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 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 reference(prefix: str, identifier: str):
"""Serve the a Bioregistry reference page."""
return render_template(
'reference.html',
prefix=prefix,
name=bioregistry.get_name(prefix),
identifier=identifier,
providers=_get_resource_providers(prefix, identifier),
)
def warnings():
"""Make warnings list."""
miriam_pattern_wrong = [
dict(
prefix=prefix,
name=bioregistry.get_name(prefix),
correct=entry['pattern'],
miriam=entry['miriam']['pattern'],
) for prefix, entry in items
if 'miriam' in entry and 'pattern' in entry
and entry['pattern'] != entry['miriam']['pattern']
]
miriam_embedding_rewrites = [
dict(
prefix=prefix,
name=bioregistry.get_name(prefix),
pattern=bioregistry.get_pattern(prefix),
correct=entry['namespace.embedded'],
miriam=entry['miriam']['namespaceEmbeddedInLui'],
) for prefix, entry in items if 'namespace.embedded' in entry
]
# When are namespace rewrites required?
miriam_prefix_rewrites = [
dict(
prefix=prefix,
name=bioregistry.get_name(prefix),
pattern=bioregistry.get_pattern(prefix),
correct=entry['namespace.rewrite'],
) for prefix, entry in items if 'namespace.rewrite' in entry
]
with open(os.path.join(DOCS_DATA, 'warnings.yml'), 'w') as file:
yaml.safe_dump(
{
'wrong_patterns': miriam_pattern_wrong,
'embedding_rewrites': miriam_embedding_rewrites,
'prefix_rewrites': miriam_prefix_rewrites,
},
file,
)
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 test_name_expansions(self):
"""Test that default names are not capital acronyms."""
for prefix in bioregistry.read_registry():
if bioregistry.is_deprecated(prefix):
continue
entry = bioregistry.get(prefix)
if 'name' in entry:
continue
name = bioregistry.get_name(prefix)
if prefix == name.lower() and name.upper() == name:
with self.subTest(prefix=prefix):
self.fail(msg=f'{prefix} acronym ({name}) is not expanded')
if '.' in prefix and prefix.split('.')[0] == name.lower():
with self.subTest(prefix=prefix):
self.fail(msg=f'{prefix} acronym ({name}) is not expanded')
def make_resource_node(cx: NiceCXBuilder, prefix: str) -> int:
"""Generate a CX node for a resource."""
node = cx.add_node(
name=bioregistry.get_name(prefix),
represents=f'bioregistry.resource:{prefix}',
)
homepage = bioregistry.get_homepage(prefix)
if homepage:
cx.add_node_attribute(node, 'homepage', homepage)
description = bioregistry.get_description(prefix)
if description:
cx.add_node_attribute(node, 'description', description)
pattern = bioregistry.get_pattern(prefix)
if pattern:
cx.add_node_attribute(node, 'pattern', pattern)
# TODO add more
return node
def test_no_redundant_acronym(self):
"""Test that there is no redundant acronym in the name.
For example, "Amazon Standard Identification Number (ASIN)" is a problematic
name for prefix "asin".
"""
for prefix in bioregistry.read_registry():
if bioregistry.is_deprecated(prefix):
continue
entry = bioregistry.get(prefix)
if 'name' in entry:
continue
name = bioregistry.get_name(prefix)
try:
_, rest = name.rstrip(')').rsplit('(', 1)
except ValueError:
continue
if rest.lower() == prefix.lower():
with self.subTest(prefix=prefix):
self.fail(msg=f'{prefix} has redundany acronym in name "{name}"')
def namespaces_table_df(graph: BELGraph,
examples: bool = True) -> pd.DataFrame:
"""Create a dataframe describing the namespaces in the graph."""
namespace_mapping = multidict((node.namespace, node) for node in graph
if isinstance(node, BaseConcept))
namespace_c = count_namespaces(graph)
if not examples:
return pd.DataFrame(namespace_c.most_common(),
columns=['Namespace', 'Count'])
return pd.DataFrame(
[
(
prefix,
bioregistry.get_name(prefix),
count,
random.choice(namespace_mapping[prefix])
if prefix in namespace_mapping else '', # noqa:S311
) for prefix, count in namespace_c.most_common()
],
columns=['Prefix', 'Name', 'Count', 'Example'],
)
def __post_init__(self):
"""Run post-init checks."""
if self.ontology != bioregistry.normalize_prefix(self.ontology):
raise BioregistryError(self.ontology)
# The type ignores are because of the hack where we override the
# class variables in the instance
if self.name is None:
self.name = bioregistry.get_name(self.ontology) # type:ignore
if not self.data_version:
if self.static_version:
self.data_version = self.static_version
else:
self.data_version = self._get_version()
if not self.dynamic_version:
if self.data_version is None:
raise ValueError(f"{self.ontology} is missing data_version")
elif "/" in self.data_version:
raise ValueError(
f"{self.ontology} has a slash in version: {self.data_version}"
)
if self.auto_generated_by is None:
self.auto_generated_by = f"bio2obo:{self.ontology}" # type:ignore
def _g(predicate):
return [{
'prefix': bioregistry_id,
'name': bioregistry.get_name(bioregistry_id),
} for bioregistry_id, bioregistry_entry in items
if predicate(bioregistry_id, bioregistry_entry)]
