def serialize(graph, outpath):
def prefix_cleanup(ps, graph):
if 'parcellation/' in outpath:
nsl = {p: n for p, n in graph.namespaces()}
if '' in nsl:
ps[''] = nsl['']
elif 'hbaslim' in outpath:
ps['HBA'] = nsl['HBA']
elif 'mbaslim' in outpath:
ps['MBA'] = nsl['MBA']
elif 'cocomac' in outpath:
ps['cocomac'] = nsl['cocomac']
# special cases for NIFORG, NIFINV, NIFRET where there identifiers in
# annotation properties that share the prefix, so we warn incase
# at some point in the future for some reason want them again...
if outpath == 'NIF-Organism.ttl':
print('WARNING: special case for NIFORG')
ps.pop('NIFORG')
elif outpath == 'NIF-Investigation.ttl':
print('WARNING: special case for NIFINV')
ps.pop('NIFINV')
elif outpath == 'unused/NIF-Retired.ttl':
print('WARNING: special case for NIFRET')
ps.pop('NIFGA')
pc = prefix_cleanup if isinstance(outpath, str) else lambda a, b: None
graph = cull_prefixes(graph, cleanup=pc)
out = graph.g.serialize(format='nifttl', gen_prefix=bool(PREFIXES))
if not isinstance(outpath, str): # FIXME not a good test that it is stdout
outpath.buffer.write(out)
else:
with open(outpath, 'wb') as f:
f.write(out)
def graph(self, graph=Graph()):
[graph.add(t) for t in self.triples]
self.post_graph(graph)
out_mgraph = cull_prefixes(graph,
prefixes={
**dict(workflow=workflow,
RRIDCUR=RRIDCUR),
**uPREFIXES
})
for c, i in out_mgraph.g.namespaces(): # will add but not take away
graph.bind(c, i)
return out_mgraph.g
def main():
args = docopt(__doc__, version='0.1')
outloc = args['--output-location']
filename = os.path.splitext(os.path.basename(args['<file>']))[0]
mgraph = makeGraph(filename, prefixes=uPREFIXES, writeloc=outloc)
if args['workflow']:
w = WorkflowMapping(args['<file>'])
[mgraph.g.add(t) for t in w.triples]
w.post_graph(mgraph.g)
elif args['paper']:
w = PaperIdMapping(args['<file>'])
[mgraph.g.add(t) for t in w.triples]
w.post_graph(mgraph.g)
elif args['methods']:
parser = etree.XMLParser(remove_blank_text=True)
e = etree.parse(args['<file>'], parser)
#graph = e.find(abv['graph'])
nodes = xpath(e, '//' + abv['node'])
edges = xpath(e, '//' + abv['edge'])
node_dict = {}
for node in nodes: # slow but who cares
id_ = node.get('id')
#label = xpath(node, '//'+abv['NodeLabel'])[0].text
idstr = '[@id="%s"]//' % id_
label = xpath(e, '//' + abv['node'] + idstr +
abv['NodeLabel'])[0].text
targets = []
node_dict['FIXME:' + id_] = label, targets
edge_dict = {}
edge_types = set()
for edge in edges:
id_ = edge.get('id')
#print(id_)
idstr = '[@id="%s"]//' % id_
source = 'FIXME:' + edge.get('source')
target = 'FIXME:' + edge.get('target')
out = xpath(edge, '//' + abv['edge'] + idstr + abv['EdgeLabel'])
edge_type = out[0].text if out else None
#print(edge_type)
edge_dict[id_] = source, target, edge_replace(edge_type)
edge_types.add(edge_type)
for et in set(edge_to_ttl.values()):
if et != 'SKIP':
mgraph.add_op(et)
for eid, (source, target, edge_type) in edge_dict.items():
node_dict[source][1].append((edge_type, target))
#print(source, edge_type, target)
if edge_type == 'SKIP':
mgraph.add_trip(source, 'rdf:type', 'owl:Class')
elif edge_type is not None:
mgraph.add_class(source)
mgraph.add_class(target)
try:
if edge_type == 'rdfs:subClassOf':
mgraph.add_trip(source, edge_type, target)
else:
mgraph.add_hierarchy(target, edge_type, source)
except ValueError as e:
raise ValueError(f'{source} {edge_type} {target}') from e
label = node_dict[source][0]
if '(id' in label:
label, rest = label.split('(id')
id_, rest = rest.split(')', 1)
mgraph.add_trip(source, 'FIXME:REPLACEID', id_)
label = label.strip() + rest.strip()
if '(syns' in label:
label, rest = label.split('(syns')
syns, rest = rest.split(')', 1)
if ',' in syns:
syns = [
mgraph.add_trip(source, 'NIFRID:synonym', s.strip())
for s in syns.split(',') if s
] #FIXME
else:
syns = [
mgraph.add_trip(source, 'NIFRID:synonym', s)
for s in syns.split(' ') if s
] #FIXME
label = label.strip() + rest.strip()
if '(note' in label:
while '(note' in label:
label, rest = label.split('(note', 1)
note, rest = rest.split(')', 1)
mgraph.add_trip(source, 'rdfs:comment', note)
label = label.strip() + rest.strip()
if '(def' in label:
label, rest = label.split('(def')
def_, rest = rest.split(')', 1)
def_ = def_.replace('\n', ' ')
mgraph.add_trip(source, 'NIFRID:definition', def_.strip())
label = label.strip() + rest
if '#FIXME' in label:
label, note = label.split('#FIXME')
label = label.replace('\n', '').strip()
note = note.replace('\n', ' ').strip()
mgraph.add_trip(source, 'rdfs:comment', note)
if args['methods']:
clabel = label.capitalize()
else:
clabel = label
mgraph.add_trip(source, 'rdfs:label', clabel)
Query = namedtuple('Query',
['root', 'relationshipType', 'direction', 'depth'])
json = mgraph.make_scigraph_json('rdfs:subClassOf', direct=True)
t, te = creatTree(*Query('FIXME:n0', 'rdfs:subClassOf', 'INCOMING',
20),
json=json) # methods
t, te = creatTree(*Query('FIXME:n236', 'rdfs:subClassOf', 'INCOMING',
20),
json=json) # techniques
print(t)
with open(os.path.join(outloc, filename + '.txt'), 'wt') as f:
f.write(str(t))
with open(os.path.join(outloc, filename + '.html'), 'wt') as f:
f.write(te.html)
out_graph = cull_prefixes(mgraph.g,
prefixes={
**dict(workflow=workflow, RRIDCUR=RRIDCUR),
**uPREFIXES
})
out_graph.filename = mgraph.filename
out_graph.write()