app.py

from flask import Flask, render_template, request, jsonify, send_file
from huggingface_hub import HfApi
import requests
import base64
import markdown
import json
import mimetypes
import os
import io
from pathlib import Path

app = Flask(__name__)

GITHUB_API = "https://api.github.com/repos/"

def generate_file_tree(paths):
    """Generate a simple file tree from a list of paths."""
    tree = ["📁 Root"]
    sorted_paths = sorted(paths)
    for path in sorted_paths:
        parts = path.split('/')
        indent = "  " * (len(parts) - 1)
        tree.append(f"{indent}📄 {parts[-1]}")
    return "\n".join(tree) + "\n\n"

def get_all_files(owner, repo, path=""):
    """Recursively fetch all files from a GitHub repository."""
    api_url = f"{GITHUB_API}{owner}/{repo}/contents/{path}".rstrip('/')
    try:
        response = requests.get(api_url, headers={"Accept": "application/json"}, timeout=10)
        response.raise_for_status()
        content_type = response.headers.get('Content-Type', '')
        if not content_type.startswith('application/json'):
            print(f"Error: Received non-JSON response from GitHub for {api_url}: {response.text[:100]}...")
            return None
        try:
            items = response.json()
        except json.JSONDecodeError as e:
            print(f"Error: Failed to parse JSON from GitHub for {api_url}: {str(e)}. Response: {response.text[:100]}...")
            return None
        files = []
        for item in items:
            if isinstance(item, dict) and item.get('type') == 'file':
                files.append(item)
            elif isinstance(item, dict) and item.get('type') == 'dir':
                sub_files = get_all_files(owner, repo, item['path'])
                if sub_files:
                    files.extend(sub_files)
        return files
    except requests.RequestException as e:
        print(f"Error fetching GitHub contents from {api_url}: {str(e)}")
        return None

def get_hf_files(owner, repo):
    """Fetch all files from a Hugging Face Space."""
    api = HfApi()
    repo_id = f"{owner}/{repo}"
    try:
        file_list = api.list_repo_files(repo_id=repo_id, repo_type="space")
        print(f"Files in HF Space {repo_id}: {file_list}")
    except Exception as e:
        print(f"Error listing files for HF Space {repo_id}: {str(e)}")
        return []
    
    processed_files = []
    local_dir = repo
    if not os.path.exists(local_dir):
        os.makedirs(local_dir)

    for file_path in file_list:
        raw_url = f"https://huggingface.co/spaces/{owner}/{repo}/raw/main/{file_path}"
        try:
            response = requests.get(raw_url, timeout=10)
            response.raise_for_status()
            content_type = response.headers.get('Content-Type', '')
            if content_type.startswith('text/html'):
                print(f"Error: Received HTML instead of raw content for {raw_url}: {response.text[:100]}...")
                continue
            local_path = os.path.join(local_dir, file_path)
            os.makedirs(os.path.dirname(local_path), exist_ok=True)
            with open(local_path, 'wb') as file:
                file.write(response.content)
            processed_files.append({"path": file_path})
        except requests.RequestException as e:
            print(f"Error downloading {file_path} from {raw_url}: {str(e)}")
    
    print(f"Processed files from HF Space {repo_id}: {processed_files}")
    return processed_files

def get_repo_contents(url):
    """Parse URL and fetch repository contents."""
    try:
        parts = url.rstrip('/').split('/')
        owner, repo = parts[-2], parts[-1]
        if "huggingface.co" in url.lower():
            files = get_hf_files(owner, repo)
            if not files and files is not None:  # Empty list is valid, None indicates error
                raise Exception("No files found in the Hugging Face Space")
            return owner, repo, files, True
        else:  # Assume GitHub URL
            files = get_all_files(owner, repo)
            if files is None:
                raise Exception("Failed to fetch GitHub repository contents")
            return owner, repo, files, False
    except Exception as e:
        print(f"Error processing URL {url}: {str(e)}")
        return None, None, f"Error fetching repo contents: {str(e)}", False

def process_file_content(file_info, owner, repo, is_hf=False):
    """Process individual file content from a repository."""
    content = ""
    file_path = file_info['path']
    try:
        if is_hf:
            file_url = f"https://huggingface.co/spaces/{owner}/{repo}/raw/main/{file_path}"
            response = requests.get(file_url, timeout=10)
            response.raise_for_status()
            content_raw = response.content
            content_type = response.headers.get('Content-Type', '')
            if content_type.startswith('text/html'):
                raise Exception(f"Received HTML instead of raw content: {response.text[:100]}...")
        else:
            file_url = f"{GITHUB_API}{owner}/{repo}/contents/{file_path}"
            response = requests.get(file_url, headers={"Accept": "application/json"}, timeout=10)
            response.raise_for_status()
            data = response.json()
            if 'content' not in data:
                raise Exception("No content available in GitHub API response")
            content_raw = base64.b64decode(data['content'])

        size = len(content_raw)
        file_extension = file_path.split('.')[-1] if '.' in file_path else ''
        mime_type, _ = mimetypes.guess_type(file_path)
        is_text = (mime_type and mime_type.startswith('text')) or file_extension in ['py', 'md', 'txt', 'js', 'html', 'css', 'json']

