app.py

import requests
import gradio as gr
import json
import base64
import re

# GitHub Configuration
GITHUB_TOKEN = "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"  # Replace with your token
REPO_NAME = "arssite/CBTLLM"  # Replace with your repo name
FILE_PATH = "url.json"  # Path to the file in your repo
BRANCH = "main" 

# Regex for URL Validation
URL_REGEX = re.compile(
    r'^(https?://)?'  # Protocol (optional)
    r'([a-zA-Z0-9.-]+)'  # Domain
    r'(\.[a-zA-Z]{2,})'  # Top-level domain
    r'(:[0-9]{1,5})?'  # Port (optional)
    r'(/.*)?$',  # Path (optional)
    re.IGNORECASE
)

# Function to validate URLs
def validate_urls(urls):
    valid_urls = []
    for url in urls:
        url = url.strip()  # Remove extra spaces
        if re.match(URL_REGEX, url):
            valid_urls.append(url if url.startswith("http") else "http://" + url)  # Ensure URL has a protocol
    return valid_urls

# Fetch the file content from GitHub
def fetch_file_content():
    url = f"https://api.github.com/repos/{REPO_NAME}/contents/{FILE_PATH}"
    headers = {"Authorization": f"token {GITHUB_TOKEN}"}
    response = requests.get(url, headers=headers)
    if response.status_code == 200:
        content = response.json()
        file_content = base64.b64decode(content['content']).decode('utf-8')
        sha = content['sha']
        return json.loads(file_content), sha
    else:
        raise Exception(f"Error fetching file: {response.status_code} - {response.text}")

# Update the file on GitHub
def update_file_on_github(new_urls):
    # Fetch current content and SHA
    current_content, sha = fetch_file_content()

    # Parse, validate, and add unique URLs
    input_urls = validate_urls(new_urls.split("\n"))  # Validate input URLs
    if not input_urls:
        raise ValueError("No valid URLs provided. Please enter valid URLs.")

    updated_urls = list(set(current_content) | set(input_urls))  # Merge current and input URLs, ensuring uniqueness

    # Prepare updated content
    updated_content = json.dumps(updated_urls, indent=2)
    encoded_content = base64.b64encode(updated_content.encode('utf-8')).decode('utf-8')

    # Push updated content
    url = f"https://api.github.com/repos/{REPO_NAME}/contents/{FILE_PATH}"
    headers = {"Authorization": f"token {GITHUB_TOKEN}"}
    data = {
        "message": "Update URLs",
        "content": encoded_content,
        "sha": sha,
        "branch": BRANCH
    }
    response = requests.put(url, headers=headers, json=data)
    if response.status_code == 200:
        return len(input_urls), len(updated_urls)
    else:
        raise Exception(f"Error updating file: {response.status_code} - {response.text}")

# Gradio Interface
def submit_urls(new_urls):
    try:
        if not new_urls.strip():
            return "No URLs provided. Please enter at least one URL."
        
        user_count, total_count = update_file_on_github(new_urls)
        total_count+=108
        return (
            f"Thanks for your contribution!\n"
            f"You added {user_count} unique URLs.\n"
            f"The file now contains {total_count} unique URLs."
        )
    except ValueError as e:
        return str(e)
    except Exception as e:
        return f"An error occurred: {str(e)}"

def resubmit():
    return gr.update(value="")  # Clears the input area for a new submission

# Gradio UI
with gr.Blocks() as app:
    with gr.Row():
        gr.Markdown("## URLink Submission Program For CBT LLM ")
        gr.Markdown("Requesting To contribute with Link or Url of Any News Article, Blog, Case Study, EBook, Any Source Related to  Cognitive Behavioral.")
    with gr.Row():
        input_area = gr.Textbox(
            label="Enter URLs ( try one per line):",
            placeholder="https://example1.com\nhttps://example2.com\n...",
            lines=5
        )
    with gr.Row():
        submit_button = gr.Button("Submit")
    with gr.Row():
        output_area = gr.Textbox(
            label="Response ",
            interactive=False
        )
    with gr.Row():
        resubmit_button = gr.Button("Resubmit")

    # Define functionality
    submit_button.click(submit_urls, inputs=input_area, outputs=output_area)
    resubmit_button.click(resubmit, inputs=None, outputs=input_area)

# Launch the app
app.launch()