def parse(self, document_path, mime_type):
self.log("info",
f"[TIKA_PARSE] Sending {document_path} to Tika server")
try:
parsed = parser.from_file(document_path)
except requests.exceptions.HTTPError as err:
raise ParseError(
f"Could not parse {document_path} with tika server: {err}")
try:
content = parsed["content"].strip()
except:
content = ""
try:
creation_date = dateutil.parser.isoparse(
parsed["metadata"]["Creation-Date"])
except:
creation_date = None
archive_path = os.path.join(self.tempdir, "convert.pdf")
convert_to_pdf(self, document_path, archive_path)
self.archive_path = archive_path
self.date = creation_date
self.text = content
def extract_metadata(self, document_path, mime_type):
tika_server = settings.PAPERLESS_TIKA_ENDPOINT
try:
parsed = parser.from_file(document_path, tika_server)
except Exception as e:
self.log(
"warning", f"Error while fetching document metadata for "
f"{document_path}: {e}")
return []
return [{
"namespace": "",
"prefix": "",
"key": key,
"value": parsed['metadata'][key]
} for key in parsed['metadata']]
def parse(self, document_path, mime_type, file_name=None):
self.log("info", f"Sending {document_path} to Tika server")
tika_server = settings.PAPERLESS_TIKA_ENDPOINT
try:
parsed = parser.from_file(document_path, tika_server)
except Exception as err:
raise ParseError(
f"Could not parse {document_path} with tika server at "
f"{tika_server}: {err}")
self.text = parsed["content"].strip()
try:
self.date = dateutil.parser.isoparse(
parsed["metadata"]["Creation-Date"])
except Exception as e:
self.log(
"warning", f"Unable to extract date for document "
f"{document_path}: {e}")
self.archive_path = self.convert_to_pdf(document_path, file_name)