update app.py

app.py

@@ -1,33 +1,19 @@
 import gradio as gr
 from datasets import load_dataset, Dataset
-from huggingface_hub import hf_hub_download
 import os
 
-# Define the dataset name and a split to load
-DATASET_NAME = "wikides" # The actual dataset is likely "wikides" or a related variant.
-# We'll use a local file for demonstration since the "wikides" dataset isn't directly loadable with load_dataset()
-# For a real application, you'd use a known, public Hugging Face dataset.
-# The user might need to upload a wikides file.
-# Let's assume a dummy local file for the sake of the API demonstration.
-# In a real-world scenario, you would use `load_dataset("wikides", split="train")`
-# However, the search results indicate that the dataset is not a simple direct load.
-# So we'll use a simplified example to demonstrate the API functionality.
-# Assuming a local dummy CSV file named 'wikides_data.csv' exists.
-# We'll create a dummy file for the demonstration.
+# Define the dataset name and split
+DATASET_NAME = "Thang/wikides"
+SPLIT = "train"
 
-# Create a dummy CSV file for demonstration purposes
-if not os.path.exists("wikides_data.csv"):
-    with open("wikides_data.csv", "w") as f:
-        f.write("id,title,text\n")
-        for i in range(100):
-            f.write(f"{i},Title {i},This is the text for article {i} from WikiDES.\n")
-
-# Load the dataset from the local file
-# In a real scenario you would have the actual dataset files.
+# Load the actual dataset from Hugging Face Hub
 try:
-    dataset = load_dataset("csv", data_files="wikides_data.csv", split="train")
+    dataset = load_dataset(DATASET_NAME, split=SPLIT)
 except Exception as e:
-    raise RuntimeError(f"Failed to load dataset: {e}")
+    # If the dataset fails to load, raise a more descriptive error.
+    # This will cause the Hugging Face Space to fail during startup,
+    # which is the desired behavior for a critical dependency.
+    raise RuntimeError(f"Failed to load dataset '{DATASET_NAME}': {e}")
 
 # Function to fetch data with a limit
 def get_data_with_limit(limit: int = 50):
@@ -35,6 +21,7 @@ def get_data_with_limit(limit: int = 50):
     Fetches data from the dataset with a specified limit.
     The limit defaults to 50 if not provided or if it's invalid.
     """
+    # Type and value checking for the limit parameter
     if not isinstance(limit, int) or limit < 1:
         limit = 50
     
@@ -44,16 +31,13 @@ def get_data_with_limit(limit: int = 50):
         limit = num_rows
 
     try:
-        # Slice the dataset to get the required number of rows
+        # Slice the dataset to get the required number of rows and convert to a dictionary
         return dataset.select(range(limit)).to_dict()
     except Exception as e:
-        # Return a structured error response
+        # Return a structured error response if something goes wrong during data retrieval
         return {"error": "Internal Server Error", "message": str(e)}
 
 # Create the Gradio interface
-# We use `gr.Interface` to define the API endpoint.
-# The `inputs` and `outputs` define the schema.
-# We set `api_name="data"` to expose the function at the `/data` endpoint.
 demo = gr.Interface(
     fn=get_data_with_limit,
     inputs=[
@@ -63,7 +47,7 @@ demo = gr.Interface(
         gr.JSON(label="Data")
     ],
     title="WikiDES Data API",
-    description="A simple API to access data from the WikiDES dataset with a configurable limit. The default limit is 50.",
+    description=f"A simple API to access data from the **{DATASET_NAME}** dataset with a configurable limit. The default limit is 50.",
     examples=[
         [10],
         [25],