Update app.py

app.py

@@ -1,164 +1,76 @@
 # --------------------------------------------------------------
-# gradio_s3_sql.py   (Gradio-compatible, no 'height' error)
+# simple_s3_sql.py
 # --------------------------------------------------------------
 import boto3
 import pandas as pd
 import duckdb
 from io import StringIO
-from botocore.exceptions import ClientError
 import gradio as gr
 
-# === YOUR IDRIVE E2 CONFIG (hardcoded) ===
+# === YOUR CREDENTIALS ===
 ENDPOINT_URL = "https://s3.us-west-1.idrivee2.com"
 ACCESS_KEY = "rNuPBAQetemqpEeBospZ"
 SECRET_KEY = "BU4FccUYxzXVqiWjPSJM1CWEX1cNhBqbU9NeGidE"
-BUCKET_NAME = "accusagas3"
+BUCKET = "accusagas3"
 
-s3 = boto3.client(
-    "s3",
-    endpoint_url=ENDPOINT_URL,
-    aws_access_key_id=ACCESS_KEY,
-    aws_secret_access_key=SECRET_KEY,
-)
+s3 = boto3.client("s3", endpoint_url=ENDPOINT_URL,
+                  aws_access_key_id=ACCESS_KEY,
+                  aws_secret_access_key=SECRET_KEY)
 
 
 def run_sql(path: str, sql: str) -> pd.DataFrame:
-    """S3 → CSV → DuckDB → SQL → DataFrame (auto-cast + auto-fix)"""
-    # --- Load CSV ---
     try:
-        obj = s3.get_object(Bucket=BUCKET_NAME, Key=path)
+        obj = s3.get_object(Bucket=BUCKET, Key=path)
         df = pd.read_csv(StringIO(obj["Body"].read().decode("utf-8")))
-    except ClientError as e:
-        if e.response["Error"]["Code"] == "NoSuchKey":
-            return pd.DataFrame({"error": [f"File not found: {path}"]})
-        return pd.DataFrame({"error": [f"S3 Error: {e}"]})
     except Exception as e:
-        return pd.DataFrame({"error": [f"CSV read failed: {e}"]})
+        return pd.DataFrame({"error": [str(e)]})
 
     if df.empty:
-        return pd.DataFrame({"error": ["CSV is empty"]})
+        return pd.DataFrame({"error": ["Empty CSV"]})
 
-    # --- Auto-convert numeric columns ---
-    numeric_keywords = ["price", "amount", "value", "cost", "revenue", "total", "volume", "open", "high", "low", "close"]
+    # Auto-convert Price, Volume, etc.
     for col in df.columns:
-        if any(kw in col.lower() for kw in numeric_keywords):
-            cleaned = df[col].astype(str).str.replace(r"[^\d.-]", "", regex=True)
-            df[col] = pd.to_numeric(cleaned, errors="coerce")
+        if any(x in col.lower() for x in ["price", "volume", "amount"]):
+            df[col] = pd.to_numeric(df[col].astype(str).str.replace(r"[^\d.-]", "", regex=True), errors="coerce")
 
-    # --- DuckDB ---
     con = duckdb.connect(":memory:")
     con.register("data", df)
 
-    norm_sql = sql.strip().lower()
-    if not norm_sql.startswith(("select", "with")):
+    if not sql.strip().lower().startswith(("select", "with")):
         con.close()
-        return pd.DataFrame({"error": ["Only SELECT or WITH queries allowed"]})
+        return pd.DataFrame({"error": ["Only SELECT allowed"]})
 
