Update app.py

app.py

@@ -2,13 +2,15 @@ import gradio as gr
 import pandas as pd
 from datetime import datetime
 import pytz
+import re
 
 # Updated Gradio app for YunExpress:
 # - New output headers (as provided)
 # - Email -> Email mapping
 # - FOBPrice1 = 2 (replaces UnitPrice1)
-# - ZIP padding fix for US 4-digit ZIPs
-# - RoutingCode logic: US/NO -> HK-ASS-PF; others -> HKTHZXR
+# - ZIP padding fix for US 3- and 4-digit ZIPs (no leading apostrophe; sanitize non [A-Za-z0-9 ] chars)
+# - RoutingCode logic: US/NO/FR -> HK-ASS-PF; others -> HKTHZXR
+# - Save Excel with ZipCode column set to Text (@) to preserve leading zeros without apostrophes
 
 def process_file(file):
     file_name = file.name.lower()
@@ -73,7 +75,7 @@ def process_file(file):
     if "Shipping Country Code" in df.columns:
         output_df["CountryCode"] = df["Shipping Country Code"]
 
-    # 7) ZipCode (pad US 4-digit to 5, prefix apostrophe)
+    # 7) ZipCode (sanitize; US 3/4-digit -> pad to 5; do NOT add apostrophe)
     if "Shipping ZIP" in df.columns:
         zip_raw = (
             df["Shipping ZIP"]
@@ -81,10 +83,16 @@ def process_file(file):
             .str.strip()
             .str.replace(r"\.0$", "", regex=True)  # clean "1234.0"
         )
-        mask_us = output_df["CountryCode"].eq("US")
-        mask_4  = zip_raw.str.fullmatch(r"\d{4}")
-        zip_padded = zip_raw.where(~(mask_us & mask_4), "0" + zip_raw)
-        output_df["ZipCode"] = zip_padded.where(~(mask_us & mask_4), "'" + zip_padded)
+        # Keep only letters, digits, spaces (avoid "only letters, numbers and spaces" API error)
+        zip_clean = zip_raw.str.replace(r"[^A-Za-z0-9 ]+", "", regex=True)
+
+        mask_us  = output_df["CountryCode"].eq("US")
+        # For US: if 3 or 4 numeric digits -> zero-fill to 5
+        mask_3_4_digits = zip_clean.str.fullmatch(r"\d{3,4}")
+        zip_padded = zip_clean.where(~(mask_us & mask_3_4_digits), zip_clean.str.zfill(5))
+
+        # Do NOT add any leading apostrophe
+        output_df["ZipCode"] = zip_padded
 
     # 8) phone
     if "Shipping Address Phone" in df.columns:
@@ -105,10 +113,10 @@ def process_file(file):
     # Fixed defaults & RoutingCode
     mask = output_df["CustomerOrderNo."].astype(str).str.len() > 0
 
-    # RoutingCode: HK-ASS-PF if US/NO else HKTHZXR
-    mask_us_or_no = mask & output_df["CountryCode"].isin(["US", "NO"])
-    mask_other    = mask & ~output_df["CountryCode"].isin(["US", "NO"])
-    output_df.loc[mask_us_or_no, "RoutingCode"] = "HK-ASS-PF"
+    # RoutingCode: HK-ASS-PF if US/NO/FR else HKTHZXR
+    mask_us_no_fr = mask & output_df["CountryCode"].isin(["US", "NO", "FR"])
+    mask_other    = mask & ~output_df["CountryCode"].isin(["US", "NO", "FR"])
+    output_df.loc[mask_us_no_fr, "RoutingCode"] = "HK-ASS-PF"
     output_df.loc[mask_other,    "RoutingCode"] = "HKTHZXR"
 
     # Pricing / descriptions / weights
@@ -133,8 +141,26 @@ def process_file(file):
     today_hk = datetime.now(hk_tz).strftime("%y%m%d")
     output_file_name = f"yunexpress {today_hk}.xlsx"
 
-    # Save to Excel
-    output_df.to_excel(output_file_name, index=False)
+    # Save to Excel with ZipCode column forced to Text (@) using xlsxwriter
+    try:
+        import xlsxwriter
+        from xlsxwriter.utility import xl_col_to_name
+
+        with pd.ExcelWriter(output_file_name, engine="xlsxwriter") as writer:
+            output_df.to_excel(writer, index=False, sheet_name="Sheet1")
+            workbook  = writer.book
+            worksheet = writer.sheets["Sheet1"]
+
+            # Text format to preserve leading zeros without adding apostrophes
+            text_fmt = workbook.add_format({"num_format": "@"})
+
+            # Locate ZipCode column and set entire column to text
+            zip_col_idx = output_df.columns.get_loc("ZipCode")  # 0-based index
+            col_letter = xl_col_to_name(zip_col_idx)
+            worksheet.set_column(f"{col_letter}:{col_letter}", None, text_fmt)
+    except Exception:
+        # Fallback to default writer if xlsxwriter not available
+        output_df.to_excel(output_file_name, index=False)
 
     return output_df, output_file_name