def process_paper(self):
# filename = "{0}/data/manual/full_iswc_paper_pdf.csv".format(
filename = "{0}/data/manual/iswc-publication-paper.csv".format(self.global_config["home"])
counter_paper = MyCounter()
with open(filename) as f:
csvreader = UnicodeReader(f)
headers = csvreader.next()
for row in csvreader:
if len(row) != len(headers):
# print "skipping mismatch row %s" % row
continue
entry = dict(zip(headers, row))
if entry["year"] != self.local_config["year"]:
# skip mismatched year
continue
if len(entry["title"]) == 0:
print "skipping empty title row %s" % entry
continue
if len(entry["proceedings_uri"]) == 0:
print "skipping empty proceedings row %s" % entry
continue
counter_paper.inc(entry["proceedings_uri"])
id_paper = counter_paper.data[entry["proceedings_uri"]]
uri_paper = "%s/paper-%02d" % (entry["proceedings_uri"], id_paper)
uri_paper_author_list = "%s/paper-%02d/author_list" % (entry["proceedings_uri"], id_paper)
# print json.dumps(entry, indent=4)
# print uri_paper
res_proceedings = URIRef(entry["proceedings_uri"])
res_paper = URIRef(uri_paper)
self.graph.add((res_paper, RDF.type, SWRC.InProceedings))
# part-of proceedings
self.graph.add((res_paper, SWC.isPartOf, res_proceedings))
self.graph.add((res_proceedings, SWC.hasPart, res_paper))
# author
self.graph.add((res_paper, SWRC.listAuthor, Literal(entry["author"])))
list_res_author = []
for author in entry["author"].split(","):
res_author = self.create_named_entity(self.get_namespace(DataIswc.PREFIX_PERSON), author)
self.graph.add((res_author, RDF.type, FOAF.Person))
list_res_author.append(res_author)
self.graph.add((res_paper, SWRC.author, res_author))
self.graph.add((res_paper, FOAF.maker, res_author))
self.graph.add((res_author, FOAF.made, res_paper))
res_paper_author_list = self.create_container(list_res_author, RDF.Seq, uri_paper_author_list)
self.graph.add((res_paper, BIBO.authorList, res_paper_author_list))
# simple properties
self.create_triple_complex(
res_paper,
[
"abstract",
"keywords",
"year",
"pages",
"title",
"category",
"link_open_access",
"link_publisher",
],
entry,
)
# cache
self.map_name_res[entry["title"]] = res_paper
self.data[key]=0
self.data[key] += cnt
def list(self, min_count=0):
ret = {}
for k,v in self.data:
if v >= min_count:
ret[k]=v
return ret
with open(filename) as f:
csvreader = UnicodeReader(f)
headers = csvreader.next()
counter = MyCounter()
for row in csvreader:
entry = dict(zip(headers, row))
entry["proceedings_uri"] = entry["proceedings_uri"].replace("demos/proceedings","demos-proceedings")
dir_id = entry["year"]+"-"+ os.path.basename(entry["proceedings_uri"])
counter.inc(dir_id)
counter.inc(dir_id+"-downloaded", 0)
state = {"paper_uri", entry["paper_uri"]}
if len(entry["link_open_access"])>0:
if entry["link_open_access"].lower().endswith(".html") or entry["link_open_access"].lower().endswith(".htm"):
print "skip ", entry["link_open_access"]
continue
def process_event(self):
filename = "{0}/data/manual/{1}-event.csv".format(self.global_config["home"], self.local_config["tag"])
counter_event = MyCounter()
with open(filename) as f:
csvreader = UnicodeReader(f)
headers = csvreader.next()
for row in csvreader:
if len(row) != len(headers):
# print "skipping mismatch row %s" % row
continue
entry = dict(zip(headers, row))
if len(entry["label"]) == 0:
# print "skipping empty label row %s" % entry
continue
if len(entry["event_type"]) == 0:
# print "skipping empty event_type row %s" % entry
continue
if entry["event_uri"].startswith("#"):
# print "skipping empty commented row %s" % entry
continue
# set default super event
if len(entry["super_event_uri"]) == 0:
entry["super_event_uri"] = "[ME]"
uri_super_event = self.expand_uri(entry["super_event_uri"])
res_super_event = URIRef(uri_super_event)
if len(entry["event_uri"]) == 0:
counter_event.inc(uri_super_event)
entry["event_uri"] = "%s/event-%02d" % (uri_super_event, counter_event.data[uri_super_event])
uri_event = self.expand_uri(entry["event_uri"])
res_event = URIRef(uri_event)
# event type
self.graph.add((res_event, RDF.type, SWC[entry["event_type"]]))
# super event
self.graph.add((res_event, SWC.isSubEventOf, res_super_event))
self.graph.add((res_super_event, SWC.isSuperEventOf, res_event))
# simple properties
self.create_triple_complex(
res_event,
[
"label",
"acronym",
"abstract",
"order_in_super_event",
"start",
"end",
"tzid",
"room",
"address",
"homepage",
"link_document",
"logo",
],
entry,
)
# linking paper event
if "TalkEvent" == entry["event_type"]:
if entry["label"] in self.map_name_res:
res_paper = self.map_name_res[entry["label"]]
self.graph.add((res_event, SWC.hasRelatedDocument, res_paper))
self.graph.add((res_paper, SWC.relatedToEvent, res_event))
else:
print "missing paper link " + entry["label"]
sys.exit(0)
# role -chair
for role in ["Chair", "Presenter"]:
role_lower = role.lower()
if len(entry[role_lower + "_person"]) > 0:
for name in entry[role_lower + "_person"].split(","):
if len(name) == 0:
continue
res_person = self.create_named_entity(self.get_namespace(DataIswc.PREFIX_PERSON), name)
self.create_role_to_event(
uri_event, "swc:" + role, entry[role_lower + "_label"], res_person
)