add delete_rows endpoint

.gitignore

@@ -1,2 +1,3 @@
 *pycache*
-*.env*
+*.env*
+.python-version

src/api/models/embedding_models.py

@@ -42,6 +42,11 @@ class DeleteEmbeddingRequest(BaseModel):
     dataset_name: str
 
 
+class DeleteRowsRequest(BaseModel):
+    dataset_name: str
+    product_types_to_delete: List[str]
+
+
 # Request model for the /embed endpoint
 class EmbedRequest(BaseModel):
     texts: List[str]  # List of strings to generate embeddings for

src/api/services/huggingface_service.py

@@ -100,3 +100,29 @@ class HuggingFaceService:
         except Exception as e:
             logger.error(f"Failed to delete dataset: {e}")
             raise DatasetDeleteError(f"Failed to delete dataset: {e}")
+
+
+    async def delete_rows_from_dataset(self, dataset_name: str, product_types_to_delete: List[str]):
+        """
+        Loads a dataset, filters out rows based on a list of product types, and pushes it back.
+        """
+        if not product_types_to_delete:
+            return
+
+        # Step 1: Load the existing dataset
+        logger.info(f"Loading dataset {dataset_name} to delete rows.")
+        dataset = await self.read_dataset(dataset_name)
+
+        # Step 2 : Filter the dataset to EXCLUDE the rows with the given product_types
+        logger.info(f"Filtering out product_types: {product_types_to_delete}")
+        initial_row_count = len(dataset)
+
+        filtered_dataset = dataset.filter(lambda product: product['product_type'] not in product_types_to_delete)
+
+        final_row_count = len(filtered_dataset)
+        logger.info(f"{initial_row_count - final_row_count} rows deleted.")
+
+        # Step 3 : Push the modified dataset back to the hub
+        await self.push_to_hub(filtered_dataset, dataset_name)
+
+        return filtered_dataset

src/main.py

@@ -9,6 +9,7 @@ from src.api.models.embedding_models import (
     CreateEmbeddingRequest,
     ReadEmbeddingRequest,
     UpdateEmbeddingRequest,
+    DeleteRowsRequest,
     DeleteEmbeddingRequest,
     EmbedRequest,
     SearchEmbeddingRequest,
@@ -230,6 +231,29 @@ async def delete_embeddings(
         raise HTTPException(status_code=500, detail=f"An error occurred: {e}")
 
 
+@app.post("/delete_rows")
+async def delete_rows(
+    request: DeleteRowsRequest,
+    huggingface_service: HuggingFaceService = Depends(get_huggingface_service)
+):
+    """
+    Deletes specific rows from a Hugging Face dataset based on their product_types"""
+    try:
+        await huggingface_service.delete_rows_from_dataset(
+            request.dataset_name, request.product_types_to_delete
+        )
+        return {
+            "message": "Rows deleted succesfully from dataset.",
+            "dataset_name": request.dataset_name,
+            "deleted_product_types": request.product_types_to_delete,
+        }
+    except DatasetNotFoundError as e:
+        raise HTTPException(status_code=404, detail=str(e))
+    except Exception as e:
+        logger.error(f"An error occured while deleting rows: {e}")
+        raise HTTPException(status_code=500, detail=str(e))
+
+
 @app.post("/search_embeddings")
 async def search_embedding(
     request: SearchEmbeddingRequest,