     try:
         result = con.execute(sql).df()
     except Exception as e:
-        # Auto-fix VARCHAR vs number
-        if "Cannot compare values of type VARCHAR" in str(e):
-            import re
-            match = re.search(r"column ([a-zA-Z0-9_]+)", str(e), re.I)
-            col = match.group(1) if match else None
-            if col and col in df.columns:
-                fixed_sql = sql.replace(f"{col}", f"CAST({col} AS DOUBLE)")
-                try:
-                    result = con.execute(fixed_sql).df()
-                except Exception as e2:
-                    con.close()
-                    return pd.DataFrame({"error": [f"SQL failed even after CAST: {e2}"]})
+        # Try auto-casting
+        if "VARCHAR" in str(e):
+            col = __import__("re").search(r"column ([a-zA-Z0-9_]+)", str(e))
+            if col and (c := col.group(1)) in df.columns:
+                sql = sql.replace(c, f"CAST({c} AS DOUBLE)")
+                result = con.execute(sql).df()
             else:
                 con.close()
-                return pd.DataFrame({"error": [f"Type error: {e}"]})
+                return pd.DataFrame({"error": [str(e)]})
         else:
             con.close()
-            return pd.DataFrame({"error": [f"SQL Error: {e}"]})
+            return pd.DataFrame({"error": [str(e)]})
     finally:
         con.close()
 
-    return result.head(10_000)
+    return result.head(10000)
 
 
-# --------------------------------------------------------------
-# Gradio UI (no 'height' on Dataframe)
-# --------------------------------------------------------------
-with gr.Blocks(title="S3 SQL Query (iDrive e2)") as demo:
-    gr.Markdown(
-        """
-        # S3 CSV SQL Explorer  
-        Query any CSV in your iDrive e2 bucket using SQL  
-        Table name: `data` | Auto-casts `Price`, `Amount`, etc.
-        """
-    )
-
-    with gr.Row():
-        path_input = gr.Textbox(
-            label="S3 Path (Key)",
-            placeholder="vatsav_123/reports/Gold Futures Historical Data.csv",
-            lines=1,
-        )
-        sql_input = gr.Textbox(
-            label="SQL Query",
-            placeholder="SELECT Date, Price FROM data WHERE Price > 1000 ORDER BY Date DESC LIMIT 10",
-            lines=4,
-        )
-
-    run_btn = gr.Button("Run SQL", variant="primary")
-
-    # Use DataGrid + CSS for fixed height
-    output = gr.DataGrid(
-        label="Result",
-        interactive=False,
-        wrap=True,
-    )
+# === GRADIO UI ===
+with gr.Blocks() as app:
+    gr.Markdown("## S3 CSV SQL")
+    path = gr.Textbox(label="S3 Path", placeholder="vatsav_123/reports/Gold Futures Historical Data.csv")
+    sql = gr.Textbox(label="SQL", lines=3, placeholder="SELECT Date, Price FROM data WHERE Price > 1000")
+    btn = gr.Button("Run")
+    out = gr.Dataframe()  # Only this – no height, no extras
 
-    # Apply height via CSS
-    demo.load(
-        None,
-        None,
-        None,
-        _js=f"""
-        () => {{
-            setTimeout(() => {{
-                const grid = document.querySelector('.gradio-container .data-grid');
-                if (grid) grid.style.maxHeight = '500px';
-                if (grid) grid.style.overflowY = 'auto';
-            }}, 100);
-        }}
-        """
-    )
+    btn.click(run_sql, [path, sql], out)
 
-    run_btn.click(
-        fn=run_sql,
-        inputs=[path_input, sql_input],
-        outputs=output,
-        show_progress=True,
-    )
 
-    gr.Examples(
-        examples=[
-            [
-                "vatsav_123/reports/Gold Futures Historical Data.csv",
-                "SELECT Date, Price FROM data WHERE Price > 2000 ORDER BY Date DESC LIMIT 10"
-            ],
-            [
-                "vatsav_123/reports/Gold Futures Historical Data.csv",
-                "SELECT MIN(Price) AS min_price, MAX(Price) AS max_price FROM data"
-            ],
-        ],
-        inputs=[path_input, sql_input],
-    )
-
-
-# --------------------------------------------------------------
-# Launch
-# --------------------------------------------------------------
 if __name__ == "__main__":
-    demo.launch(
-        server_name="0.0.0.0",
-        server_port=7860,
-        share=False,
-        debug=True,
-    )
+    app.launch(server_name="0.0.0.0", server_port=7860)