Update app.py

app.py

@@ -1,42 +1,59 @@
-from sepa import parser
-import re
-import pandas as pd
 import gradio as gr
+import pandas as pd
 import numpy as np
+from sepa import parser
+import re
 
-def full_function(xml_file):
 
-    # Utility function to remove additional namespaces from the XML
+##################################################################
+##################################################################
+##################################################################
+
+def full_function(xml_file):
+    
+    #for gradio: swap with xml_file for local testing
+    full_name = xml_file.name
+    #full_name = xml_file
+
+    
+    print("File name in gradio is ")
+    print(full_name)
+    
     def strip_namespace(xml):
         return re.sub(' xmlns="[^"]+"', '', xml, count=1)
 
     # Read file
-    with open(xml_file, 'r') as f:
+    with open(full_name, 'r') as f:
         input_data = f.read()
 
     # Parse the bank statement XML to dictionary
+    print("Parse full xml string")
     camt_dict = parser.parse_string(parser.bank_to_customer_statement, bytes(strip_namespace(input_data), 'utf8'))
 
     statements = pd.DataFrame.from_dict(camt_dict['statements'])
     all_entries = []
+    dd_all = []
+    
+    print("Start loop all the transactions and add to df")
     for i,_ in statements.iterrows():
         if 'entries' in camt_dict['statements'][i]:
 
             #create empty df
             df = pd.DataFrame()
-            dd = pd.DataFrame.from_records(camt_dict['statements'][i]['entries'])
+            dd = pd.DataFrame.from_records(camt_dict['statements'][i]['entries']) 
 
             df['reference'] = dd['reference']
             df['credit_debit_indicator'] = dd['credit_debit_indicator']
             df['status'] = dd['status']
             df['account_servicer_reference'] = dd['account_servicer_reference']
-            
+
             iban = camt_dict['statements'][i]['account']['id']['iban']
             name = camt_dict['statements'][i]['account']['name']
             df['iban'] = iban
             df['name'] = name
             df['currency'] = dd['amount'].str['currency']
             df['amount'] = dd['amount'].str['_value']
+            df['reference'] = dd['reference']
 
             df['value_date'] = dd['value_date'].str['date']
             df['value_date'] = pd.to_datetime(df['value_date']).dt.strftime('%Y-%m-%d')
@@ -54,38 +71,68 @@ def full_function(xml_file):
             #transaction details
             df['debtor_name'] = dd['entry_details'].str[0].str['transaction_details'].str[0].str['related_parties'].str['debtor'].str['name']
             df['debtor_iban'] = dd['entry_details'].str[0].str['transaction_details'].str[0].str['related_parties'].str['debtor_account'].str['id'].str['iban']
+            df['creditor_name'] = dd['entry_details'].str[0].str['transaction_details'].str[0].str['related_parties'].str['creditor'].str['name']
+            df['creditor_iban'] = dd['entry_details'].str[0].str['transaction_details'].str[0].str['related_parties'].str['creditor_account'].str['id'].str['iban']
+
+            df['bic'] = dd['entry_details'].str[0].str['transaction_details'].str[0].str['related_agents'].str['debtor_agent'].str['financial_institution'].str['bic']
+            df['remittance_information'] = dd['entry_details'].str[0].str['transaction_details'].str[0].str['remittance_information'].str['unstructured'].str[0]
 
             df['account_servicer_reference'] = dd['entry_details'].str[0].str['transaction_details'].str[0].str['refs'].str['account_servicer_reference']
             df['end_to_end_id'] = dd['entry_details'].str[0].str['transaction_details'].str[0].str['refs'].str['end_to_end_id']
 
+        all_entries.append(df)
 
-            all_entries.append(df)
-
+    print("merge the list into df")
     df_entries = pd.concat(all_entries)
-    df_entries.head()
 
-    return df_entries
+    #drop duplicates
+    print("remove duplicate rows")
+    df_entries = df_entries.drop_duplicates(subset=['reference'], keep='last')
+    
+    print("all done")
+    
+    df_entries_example = df_entries[['reference', 'credit_debit_indicator', 'iban', 'name', 'currency', 'amount', 'value_date', 'debtor_name', 'debtor_iban', 'creditor_name', 'creditor_iban', 'remittance_information']].head(20)
+    #print(df_entries_example)
+    
+    return df_entries, df_entries_example
 
 
 
+##################################################################
+##################################################################
+##################################################################
 
 
+def export_csv(xml_file):
+    
+    df_entries, df_entries_example = full_function(xml_file)
+    
+    df_entries.to_csv("./output.csv")
+    
+    out = gr.File.update(value="output.csv", visible=True)
+    
+    return out, df_entries_example
+
+##################################################################
+##################################################################
+##################################################################
+
 desc = "Upload XML file, convert to .csv file, and analyze transactions"
 
 with gr.Blocks() as demo: 
 
     xml_file = gr.File(label = "Upload XML file here")
      
-    # input_employees = gr.CheckboxGroup(["Transfer Solutions", "Ordina", "PwC", "Quistor", "Full Orbit", "Accenture", "Atos", "AMIS"], label="Oracle Partners", info="Who to judge?")
-    # input_question = gr.Text(label="What activity is the Oracle Partner performing?")
-    # additional_info = gr.Text(label="Additional information (optional)")
-    output_text = gr.Text(label="R")
-    df_entries = gr.DataFrame(label="Output table")
-
-
-    submit_btn = gr.Button("Run analysis on XML file")
+    #output table. 
+    df_entries_example = gr.DataFrame(label="Example output table, top 20 rows (not all columns)")
+    
+    with gr.Row():
+        #export_button = gr.Button("Export")
+        out = gr.File(label = "Output file", interactive=False, visible=False)
 
 
-gr.Interface(fn=full_function, inputs=xml_file, outputs=df_entries, title=desc).launch(share=True)
+    #submit_btn = gr.Button("Run analysis on XML file")
+    #export_button.click(export_csv, df_entries, csv)
 
 
+gr.Interface(fn=export_csv, inputs=xml_file, outputs=[out, df_entries_example], title=desc).launch(share=True, debug =True)