def setup_mock():
doc_field.uid = 'ABCDEF'
doc_field.code = 'client'
csv_text = """
,value,pattern
0,"Big Bank & Company (004578) (Knight, Bobby (Charlotte); Bryant, Koby (Charlotte); Williams, Gary (Charlotte); Johnson, Magic (Charlotte); Lobo, Rebecca (Charlotte))","\bbig\s{1,5}bank\s{1,5}.{1,5}\s{1,5}company\s{1,5}(004578)\b"
1,"Family Name Limited (173437) (Tanner, Rebecca (Houston); Saget, Bob (Houston))","family\s{1,5}name\s{1,5}\(173437\)"
2,"Financial Services & Co. (015607) (Spelling, Tori (Chicago); Priestley, Jason (Dallas); Perry, Luke (New York); Doherty, Shannon (Chicago); Garth, Jenny (Chicago))","\bfinancial\s{1,5}services\s{1,5}.{1,5}(015607)\b"
3,"Food Wholsale, Inc. (056230) (Jenner, Bruce (Chicago))","\bfood\s{1,5}wholsale,(056230)\b"
4,"All Eyes Communications (018951) (Moore, Michael (New York); Tarantino, Quentin (San Francisco); Lee, Spike (New York); Levinson, Barry (Charlotte))","\ball\s{1,5}eyes\s{1,5}communications\s{1,5}(018951)\b"
5,"Joe Smith Archives, LLC d/b/a Foxtrot (085292) (Flay, Bobby (New York))","\bfoxtrot\s{1,5}(085292)\b
\bjoe\s{1,5}smith\s{1,5}archives\b" """
detector = DocumentFieldMultilineRegexDetector()
detector.csv_content = csv_text
detector.document_field = doc_field
detector.update_checksum()
CsvRegexpsDetectionCacheMock.detector_by_field[doc_field.uid] = detector
def save_detector_settings_csv(
self, detectors_by_value: Dict[str, List[str]]) -> None:
detector = DocumentFieldMultilineRegexDetector()
detector.document_field = self.document_field
df = pd.DataFrame(columns=['value', 'pattern'])
df.set_index("pattern", inplace=True)
for field_val in detectors_by_value:
for include_reg_value in detectors_by_value[field_val]:
df = df.append(
{
'value': field_val,
'pattern': include_reg_value
},
ignore_index=True)
df.drop_duplicates(subset='pattern', inplace=True)
try:
existing = DocumentFieldMultilineRegexDetector.objects.get(
document_field_id=self.document_field.uid
) # type: DocumentFieldMultilineRegexDetector
except DocumentFieldMultilineRegexDetector.DoesNotExist:
detector.csv_content = df.to_csv()
detector.update_checksum()
detector.save()
return
# just update CSV content and hashsum
if self.drop_previous_field_detectors:
existing.csv_content = df.to_csv()
existing.update_checksum()
existing.save()
return
# join these options with existing one
# overwriting duplicates by detected_value or regexp pattern
existing.combine_with_dataframe(df)
existing.save()