def _process_html(self):
path = '//a[contains(./@href, ".pdf")]'
for link in self.html.xpath(path):
case = {}
if len(link.text.strip()) == 0:
# There seems to be some cases that are pdf-linked more than once, with the second link blank
continue
# text of case info apart from name
utext = link.getparent().text_content()[len(link.text):]
text = utext.encode("ascii", "replace")
# date
date = ""
date_match = re.search(
"(\([a-zA-Z]+[^\w]{0,3}\d\d?. \d{4}\))|(\(\d{4}\))", utext)
if date_match is not None:
date = date_match.group(0).strip("()")
else:
raise ParsingException("Date regexp failed on text" + text)
# precedent
header = link.getparent().getparent().getprevious().text
precedential = 'Unpublished'
if 'Precedential' in header:
precedential = 'Published'
# docket numbers
docket_match = re.search(
"(\d{2,4}-\d{4,6})|(\d{3},[0-9|A-Z]{3})|(\dx{2},x{3})", utext)
docket = ""
paper_match = re.search("Paper \d+", utext)
if docket_match is not None:
docket = docket_match.group(0)
if paper_match is not None:
docket = "{} ({})".format(docket, paper_match.group(0))
else:
raise ParsingException("Docket No. Regexp failed on text" +
text)
# summary
# use unicode, there are a lot of section codes
summary_match = re.search("\[.*\]", utext)
summary = ""
if summary_match is not None:
summary = summary_match.group(0).strip("[]")
else:
ParsingException(
"Summary regexp failed on text {}".format(text))
# We sometimes have repeats on this page, so dict
self.cases[docket] = {
"name": link.text,
"url": link.get('href'),
"date": convert_date_string(date),
"precedential": precedential,
"docket": docket,
"summary": summary
}
def fetch_docket_by_pacer_case_id(
session, court_id, pacer_case_id, fq,
):
"""Download the docket from PACER and merge it into CL
:param session: A PacerSession object to work with
:param court_id: The CL ID of the court
:param pacer_case_id: The pacer_case_id of the docket, if known
:param fq: The PacerFetchQueue object
:return: a dict with information about the docket and the new data
"""
report = DocketReport(map_cl_to_pacer_id(court_id), session)
report.query(pacer_case_id, **get_fq_docket_kwargs(fq))
docket_data = report.data
if not docket_data:
raise ParsingException("No data found in docket report.")
if fq.docket_id:
d = Docket.objects.get(pk=fq.docket_id)
else:
d, count = find_docket_object(
court_id, pacer_case_id, docket_data["docket_number"]
)
if count > 1:
d = d.earliest("date_created")
rds_created, content_updated = merge_pacer_docket_into_cl_docket(
d, pacer_case_id, docket_data, report, appellate=False,
)
return {
"docket_pk": d.pk,
"content_updated": bool(rds_created or content_updated),
}
def get_int_from_details(self, node):
s = self.case_details.xpath('%s/text()' % node)[0].strip()
try:
return int(s)
except ValueError:
# Can't parse string to int
logger.debug("Couldn't get int for node %s" % node)
raise ParsingException("Cannot extract int for node %s" % node)
def get_court(self):
"""Extract the court from the XML and return it as a Court object"""
court_str = self.case_details.xpath('court/text()')[0].strip()
try:
c = Court.objects.get(pk=map_pacer_to_cl_id(court_str))
except Court.DoesNotExist:
raise ParsingException("Unable to identify court: %s" % court_str)
else:
return c