
prefix=prefix,
banana=banana,
pattern=bioregistry.get_pattern(prefix),
):
identifier = bioregistry.get_example(prefix)
self.assertIsNotNone(identifier)
url = bioregistry.resolve_identifier.get_identifiers_org_url(
prefix, identifier)
res = self.session.get(url, allow_redirects=False)
self.assertEqual(302,
res.status_code,
msg=f'failed with URL: {url}')
def main(url: str, local: bool):
"""Test the API."""
url = url.rstrip('/')
if local:
url = 'http://localhost:5000'
click.echo(f'Testing resolution API on {url}')
failure = False
prefixes = tqdm(bioregistry.read_registry())
for prefix in prefixes:
identifier = bioregistry.get_example(prefix)
if identifier is None:
continue
prefixes.set_postfix({'prefix': prefix})
req_url = f'{url}/{prefix}:{identifier}'
res = requests.get(req_url, allow_redirects=False)
log = partial(_log, req_url=req_url)
if res.status_code == 302: # redirect
continue
elif res.status_code != 404:
text = res.text.splitlines()[3][len('<p>'):-len('</p>')]
log(f'HTTP {res.status_code}: {res.reason} {text}', fg='red')
elif not bioregistry.get_providers(prefix, identifier):
continue
elif '/' in identifier or SLASH_URL_ENCODED in identifier:
log('contains slash 🎩 🎸', fg='red')
elif not bioregistry.validate(prefix, identifier):
pattern = bioregistry.get_pattern(prefix)
if bioregistry.get_banana(prefix):
log(f'banana {pattern} 🍌', fg='red')
else:
log(f'invalid example does not match pattern {pattern}',
fg='red')
else:
log('404 unknown issue', fg='red')
failure = True
return sys.exit(1 if failure else 0)