Fix #3: replace empty string with np.nan #5
@ -83,6 +83,5 @@ def table2df(tables):
|
||||
df = pd.concat(dfs)
|
||||
|
||||
df["immeuble"] = df["immeuble"].apply(lambda x: x[0].capitalize())
|
||||
print(df.columns)
|
||||
df["lot"] = df["RECAPITULATIF DES OPERATIONS"].apply(get_lot)
|
||||
return df.astype(DF_TYPES, errors="ignore")
|
||||
return df.astype(DF_TYPES)
|
||||
|
@ -1,3 +1,4 @@
|
||||
import numpy as np
|
||||
import pandas as pd
|
||||
|
||||
DF_TYPES = {
|
||||
@ -33,7 +34,7 @@ def is_drop(row):
|
||||
|
||||
|
||||
def extract(table, additionnal_fields: dict = {}):
|
||||
"""Turn table to dictionary with additionnal fields"""
|
||||
"""Turn table to dictionary with additional fields"""
|
||||
extracted = []
|
||||
header = table[0]
|
||||
for row in table[1:]:
|
||||
@ -159,4 +160,7 @@ def table2df(tables):
|
||||
df["immeuble"] = df["immeuble"].apply(lambda x: x[0].capitalize())
|
||||
df["Type"] = df["Type"].apply(clean_type)
|
||||
|
||||
return df.astype(DF_TYPES, errors="ignore")
|
||||
numeric_cols = [k for k, v in DF_TYPES.items() if v == float]
|
||||
df[numeric_cols] = df[numeric_cols].replace("", np.nan)
|
||||
|
||||
return df.astype(DF_TYPES)
|
||||
|
Loading…
Reference in New Issue
Block a user