Philipp/aki_prj23_transparenzregister
1
0
Fork 0
mirror of https://github.com/fhswf/aki_prj23_transparenzregister.git synced 2025-06-22 07:53:55 +02:00
Code Issues Packages Projects Releases Wiki Activity

fix(data-extraction): Resolve regex issue in detecting auditors

Browse Source
This commit is contained in:
TrisNol
2023-10-17 17:56:26 +02:00
parent c680ac9759
commit 4058824f15
1 changed files with 2 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/aki_prj23_transparenzregister/utils/data_extraction/bundesanzeiger.py
View File

@ -102,12 +102,9 @@ class Bundesanzeiger:
list[Auditor]: List of Auditors found in the given report list[Auditor]: List of Auditors found in the given report
""" """
auditor_company = self.extract_auditor_company(report) auditor_company = self.extract_auditor_company(report)
auditor_regex = r"[a-z A-Z,.'-]+, Wirtschaftsprüfer" auditor_regex = r"([a-z A-ZÄäÜüÖö,.'-]+), Wirtschaftsprüfer(in)?"
hits = re.findall(auditor_regex, report) hits = re.findall(auditor_regex, report)
return [ return [Auditor(hit[0].strip(), auditor_company) for hit in hits]
Auditor(hit.replace(", Wirtschaftsprüfer", "").lstrip(), auditor_company)
for hit in hits
]
def __extract_kpis__(self, report: str) -> dict: def __extract_kpis__(self, report: str) -> dict:
"""Source: https://github.com/bundesAPI/deutschland/pull/87/files#diff-f5b9db5384cf523fcc677056065041e7793bfc4da9cf74c4eebd6fab732739bd. """Source: https://github.com/bundesAPI/deutschland/pull/87/files#diff-f5b9db5384cf523fcc677056065041e7793bfc4da9cf74c4eebd6fab732739bd.