app.py

import gradio as gr
import requests
from urllib.parse import urlparse, urljoin
from bs4 import BeautifulSoup
import asyncio

# HTML and JavaScript for the "Copy Code" button
copy_button_html = """
<script>
function copyCode(textareaId) {
    const text = document.querySelector(`#${textareaId} textarea`).value;
    navigator.clipboard.writeText(text).then(() => {
        alert("Text copied to clipboard!");
    }).catch(() => {
        alert("Failed to copy text.");
    });
}
</script>
"""

# Common functions
def is_valid_url(url):
    """Checks if the string is a valid URL."""
    try:
        result = urlparse(url)
        return all([result.scheme, result.netloc])  # Check for scheme and domain
    except:
        return False

async def fetch_file_content(url):
    """Fetches the content of a file (CSS, JS, etc.) from a URL."""
    try:
        response = await asyncio.to_thread(requests.get, url, timeout=5)
        response.raise_for_status()
        return response.text
    except:
        return "Failed to fetch content."

# URL to Text Converter
async def extract_additional_resources(url):
    """Extracts links to CSS, JS, and images from HTML code."""
    try:
        response = await asyncio.to_thread(requests.get, url, timeout=5)
        response.raise_for_status()

        # Check if the content is HTML
        if 'text/html' in response.headers.get('Content-Type', ''):
            soup = BeautifulSoup(response.text, "html.parser")

            # Extract CSS links (limit to 5)
            css_links = [urljoin(url, link["href"]) for link in soup.find_all("link", rel="stylesheet") if "href" in link.attrs][:5]

            # Extract JS links (limit to 5)
            js_links = [urljoin(url, script["src"]) for script in soup.find_all("script") if "src" in script.attrs][:5]

            # Extract image links (limit to 5)
            img_links = [urljoin(url, img["src"]) for img in soup.find_all("img") if "src" in img.attrs][:5]

            # Fetch CSS and JS content asynchronously
            css_content = await asyncio.gather(*[fetch_file_content(link) for link in css_links])
            js_content = await asyncio.gather(*[fetch_file_content(link) for link in js_links])

            return css_links, js_links, img_links, css_content, js_content
        else:
            # If it's not HTML, treat it as a file
            return [], [], [], [response.text], []
    except Exception as e:
        return [], [], [], [], []

async def convert_to_text(url):
    # Handle view-source: URLs
    if url.startswith("view-source:"):
        url = url[len("view-source:"):]

    if not is_valid_url(url):
        return "Error: Please enter a valid URL.", "", None, [], [], [], [], []  # Return error message and empty data

    try:
        # Set headers to mimic a browser request
        headers = {
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"
        }
        response = await asyncio.to_thread(requests.get, url, headers=headers, timeout=5)
        response.raise_for_status()  # Check for HTTP errors (e.g., 404, 500)

        # Return results
        status = f"Request status: {response.status_code}"
        content_length = f"Content size: {len(response.text)} characters"
        results = f"{status}\n{content_length}"

        # Save text content to a file
        file_path = "downloaded_content.txt"
        with open(file_path, "w", encoding="utf-8") as file:
            file.write(response.text)

        # Extract additional resources
        css_links, js_links, img_links, css_content, js_content = await extract_additional_resources(url)

        return results, response.text, file_path, css_links, js_links, img_links, css_content, js_content
    except requests.exceptions.RequestException as e:
        return f"Error: {e}", "", None, [], [], [], [], []  # Return error message and empty data

# Model to Text Converter
async def fetch_model_info(model_url):
    """Fetches model description and installation instructions."""
    try:
        if "huggingface.co" in model_url:
            # Fetch model card from Hugging Face
            response = await asyncio.to_thread(requests.get, model_url, timeout=5)
            response.raise_for_status()
            soup = BeautifulSoup(response.text, "html.parser")

            # Extract model description
            description = soup.find("div", {"class": "prose"}).get_text(strip=True) if soup.find("div", {"class": "prose"}) else "No description available."

            # Generate installation instructions
            model_name = model_url.split("/")[-1]
            install_instructions = f"To install this model, run:\n```bash\npip install transformers\n```\nThen load the model in Python:\n```python\nfrom transformers import AutoModel, AutoTokenizer\nmodel = AutoModel.from_pretrained('{model_name}')\ntokenizer = AutoTokenizer.from_pretrained('{model_name}')\n```"

            return description, install_instructions
        elif "github.com" in model_url:
            # Fetch README from GitHub
            readme_url = f"{model_url}/raw/main/README.md"
            response = await asyncio.to_thread(requests.get, readme_url, timeout=5)
            response.raise_for_status()

            # Extract description from README
            description = response.text if response.text else "No description available."

            # Generate installation instructions
            install_instructions = f"To install this model, clone the repository:\n```bash\ngit clone {model_url}.git\ncd {model_url.split('/')[-1]}\n```"

            return description, install_instructions
        else:
            return "Unsupported repository.", ""
    except Exception as e:
        return f"Error: {e}", ""

async def fetch_model_file_content(model_url, file_path):
    """Fetches the content of a file from a model repository (Hugging Face or GitHub)."""
    try:
        # Construct the full URL to the file
        if "huggingface.co" in model_url:
            # Убираем /blob/main/ из URL, если он есть
            if "/blob/main/" in model_url:
                model_url = model_url.replace("/blob/main/", "/")
            # Hugging Face URL format: https://huggingface.co/{model}/raw/main/{file_path}
            full_url = f"{model_url}/raw/main/{file_path}"
        elif "github.com" in model_url:
            # GitHub URL format: https://github.com/{user}/{repo}/raw/main/{file_path}
            full_url = f"{model_url}/raw/main/{file_path}"
        else:
            return "Error: Unsupported repository."

