def _parse_headers(self):
header_arr = self._get_next_fields()
summary_line = self._get_next_fields()
self.form_row = summary_line
self.headers = parse_header(header_arr, self.is_paper)
self.headers['filing_amended'] = None
self.headers['report_num'] = None
self.version = self.headers['fec_version']
try:
self.headers['form'] = clean_entry(summary_line[0])
self.headers['fec_id'] = clean_entry(summary_line[1])
except IndexError:
return False
# Amendment discovery.
# Identify if this is an amemndment to a filing.
# If so, identify which filing it amends.
form_last_char = self.headers['form'][-1].upper()
if form_last_char == 'A':
self.is_amendment = True
self.headers['is_amendment'] = self.is_amendment
if self.is_paper:
self.headers['filing_amended'] = None
else:
# Listing the original only works for electonic filings, of course!
print "Found amendment %s : %s " % (self.filing_number,
self.headers['report_id'])
amendment_match = re.search('^FEC\s*-\s*(\d+)',
self.headers['report_id'])
if amendment_match:
original = amendment_match.group(1)
self.headers['filing_amended'] = original
else:
raise Exception(
"Can't find original filing in amended report %s" %
(self.filing_number))
else:
self.is_amendment = False
self.headers['is_amendment'] = self.is_amendment
return True
def _parse_headers(self):
header_arr = self._get_next_fields()
summary_line = self._get_next_fields()
self.form_row = summary_line
self.headers = parse_header(header_arr, self.is_paper)
self.headers['filing_amended'] = None
self.headers['report_num'] = None
self.version = self.headers['fec_version']
try:
self.headers['form'] = clean_entry(summary_line[0])
self.headers['fec_id'] = clean_entry(summary_line[1])
except IndexError:
return False
# Amendment discovery.
# Identify if this is an amemndment to a filing.
# If so, identify which filing it amends.
form_last_char = self.headers['form'][-1].upper()
if form_last_char == 'A':
self.is_amendment = True
self.headers['is_amendment'] = self.is_amendment
if self.is_paper:
self.headers['filing_amended'] = None
else:
# Listing the original only works for electonic filings, of course!
print "Found amendment %s : %s " % (self.filing_number, self.headers['report_id'])
amendment_match = re.search('^FEC\s*-\s*(\d+)', self.headers['report_id'])
if amendment_match:
original = amendment_match.group(1)
self.headers['filing_amended'] = original
else:
raise Exception("Can't find original filing in amended report %s" % (self.filing_number))
else:
self.is_amendment = False
self.headers['is_amendment'] = self.is_amendment
return True
def parse_header(header_array, is_paper=False):
""" Decides which version of the headers to use."""
if not is_paper:
version = clean_entry(header_array[2])
if old_eheaders_re.match(version):
headers_list = old_eheaders
elif new_eheaders_re.match(version):
headers_list = new_eheaders
else:
raise UnknownHeaderError(
"Couldn't find parser for electronic version %s" % (version))
else:
version = clean_entry(header_array[1])
if paper_headers_v1_re.match(version):
headers_list = paper_headers_v1
elif paper_headers_v2_2_re.match(version):
headers_list = paper_headers_v2_2
elif paper_headers_v2_6_re.match(version):
headers_list = paper_headers_v2_6
else:
raise UnknownHeaderError(
"Couldn't find parser for paper version %s" % (version))
headers = {}
for i in range(0, len(headers_list)):
this_arg = "" # It's acceptable for header rows to leave off delimiters, so enter missing trailing args as blanks.
try:
this_arg = clean_entry(header_array[i])
except IndexError:
# [JACOB WHAT DOES THIS INDICATE?]
pass
headers[headers_list[i]] = this_arg
return headers
def parse_header(header_array, is_paper=False):
""" Decides which version of the headers to use."""
if not is_paper:
version = clean_entry(header_array[2])
if old_eheaders_re.match(version):
headers_list = old_eheaders
elif new_eheaders_re.match(version):
headers_list = new_eheaders
else:
raise UnknownHeaderError ("Couldn't find parser for electronic version %s" % (version))
else:
version = clean_entry(header_array[1])
if paper_headers_v1_re.match(version):
headers_list = paper_headers_v1
elif paper_headers_v2_2_re.match(version):
headers_list = paper_headers_v2_2
elif paper_headers_v2_6_re.match(version):
headers_list = paper_headers_v2_6
else:
raise UnknownHeaderError ("Couldn't find parser for paper version %s" % (version))
headers = {}
for i in range(0, len(headers_list)):
this_arg = "" # It's acceptable for header rows to leave off delimiters, so enter missing trailing args as blanks.
try:
this_arg = clean_entry(header_array[i])
except IndexError:
# [JACOB WHAT DOES THIS INDICATE?]
pass
headers[headers_list[i]] = this_arg
return headers
def parse_line(self, line_array, version):
""" Return a dict of all variables"""
found_version = False
regex_key = None
# make sure we have this version; since these regexes are non-overlapping, we don't care about the order, and can iterate over the hash keys (instead of an ordered array)
for regex in self.regex_dict:
if (re.match(regex, version)):
regex_key = regex
found_version = True
if not found_version:
raise Exception("Can't find data to parse line type=%s version=%s" % (self.form, version))
line_dict = {}
for column in self.column_locations_dict[regex_key]:
col_position = self.column_locations_dict[regex_key][column]
# sometimes trailing commas are omitted, so test that there actually is a value
if (col_position <= len(line_array) - 1):
line_dict[column] = clean_entry(line_array[col_position])
else:
line_dict[column] = ''
return line_dict
