def json_export_type_converter(obj):
if isinstance(obj, deque):
return list(obj)
elif isinstance(obj, AJ):
return obj.asJson()
elif isinstance(obj, ProtcurExpression):
return obj.json()
elif isinstance(obj, PurePath):
return obj.as_posix()
elif isinstance(obj, Quantity):
return obj.json()
elif isinstance(obj, Measurement):
return obj.json()
elif isinstance(obj, oq.OntTerm):
return obj.iri
#return obj.asDict() # FIXME need a no network/scigraph version
elif isinstance(obj, idlib.Stream) and hasattr(obj, '_id_class'):
if obj._id_class is str:
return obj.identifier
else:
return json_export_type_converter(obj.identifier)
#return obj.asDict() # FIXME need a no network/scigraph version
elif isinstance(obj, datetime):
return isoformat(obj)
elif isinstance(obj, time):
return isoformat(obj)
elif isinstance(obj, BaseException):
# FIXME hunt down where these are sneeking in from
return repr(obj)
def triples_header(self):
ontid = self.ontid
nowish = utcnowtz()
epoch = nowish.timestamp()
iso = isoformat(nowish)
ver_ontid = rdflib.URIRef(ontid + f'/version/{epoch}/{self.id}')
sparc_methods = rdflib.URIRef('https://raw.githubusercontent.com/SciCrunch/'
'NIF-Ontology/sparc/ttl/sparc-methods.ttl')
sparc_mis_helper = rdflib.URIRef('https://raw.githubusercontent.com/SciCrunch/'
'NIF-Ontology/sparc/ttl/sparc-mis-helper.ttl')
pos = (
(rdf.type, owl.Ontology),
(owl.versionIRI, ver_ontid),
(owl.versionInfo, rdflib.Literal(iso)),
(isAbout, rdflib.URIRef(self.uri_api)),
(TEMP.hasUriHuman, rdflib.URIRef(self.uri_human)),
(rdfs.label, rdflib.Literal(f'{self.folder_name} curation export graph')),
(rdfs.comment, self.header_graph_description),
(owl.imports, sparc_methods),
(owl.imports, sparc_mis_helper),
(TEMP.TimestampExportStart, rdflib.Literal(self.timestamp_export_start)),
)
for p, o in pos:
yield ontid, p, o
def render_manifest(rows):
# FIXME checksum cypher
return [[
filename.as_posix(),
isoformat(timestamp),
desc,
ft,
at,
checksum.hex() if checksum else checksum,
] for filename, timestamp, desc, ft, at, checksum in rows]
def _datetime(self, value):
if not isinstance(value, datetime):
raise TypeError(f'{type(value)} is not a datetime for {value}')
has_tz = value.tzinfo is not None and value.tzinfo.utcoffset(
None) is not None
value = isoformat(value)
if not has_tz:
log.warning('why do you have a timestamp without a timezone ;_;')
return value
def subject_id(self, v, species=None): # TODO species for human/animal
if not isinstance(v, str):
#loge.critical('darn it max normlize your ids!') # now caught by the schemas
if isinstance(v, datetime):
# XXX datetime is a REALLY bad case because there is not a
# canonical representation
v = isoformat(v)
else:
v = str(v)
v = quote(v, safe=tuple())
s = rdflib.URIRef(self.dsid + '/subjects/' + v)
return s
def triples_header(self):
ontid = self.ontid
nowish = utcnowtz()
epoch = nowish.timestamp()
iso = isoformat(nowish)
ver_ontid = rdflib.URIRef(ontid + f'/version/{epoch}/{self.id}')
pos = (
(rdf.type, owl.Ontology),
(owl.versionIRI, ver_ontid),
(owl.versionInfo, rdflib.Literal(iso)),
(rdfs.comment, self.header_graph_description),
(TEMP.TimestampExportStart, rdflib.Literal(self.timestamp_export_start)),
)
for p, o in pos:
yield ontid, p, o
def _rowcellify(cell_value):
if isinstance(cell_value, idlib.Stream):
try:
return cell_value.asUri()
except AttributeError as e:
breakpoint()
elif (isinstance(cell_value, date)
or isinstance(cell_value, datetime)):
# the default json encoder doesn't know what to do with these
return isoformat(cell_value)
elif isinstance(cell_value, str):
if cell_value.lower() == 'x':
return True
return cell_value
def default(self, obj):
if isinstance(obj, deque):
return list(obj)
elif isinstance(obj, ProtcurExpression):
return obj.json()
elif isinstance(obj, Path):
return obj.as_posix()
elif isinstance(obj, Quantity):
return obj.json()
elif isinstance(obj, idlib.Stream):
return obj.identifier
elif isinstance(obj, datetime):
return isoformat(obj)
#else:
#log.critical(f'{type(obj)} -> {obj}')
# Let the base class default method raise the TypeError
return json.JSONEncoder.default(self, obj)
def makeKwargs(repo_path, filepath):
path = repo_path / filepath
kwargs = {}
kwargs['title'] = filepath
kwargs['authors'] = sorted(name.strip() for name in set(
repo_path.repo.git.log(
['--follow', '--pretty=format:%an%x09', filepath]).split('\n')))
kwargs['date'] = isoformat(path.latest_commit().authored_datetime)
repo_url = Path(
next(
next(r for r in repo_path.repo.remotes
if r.name == 'origin').urls))
kwargs['org'] = repo_url.parent.name
kwargs['repo'] = repo_url.stem
kwargs[
'branch'] = 'master' # TODO figure out how to get the default branch on the remote
kwargs['crumbs'] = _crumbs
return kwargs
def triples_header(self):
# TODO TODO
ontid = self.ontid
nowish = utcnowtz() # FIXME pass in so we can align all times per export??
epoch = nowish.timestamp()
iso = isoformat(nowish)
ver_ontid = rdflib.URIRef(ontid + f'/version/{epoch}/{self.id}')
#sparc_methods = rdflib.URIRef('https://raw.githubusercontent.com/SciCrunch/'
#'NIF-Ontology/sparc/ttl/sparc-methods.ttl')
pos = (
(a, owl.Ontology),
(owl.versionIRI, ver_ontid),
(owl.versionInfo, rdflib.Literal(iso)),
#(isAbout, rdflib.URIRef(self.uri_api)),
#(TEMP.hasHumanUri, rdflib.URIRef(self.uri_human)),
(rdfs.label, rdflib.Literal(f'TODO export graph')),
#(rdfs.comment, self.header_graph_description),
#(owl.imports, sparc_methods),
)
for p, o in pos:
yield ontid, p, o
def START_TIMESTAMP_LOCAL(self):
return isoformat(self._start_time_local)
def START_TIMESTAMP(self):
return isoformat(self._start_time)
def test_isoformat(self):
now = datetime.now()
n = utils.isoformat(now)
today = date.today()
t = utils.isoformat(today)
