Spaces:
Sleeping
Sleeping
Update appStore/prep_data.py
Browse files- appStore/prep_data.py +5 -1
appStore/prep_data.py
CHANGED
@@ -100,8 +100,12 @@ def process_giz_worldwide():
|
|
100 |
giz_df['url'] = ''
|
101 |
|
102 |
# Convert CRS value to numeric then to integer if possible.
|
|
|
103 |
if 'crs_key' in giz_df.columns:
|
104 |
-
giz_df['crs_key'] = giz_df['crs_key'].apply(
|
|
|
|
|
|
|
105 |
|
106 |
# Compute text_size based on merged_text and assign full text to the 'chunks' column
|
107 |
giz_df['text_size'] = giz_df['merged_text'].apply(lambda text: len(text.split()) if isinstance(text, str) else 0)
|
|
|
100 |
giz_df['url'] = ''
|
101 |
|
102 |
# Convert CRS value to numeric then to integer if possible.
|
103 |
+
# After reading the JSON file into giz_df, convert the crs_key column:
|
104 |
if 'crs_key' in giz_df.columns:
|
105 |
+
giz_df['crs_key'] = giz_df['crs_key'].apply(
|
106 |
+
lambda x: str(int(float(x))).replace('.0', '') if pd.notnull(x) and str(x).strip() != "" else x
|
107 |
+
)
|
108 |
+
|
109 |
|
110 |
# Compute text_size based on merged_text and assign full text to the 'chunks' column
|
111 |
giz_df['text_size'] = giz_df['merged_text'].apply(lambda text: len(text.split()) if isinstance(text, str) else 0)
|