Philipp/aki_prj23_transparenzregister
Philipp/aki_prj23_transparenzregister
1
0
Fork 0
mirror of https://github.com/fhswf/aki_prj23_transparenzregister.git synced 2025-04-25 12:32:34 +02:00
Code Issues Packages Projects Releases Wiki Activity

feat: Add simple wrapper to update particual financial entries

Browse Source
This commit is contained in:
TrisNol 2023-09-25 19:33:23 +02:00
parent 2050b49fde
commit 7b5cf16e49
1 changed files with 5 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

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

@ -61,12 +61,15 @@ class Bundesanzeiger:
Returns: Returns:
pd.DataFrame: Filtered and pruned DataFrame pd.DataFrame: Filtered and pruned DataFrame
""" """
date_regex = r"\d{2}\.\d{2}.\d{4}"
df_reports["type"] = df_reports.name.apply(lambda name: name.split(" ")[0]) df_reports["type"] = df_reports.name.apply(lambda name: name.split(" ")[0])
df_reports = df_reports.loc[ df_reports = df_reports.loc[
(df_reports.type == "Jahresabschluss") | (df_reports.type == "Jahres-") ((df_reports.type == "Jahresabschluss") | (df_reports.type == "Jahres-"))
& df_reports.name.str.contains(date_regex, regex=True)
] ]
df_reports["jahr"] = df_reports.name.apply( df_reports["jahr"] = df_reports.name.apply(
lambda name: re.findall(r"\d{2}\.\d{2}.\d{4}", name)[0].split(".")[-1] lambda name: re.findall(date_regex, name)[0].split(".")[-1]
) )
return df_reports.drop(["name", "report", "type"], axis=1) return df_reports.drop(["name", "report", "type"], axis=1)