Create app.py

app.py

@@ -0,0 +1,138 @@
+import gradio as gr
+import pandas as pd
+import re
+import unicodedata
+import io
+
+# ---------- Helper Functions ----------
+
+def clean_text(s: str) -> str:
+    """Remove unwanted characters and normalize."""
+    if pd.isna(s):
+        return ""
+    s = str(s).replace("ÿ", "")
+    s = unicodedata.normalize("NFKD", s)
+    s = "".join(ch for ch in s if 32 <= ord(ch) <= 126)
+    return s.strip()
+
+def format_zip(zip_code):
+    """Pad ZIP codes to 5 digits."""
+    if pd.isna(zip_code):
+        return ""
+    z = str(zip_code).strip()
+    z = re.sub(r"[^\d]", "", z)
+    if not z:
+        return ""
+    return z.zfill(5)[:5]
+
+def flow_address_lines(lines, maxlen=35, maxlines=3):
+    """Split long address lines into multiple lines."""
+    tokens = []
+    for ln in lines:
+        txt = clean_text(ln)
+        if txt:
+            tokens.extend(txt.split())
+    out = ["", "", ""]
+    line_i = 0
+    for tok in tokens:
+        while len(tok) > maxlen:
+            chunk, tok = tok[:maxlen], tok[maxlen:]
+            if line_i >= maxlines:
+                return out
+            if out[line_i]:
+                line_i += 1
+                if line_i >= maxlines:
+                    return out
+            out[line_i] = chunk
+            line_i += 1
+            if line_i >= maxlines:
+                return out
+        if line_i >= maxlines:
+            return out
+        add_len = len(tok) if not out[line_i] else len(tok) + 1
+        if len(out[line_i]) + add_len <= maxlen:
+            out[line_i] = (out[line_i] + (" " if out[line_i] else "") + tok).strip()
+        else:
+            line_i += 1
+            if line_i >= maxlines:
+                return out
+            out[line_i] = tok
+    return [ln[:maxlen] for ln in out]
+
+def convert_dry_ice_kg(x):
+    """Convert lbs -> kg and round."""
+    if pd.isna(x) or str(x).strip() == "":
+        return ""
+    try:
+        kg = round(float(str(x).strip()) / 2.2)
+        return str(int(kg))
+    except:
+        return ""
+
+# ---------- Main Cleaning Function ----------
+
+def clean_csv(file):
+    try:
+        df = pd.read_csv(file.name, encoding="latin1")
+    except Exception:
+        df = pd.read_csv(file.name, encoding="utf-8-sig")
+
+    df.columns = df.columns.str.strip()
+
+    # --- Cleaning operations ---
+    if "ZipCode" in df.columns:
+        df["ZipCode"] = df["ZipCode"].map(format_zip)
+
+    # Address split logic
+    addr1, addr2, addr3 = [], [], []
+    for _, row in df.iterrows():
+        a1, a2, a3 = flow_address_lines([
+            row.get("Address1", ""), row.get("Address2", ""), row.get("Address3", "")
+        ])
+        addr1.append(a1)
+        addr2.append(a2)
+        addr3.append(a3)
+    df["Address1"] = addr1
+    df["Address2"] = addr2
+    df["Address3"] = addr3
+
+    # Clean text fields
+    text_cols = ["Company Name", "Contact Name", "City", "State", "Phone Number", "Email"]
+    for col in text_cols:
+        if col in df.columns:
+            df[col] = df[col].map(clean_text)
+
+    # Dry Ice conversion
+    if "Dry Ice Weight" in df.columns:
+        df["Dry Ice Weight (kg)"] = df["Dry Ice Weight"].map(convert_dry_ice_kg)
+
+    # Save to BytesIO for Gradio download
+    buffer = io.BytesIO()
+    df.to_csv(buffer, index=False, encoding="utf-8-sig")
+    buffer.seek(0)
+    return buffer, "cleaned_output.csv"
+
+# ---------- Gradio UI ----------
+
+title = "UPS Shipment CSV Cleaner"
+description = """
+Upload your **raw shipment CSV file** below.  
+This tool will:
+- Remove strange characters (e.g. ÿ)
+- Pad ZIP codes to 5 digits
+- Split long addresses into ≤ 35 characters
+- Convert Dry Ice Weight from lbs → kg  
+Then download the cleaned CSV ready for UPS Batch import.
+"""
+
+demo = gr.Interface(
+    fn=clean_csv,
+    inputs=gr.File(label="📤 Upload CSV File"),
+    outputs=gr.File(label="📥 Download Cleaned CSV"),
+    title=title,
+    description=description,
+    allow_flagging="never"
+)
+
+if __name__ == "__main__":
+    demo.launch()