        if is_text:
            try:
                text_content = content_raw.decode('utf-8')
                if file_extension == 'json':
                    try:
                        json_data = json.loads(text_content)
                        formatted_json = json.dumps(json_data, indent=2)
                        content = f"### File: {file_path}\n```json\n{formatted_json}\n```\n\n"
                    except json.JSONDecodeError:
                        content = f"### File: {file_path}\n```json\n{text_content}\n```\n[Note: Invalid JSON format]\n\n"
                else:
                    content = f"### File: {file_path}\n```{file_extension or 'text'}\n{text_content}\n```\n\n"
            except UnicodeDecodeError:
                content = f"### File: {file_path}\n[Binary file - {size} bytes]\n\n"
        else:
            content = f"### File: {file_path}\n[Binary file - {size} bytes]\n\n"
    except Exception as e:
        content = f"### File: {file_path}\n[Error fetching content: {str(e)}]\n\n"
    return content

def process_uploaded_file(file):
    """Process uploaded file content."""
    content = ""
    filename = file.filename
    file_extension = filename.split('.')[-1] if '.' in filename else ''
    try:
        content_raw = file.read()
        size = len(content_raw)
        mime_type, _ = mimetypes.guess_type(filename)
        is_text = (mime_type and mime_type.startswith('text')) or file_extension in ['py', 'md', 'txt', 'js', 'html', 'css', 'json']
        
        if is_text:
            try:
                text_content = content_raw.decode('utf-8')
                if file_extension == 'json':
                    try:
                        json_data = json.loads(text_content)
                        formatted_json = json.dumps(json_data, indent=2)
                        content = f"### File: {filename}\n```json\n{formatted_json}\n```\n\n"
                    except json.JSONDecodeError:
                        content = f"### File: {filename}\n```json\n{text_content}\n```\n[Note: Invalid JSON format]\n\n"
                else:
                    content = f"### File: {filename}\n```{file_extension or 'text'}\n{text_content}\n```\n\n"
            except UnicodeDecodeError:
                content = f"### File: {filename}\n[Binary file - {size} bytes]\n\n"
        else:
            content = f"### File: {filename}\n[Binary file - {size} bytes]\n\n"
    except Exception as e:
        content = f"### File: {filename}\n[Error processing file: {str(e)}]\n\n"
    return content

def create_markdown_document(url=None, files=None):
    """Create markdown document from repo contents or uploaded files."""
    if url:
        owner, repo, contents, is_hf = get_repo_contents(url)
        if isinstance(contents, str):  # Error case
            return f"Error: {contents}"
        markdown_content = f"# {'Space' if is_hf else 'Repository'}: {owner}/{repo}\n\n"
        markdown_content += "## File Structure\n```\n"
        markdown_content += generate_file_tree([item['path'] for item in contents])
        markdown_content += "```\n\n"
        markdown_content += f"Below are the contents of all files in the {'space' if is_hf else 'repository'}:\n\n"
        for item in contents:
            markdown_content += process_file_content(item, owner, repo, is_hf)
    else:
        markdown_content = "# Uploaded Files\n\n"
        markdown_content += "## File Structure\n```\n"
        markdown_content += generate_file_tree([file.filename for file in files])
        markdown_content += "```\n\n"
        markdown_content += "Below are the contents of all uploaded files:\n\n"
        for file in files:
            markdown_content += process_uploaded_file(file)
    return markdown_content

@app.route('/')
def index():
    return render_template('index.html')

@app.route('/process', methods=['POST'])
def process():
    response_data = {'markdown': '', 'html': '', 'filename': '', 'error': None}
    
    if 'files[]' in request.files:
        files = request.files.getlist('files[]')
        if not files or all(not file.filename for file in files):
            response_data['error'] = 'No files uploaded'
            return jsonify(response_data), 400
        markdown_content = create_markdown_document(files=files)
        response_data['markdown'] = "```markdown\n" + markdown_content + "\n```"
        response_data['html'] = markdown.markdown(markdown_content)
        response_data['filename'] = "uploaded_files_summary.md"
    else:
        repo_url = request.json.get('repo_url', '').strip()
        if not repo_url:
            response_data['error'] = 'Please provide a repository URL or upload files'
            return jsonify(response_data), 400
        markdown_content = create_markdown_document(repo_url)
        owner, repo, contents, is_hf = get_repo_contents(repo_url)
        if not owner:  # Error case
            response_data['error'] = markdown_content
            return jsonify(response_data), 400
        response_data['markdown'] = markdown_content
        response_data['html'] = markdown.markdown(markdown_content)
        response_data['filename'] = f"{owner}_{repo}_summary.md"
    
    return jsonify(response_data)

@app.route('/download', methods=['POST'])
def download():
    markdown_content = request.json.get('markdown', '')
    filename = request.json.get('filename', 'document.md')
    buffer = io.BytesIO()
    buffer.write(markdown_content.encode('utf-8'))
    buffer.seek(0)
    return send_file(
        buffer,
        as_attachment=True,
        download_name=filename,
        mimetype='text/markdown'
    )

if __name__ == '__main__':
    # Ensure templates directory and index.html exist
    if not os.path.exists('templates'):
        os.makedirs('templates')
    # You’ll need to provide your own index.html or use a placeholder
    if not os.path.exists('templates/index.html'):
        with open('templates/index.html', 'w') as f:
            f.write('<!DOCTYPE html><html><body><h1>Repo to Markdown</h1><p>Use POST /process to convert repo contents.</p></body></html>')
    app.run(host="0.0.0.0", port=7860, debug=True)