        # Fetch the file content
        response = await asyncio.to_thread(requests.get, full_url, timeout=5)
        response.raise_for_status()
        return response.text
    except Exception as e:
        return f"Error: {e}"

# Space to Text Converter
async def fetch_space_file_content(space_url, file_path):
    """Fetches the content of a file from a Hugging Face Space."""
    try:
        # Construct the full URL to the file
        if "huggingface.co/spaces" in space_url:
            # Hugging Face Spaces URL format: https://huggingface.co/spaces/{user}/{space}/raw/main/{file_path}
            full_url = f"{space_url}/raw/main/{file_path}"
        else:
            return "Error: Unsupported repository. Please provide a Hugging Face Space URL."

        # Fetch the file content
        response = await asyncio.to_thread(requests.get, full_url, timeout=5)
        response.raise_for_status()
        return response.text
    except Exception as e:
        return f"Error: {e}"

# Create the Gradio interface
with gr.Blocks() as demo:
    gr.HTML(copy_button_html)  # Add the "Copy Code" script

    with gr.Tabs():
        # Tab 1: URL to Text Converter
        with gr.Tab("URL to Text Converter"):
            gr.Markdown("## URL to Text Converter")
            gr.Markdown("Enter a URL to fetch its text content and download it as a .txt file.")
            
            with gr.Row():
                url_input = gr.Textbox(label="Enter URL", placeholder="https://example.com or view-source:https://example.com")
            
            with gr.Row():
                results_output = gr.Textbox(label="Request Results", interactive=False)
                text_output = gr.Textbox(label="Text Content", interactive=True, elem_id="output-text")
            
            with gr.Row():
                gr.HTML("<button onclick='copyCode(\"output-text\")'>Copy Code</button>")  # Add the "Copy Code" button
                file_output = gr.File(label="Download File", visible=False)  # Hidden file download component
            
            submit_button = gr.Button("Fetch Content")
            submit_button.click(
                fn=convert_to_text,
                inputs=url_input,
                outputs=[
                    results_output, text_output, file_output,
                    gr.Textbox(label="CSS Files"), gr.Textbox(label="JS Files"), gr.Textbox(label="Images"),
                    gr.Textbox(label="CSS Content"), gr.Textbox(label="JS Content")
                ]
            )

            # Add an Accordion to show/hide additional resources
            with gr.Accordion("Show/Hide Additional Resources", open=False):
                gr.Markdown("### CSS Files")
                css_output = gr.Textbox(label="CSS Files", interactive=False)
                
                gr.Markdown("### JS Files")
                js_output = gr.Textbox(label="JS Files", interactive=False)
                
                gr.Markdown("### Images")
                img_output = gr.Textbox(label="Images", interactive=False)
                
                gr.Markdown("### CSS Content")
                css_content_output = gr.Textbox(label="CSS Content", interactive=True)
                
                gr.Markdown("### JS Content")
                js_content_output = gr.Textbox(label="JS Content", interactive=True)

        # Tab 2: Model to Text Converter
        with gr.Tab("Model to Text Converter"):
            gr.Markdown("## Model to Text Converter")
            gr.Markdown("Enter a link to a model on Hugging Face or GitHub, and specify the file path.")
            
            with gr.Row():
                model_url_input = gr.Textbox(label="Model URL", placeholder="https://huggingface.co/... or https://github.com/...")
                file_path_input = gr.Textbox(label="File Path", placeholder="e.g., config.json or README.md")
            
            with gr.Row():
                model_description_output = gr.Textbox(label="Model Description", interactive=False)
                install_instructions_output = gr.Textbox(label="Installation Instructions", interactive=False)
            
            with gr.Row():
                model_content_output = gr.Textbox(label="File Content", interactive=True, elem_id="model-content-output")
            
            with gr.Row():
                gr.HTML("<button onclick='copyCode(\"model-content-output\")'>Copy Code</button>")  # Add the "Copy Code" button
            
            submit_model_button = gr.Button("Fetch Model Info and File Content")
            submit_model_button.click(
                fn=fetch_model_info,
                inputs=[model_url_input],
                outputs=[model_description_output, install_instructions_output]
            )
            submit_model_button.click(
                fn=fetch_model_file_content,
                inputs=[model_url_input, file_path_input],
                outputs=[model_content_output]
            )

        # Tab 3: Space to Text Converter
        with gr.Tab("Space to Text Converter"):
            gr.Markdown("## Space to Text Converter")
            gr.Markdown("Enter a link to a Hugging Face Space and specify the file path to fetch its content.")
            
            with gr.Row():
                space_url_input = gr.Textbox(label="Space URL", placeholder="https://huggingface.co/spaces/...")
                space_file_path_input = gr.Textbox(label="File Path", placeholder="e.g., app.py or README.md")
            
            with gr.Row():
                space_content_output = gr.Textbox(label="File Content", interactive=True, elem_id="space-content-output")
            
            with gr.Row():
                gr.HTML("<button onclick='copyCode(\"space-content-output\")'>Copy Code</button>")  # Add the "Copy Code" button
            
            submit_space_button = gr.Button("Fetch File Content")
            submit_space_button.click(
                fn=fetch_space_file_content,
                inputs=[space_url_input, space_file_path_input],
                outputs=[space_content_output]
            )

# Launch the interface
demo.launch()