def __init__(self, api_url):
"""
"""
self.api_url = api_url
if len(self.api_url) > 0 and self.api_url[-1] != "/":
self.api_url += "/"
self.reader = ParsCitReader()
def __init__(self, api_url):
"""
"""
self.api_url = api_url
if len(self.api_url) > 0 and self.api_url[-1] !="/":
self.api_url += "/"
self.reader=ParsCitReader()
class ParsCitClient:
"""
Connects to ParsCit web service, sends documents and converts the
response to metadata.
"""
def __init__(self, api_url):
"""
"""
self.api_url = api_url
if len(self.api_url) > 0 and self.api_url[-1] !="/":
self.api_url += "/"
self.reader=ParsCitReader()
def tagFullDocument(self, doc_text):
"""
"""
raise NotImplementedError
def extractReferenceList(self, ref_list, format="raw"):
"""
Main function: interface with ParsCit.
Args:
ref_list: list of strings to be processed. Each string can
contain multiple references
"""
res=[]
if isinstance(ref_list, list):
ref_list ="\n\n".join(ref_list)
# Hack so ParsCit will actually recognize the references
if format=="raw":
ref_list=u"References\n\n%s" % ref_list
else:
ref_list=u"<references>%s</references>" % ref_list
data={"text":ref_list, "format":format}
r=requests.post(self.api_url+"extract_citations/", json=data)
if r.status_code != 200:
# TODO specialized exceptions
print(ref_list)
raise ValueError("ParsCit exception")
json_data=json.loads(r.content)
res=self.reader.parseParsCitXML(json_data["parsed_xml"])
return res
class ParsCitClient:
"""
Connects to ParsCit web service, sends documents and converts the
response to metadata.
"""
def __init__(self, api_url):
"""
"""
self.api_url = api_url
if len(self.api_url) > 0 and self.api_url[-1] != "/":
self.api_url += "/"
self.reader = ParsCitReader()
def tagFullDocument(self, doc_text):
"""
"""
raise NotImplementedError
def extractReferenceList(self, ref_list, format="raw"):
"""
Main function: interface with ParsCit.
Args:
ref_list: list of strings to be processed. Each string can
contain multiple references
"""
res = []
if isinstance(ref_list, list):
ref_list = "\n\n".join(ref_list)
# Hack so ParsCit will actually recognize the references
if format == "raw":
ref_list = u"References\n\n%s" % ref_list
else:
ref_list = u"<references>%s</references>" % ref_list
data = {"text": ref_list, "format": format}
r = requests.post(self.api_url + "extract_citations/", json=data)
if r.status_code != 200:
# TODO specialized exceptions
print(ref_list)
raise Exception
json_data = json.loads(r.content)
res = self.reader.parseParsCitXML(json_data["parsed_xml"])
return res
