From b09b36b5e1bc06aca7f573c03078e9c4e9f1dded Mon Sep 17 00:00:00 2001
From: Adarsh Kumar <kumaradarsh.cse@gmail.com>
Date: Sun, 19 Oct 2025 13:31:32 +0530
Subject: [PATCH] feat: Add /process/preview endpoint for CSV previews

---
 app/routers/process.py     | 54 +++++++++++++++++++++++++++++++++++++-
 app/services/processing.py | 17 ++++++++++++
 2 files changed, 70 insertions(+), 1 deletion(-)

diff --git a/app/routers/process.py b/app/routers/process.py
index a1e2e16..7d0dd7b 100644
--- a/app/routers/process.py
+++ b/app/routers/process.py
@@ -22,7 +22,7 @@
     ProcessRequestBody,
     ProcessResponse,
 )
-from app.services.processing import parse_input, process_dataframe
+from app.services.processing import parse_input, process_dataframe, generate_preview
 from app.utils.file_utils import download_to_temp
 
 router = APIRouter()
@@ -113,3 +113,55 @@ async def process_csv(
             status_code=400,
             content=ProcessResponse(data=None, stats=None, download_url=None, errors=[{"message": str(e)}]).dict(),
         )
+
+@router.post("/preview")
+async def get_csv_preview(
+    # We accept the same input types as the main endpoint
+    file: Optional[UploadFile] = File(default=None, description="CSV/Excel file to upload"),
+    url: Optional[str] = Form(default=None, description="URL to CSV/Excel file"),
+    raw_csv: Optional[str] = Form(default=None, description="Raw CSV text"),
+):
+    """Generate a JSON preview of a file, URL, or raw CSV.
+
+    Returns the column names and the first 5 rows.
+    """
+    try:
+        # Step 1: Handle URL input if provided
+        url_path: Optional[Path] = None
+        if url:
+            try:
+                url_path, _ = await download_to_temp(url)
+            except Exception as e:
+                raise HTTPException(status_code=400, detail=f"Failed to download URL: {e}")
+
+        # Step 2: Handle file upload input if provided
+        file_bytes: Optional[bytes] = None
+        filename: Optional[str] = None
+        if file is not None:
+            filename = file.filename
+            file_bytes = await file.read()
+
+        # Step 3: Reuse the existing parse_input function to create a DataFrame
+        # We use a default ParsingOptions since we don't need customization for a preview.
+        parsed = parse_input(
+            file_bytes=file_bytes,
+            filename=filename,
+            url_path=url_path,
+            raw_csv=raw_csv,
+            parsing=ParsingOptions(),
+        )
+
+        # Step 4: Call our new function to generate the preview from the DataFrame
+        preview_data = generate_preview(parsed.df)
+
+        return JSONResponse(status_code=200, content=preview_data)
+
+    except HTTPException:
+        # Re-raise known HTTP exceptions
+        raise
+    except Exception as e:
+        # Return a structured error for any other failures, matching the project's style
+        return JSONResponse(
+            status_code=400,
+            content={"errors": [{"message": str(e)}]},
+        )
\ No newline at end of file
diff --git a/app/services/processing.py b/app/services/processing.py
index 7c667c3..ac0ee57 100644
--- a/app/services/processing.py
+++ b/app/services/processing.py
@@ -337,3 +337,20 @@ def parse_input(
         return ParsedInput(df=df, source_name=filename)
 
     raise ValueError("No input provided: expected file upload, URL, or raw_csv text")
+
+def generate_preview(df: pd.DataFrame) -> dict:
+    """Generate a preview from a DataFrame.
+
+    Args:
+        df: The input DataFrame.
+
+    Returns:
+        A dictionary containing column names and the first 5 rows.
+    """
+    # Get the column names as a list
+    columns = df.columns.tolist()
+
+    # Get the first 5 rows and convert them to a list of dictionaries
+    rows = df.head(5).to_dict(orient="records")
+
+    return {"columns": columns, "rows": rows}
\ No newline at end of file