def extract_recap_pdf(pk, skip_ocr=False):
doc = RECAPDocument.objects.get(pk=pk)
path = doc.filepath_local.path
process = make_pdftotext_process(path)
content, err = process.communicate()
if needs_ocr(content):
if not skip_ocr:
# probably an image PDF. Send it to OCR.
success, content = extract_by_ocr(path)
if success:
doc.ocr_status = RECAPDocument.OCR_COMPLETE
elif content == u"" or not success:
content = u"Unable to extract document content."
doc.ocr_status = RECAPDocument.OCR_FAILED
else:
content = u""
doc.ocr_status = RECAPDocument.OCR_NEEDED
else:
doc.ocr_status = RECAPDocument.OCR_UNNECESSARY
doc.plain_text, _ = anonymize(content)
doc.save()
return path
def extract_recap_pdf(
pks: Union[int, List[int]],
skip_ocr: bool = False,
check_if_needed: bool = True,
) -> List[int]:
"""Extract the contents from a RECAP PDF if necessary."""
if not is_iter(pks):
pks = [pks]
processed = []
for pk in pks:
rd = RECAPDocument.objects.get(pk=pk)
if check_if_needed and not rd.needs_extraction:
# Early abort if the item doesn't need extraction and the user
# hasn't disabled early abortion.
processed.append(pk)
continue
with NamedTemporaryFile(
prefix="extract_file_",
suffix=".pdf",
buffering=0, # Make sure it's on disk when we try to use it
) as tmp:
tmp.write(rd.filepath_local.read())
process = make_pdftotext_process(tmp.name)
content, err = process.communicate()
content = content.decode()
if needs_ocr(content):
if not skip_ocr:
# probably an image PDF. Send it to OCR.
success, content = extract_by_ocr(tmp.name)
if success:
rd.ocr_status = RECAPDocument.OCR_COMPLETE
elif content == "" or not success:
content = "Unable to extract document content."
rd.ocr_status = RECAPDocument.OCR_FAILED
else:
content = ""
rd.ocr_status = RECAPDocument.OCR_NEEDED
else:
rd.ocr_status = RECAPDocument.OCR_UNNECESSARY
rd.plain_text, _ = anonymize(content)
# Do not do indexing here. Creates race condition in celery.
rd.save(index=False, do_extraction=False)
processed.append(pk)
return processed
def extract_recap_pdf(pks, skip_ocr=False, check_if_needed=True):
"""Extract the contents from a RECAP PDF if necessary."""
if not is_iter(pks):
pks = [pks]
processed = []
for pk in pks:
rd = RECAPDocument.objects.get(pk=pk)
if check_if_needed and not rd.needs_extraction:
# Early abort if the item doesn't need extraction and the user
# hasn't disabled early abortion.
processed.append(pk)
continue
path = rd.filepath_local.path
process = make_pdftotext_process(path)
content, err = process.communicate()
content = content.decode()
if needs_ocr(content):
if not skip_ocr:
# probably an image PDF. Send it to OCR.
success, content = extract_by_ocr(path)
if success:
rd.ocr_status = RECAPDocument.OCR_COMPLETE
elif content == "" or not success:
content = "Unable to extract document content."
rd.ocr_status = RECAPDocument.OCR_FAILED
else:
content = ""
rd.ocr_status = RECAPDocument.OCR_NEEDED
else:
rd.ocr_status = RECAPDocument.OCR_UNNECESSARY
rd.plain_text, _ = anonymize(content)
# Do not do indexing here. Creates race condition in celery.
rd.save(index=False, do_extraction=False)
processed.append(pk)
return processed
def extract_recap_pdf(pks, skip_ocr=False, check_if_needed=True):
"""Extract the contents from a RECAP PDF if necessary."""
if not is_iter(pks):
pks = [pks]
processed = []
for pk in pks:
rd = RECAPDocument.objects.get(pk=pk)
if check_if_needed and not rd.needs_extraction:
# Early abort if the item doesn't need extraction and the user
# hasn't disabled early abortion.
processed.append(pk)
continue
path = rd.filepath_local.path
process = make_pdftotext_process(path)
content, err = process.communicate()
if needs_ocr(content):
if not skip_ocr:
# probably an image PDF. Send it to OCR.
success, content = extract_by_ocr(path)
if success:
rd.ocr_status = RECAPDocument.OCR_COMPLETE
elif content == u'' or not success:
content = u'Unable to extract document content.'
rd.ocr_status = RECAPDocument.OCR_FAILED
else:
content = u''
rd.ocr_status = RECAPDocument.OCR_NEEDED
else:
rd.ocr_status = RECAPDocument.OCR_UNNECESSARY
rd.plain_text, _ = anonymize(content)
# Do not do indexing here. Creates race condition in celery.
rd.save(index=False, do_extraction=False)
processed.append(pk)
return processed
