def parse_protocol(row):
original_filename, ext, output_filename, full_output_filename, download_filename = get_filenames(
row)
if os.path.exists(full_output_filename):
logging.info('file exists: {}'.format(full_output_filename))
stats["existing files"] += 1
filesize = os.path.getsize(full_output_filename)
crc32c = get_crc32c(full_output_filename)
logging.info('existing file: {}'.format(full_output_filename))
elif os.path.exists(download_filename):
with open(download_filename, "rb") as f:
with CommitteeMeetingProtocol.get_from_file(f) as protocol:
os.makedirs(os.path.dirname(full_output_filename),
exist_ok=True)
with utils.temp_file() as temp_filename:
with open(temp_filename, "w") as of:
if parse_type == "text":
of.write(protocol.text)
else:
csv_writer = csv.writer(of)
csv_writer.writerow(["header", "body"])
for part in protocol.parts:
csv_writer.writerow([part.header, part.body])
shutil.copy(temp_filename, full_output_filename)
filesize = os.path.getsize(full_output_filename)
crc32c = get_crc32c(full_output_filename)
logging.info('parsed file: {}'.format(full_output_filename))
stats["parsed files"] += 1
else:
logging.warning('missing document committee session file: {}'.format(
download_filename))
ext, output_filename, filesize, crc32c = None, None, 0, None
return ext, output_filename, filesize, crc32c
def get_resource():
for row_num, row in enumerate(download_rows):
logging.info("{} / {}".format(row_num, len(download_rows)))
try:
original_filename = os.path.join("files", str(row["GroupTypeID"]),
str(row["DocumentCommitteeSessionID"])[0],
str(row["DocumentCommitteeSessionID"])[1],
str(row["DocumentCommitteeSessionID"]) + "." + row["ApplicationDesc"])
ext = os.path.splitext(original_filename)[1].lower()
output_filename = "files/{}/{}/{}.{}".format(str(row["CommitteeSessionID"])[0],
str(row["CommitteeSessionID"])[1],
str(row["CommitteeSessionID"]),
"csv" if parse_type == "parts" else "txt")
if not files_limit or stats["parsed files"] < files_limit:
if download_from_path:
download_filename = "../data/committees/download_document_committee_session/" + original_filename
if os.path.exists(download_filename):
with open(download_filename, "rb") as f:
with CommitteeMeetingProtocol.get_from_file(f) as protocol:
parse_protocol(output_filename, protocol)
else:
logging.warning("missing download_filename {}".format(download_filename))
elif download_from_remote_storage:
url = download_from_remote_storage + original_filename
with CommitteeMeetingProtocol.get_from_url(url) as protocol:
parse_protocol(output_filename, protocol)
else:
raise Exception("no valid download option")
row.update(protocol_extension=ext,
parsed_filename=output_filename)
yield row
except Exception as e:
# there is a bug in knesset-data-python which prevents getting the error message from the exception
# TODO: fix this bug
error_message = "failed to parse CommitteeSessionID {}".format(row["CommitteeSessionID"]) # , str(e))
logging.exception(error_message)
row.update(error=error_message)
errors.append(row)
def process_row(row, row_index, resource_descriptor, resource_index,
parameters, stats):
if resource_descriptor['name'] == 'kns_documentcommitteesession':
t = parameters['type']
row[t + "_protocol_extension"] = None
row[t + "_parsed_filename"] = None
row[t + "_filesize"] = 0
row[t + "_crc32c"] = None
row[t + "_error"] = None
if (row['GroupTypeID'] == 23 and row['ApplicationDesc'] == 'DOC'
and (row["FilePath"].lower().endswith('.doc')
or row["FilePath"].lower().endswith('.docx'))):
document_id = "{}-{}-{}".format(row["GroupTypeID"],
row["DocumentCommitteeSessionID"],
row["ApplicationDesc"])
original_filename, ext, output_filename, full_output_filename, download_filename, full_output_hash_filename = get_filenames(
row, parameters)
if os.path.exists(download_filename) and row.get(
'download_crc32c'):
m = BASE_HASH_OBJ.copy()
m.update(row['download_crc32c'].encode())
new_cache_hash = m.hexdigest()
if os.path.exists(full_output_filename) and os.path.exists(
full_output_hash_filename):
with open(full_output_hash_filename) as f:
old_cache_hash = f.read()
else:
old_cache_hash = None
if old_cache_hash and new_cache_hash and new_cache_hash == old_cache_hash:
stats[t + ": existing files"] += 1
row[t + "_protocol_extension"] = ext
row[t + "_parsed_filename"] = output_filename
row[t +
"_filesize"] = os.path.getsize(full_output_filename)
row[t + "_crc32c"] = get_crc32c(full_output_filename)
elif parameters.get('files-limit') and parameters[
'files-limit'] <= stats[t + ": parsed files"]:
row[t + "_error"] = 'reached files-limit, skipping'
stats[t + ": skipped files"] += 1
else:
error_string = None
try:
with open(download_filename, "rb") as f:
with CommitteeMeetingProtocol.get_from_file(
f) as protocol:
os.makedirs(
os.path.dirname(full_output_filename),
exist_ok=True)
with utils.temp_file() as temp_filename:
with open(temp_filename, "w") as of:
if parameters['type'] == "text":
of.write(protocol.text)
else:
csv_writer = csv.writer(of)
csv_writer.writerow(
["header", "body"])
for part in protocol.parts:
csv_writer.writerow(
[part.header, part.body])
shutil.copy(temp_filename,
full_output_filename)
except Exception as e:
logging.exception(
'exception parsing protocol for {}'.format(
document_id))
try:
error_string = str(e)
except Exception:
error_string = 'unexpected exception'
if error_string:
row[t + "_error"] = error_string
stats[t + ': errored files'] += 1
else:
row[t + "_protocol_extension"] = ext
row[t + "_parsed_filename"] = output_filename
row[t + "_filesize"] = os.path.getsize(
full_output_filename)
row[t + "_crc32c"] = get_crc32c(full_output_filename)
stats[t + ": parsed files"] += 1
with open(full_output_hash_filename, 'w') as f:
f.write(new_cache_hash)
else:
row[t + "_error"] = 'missing download file'
stats[t + ': missing download files'] += 1
return row
