Update app.py

app.py

@@ -1,335 +1,49 @@
-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
-import uuid
-from pathlib import Path
-
-app = Flask(__name__)
-
-GITHUB_API = "https://api.github.com/repos/"
-#HF_API = "https://huggingface.co/api/spaces/"
-
-def generate_file_tree(paths):
-    """Generate a simple file tree from a list of paths."""
-    print("generating file tree") 
-    tree = ["📁 Root"]
-    sorted_paths = sorted(paths)
-    for path in sorted_paths:
-        parts = path.split('/')
-        indent = "  " * (len(parts) - 1)
-        tree.append(f"{indent}📄 {parts[-1]}")
-    print("generating file tree - Complete") 
-    
-    return "\n".join(tree) + "\n\n"
-
-def get_all_files(owner, repo, path="", is_hf=False):
-    """Recursively fetch all files from a repository."""
-    if is_hf:
-        # Attempt to fetch file list from Hugging Face Space (publicly accessible files)
-        api_url = f"https://huggingface.co/spaces/{owner}/{repo}/tree/main/{path}".rstrip('/')
-    else:
-        api_url = f"{GITHUB_API}{owner}/{repo}/contents/{path}".rstrip('/')
-    
-    try:
-        response = requests.get(api_url)
-        print(response.content)
-        response = requests.get(api_url, headers={"Accept": "application/json"})
-        print(str(response.json()))
-        response.raise_for_status()
-        items = response.json()
-        
-        # Hugging Face might not return JSON in the same format; adjust if HTML is returned
-        if isinstance(items, str):  # If response isn’t JSON, it’s likely HTML
-            return None  # Fallback to error handling
-        
-        files = []
-        for item in items:
-            if item['type'] == 'file':
-                files.append(item)
-            elif item['type'] == 'dir':
-                files.extend(get_all_files(owner, repo, item['path'], is_hf))
-        print(files)
-        return files
-        
-    except Exception as e:
-        return None
-
-def get_hf_files(repo,name,path=""):
+def get_hf_files(repo, name, path=""):
     api = HfApi()
-    f_ist = (api.list_repo_files(repo_id=f'{repo}/{name}', repo_type="space"))
-    print (f_ist)
-    file_list = []
-    #file_out = []
+    file_list = api.list_repo_files(repo_id=f'{repo}/{name}', repo_type="space")
+    print(f"Files in {repo}/{name}: {file_list}")
+    processed_files = []
     
+    # Create base directory if it doesn't exist
     if not os.path.exists(name):
-           os.makedirs(name)
-    for d_app in f_ist:
-        if "/" in d_app:
-            dir_1=d_app.split("/",1)[0]
-            file_list.extend(get_hf_files(repo,name,dir_1))
-            '''
-            rem_1=d_app.split("/",1)[1]
-            if not os.path.exists(f'{name}/{dir_1}'):
-                   os.makedirs(f'{name}/{dir_1}')
-            
-            if "/" in rem_1:
-                dir_2=rem_1.split("/",1)[0]
-                rem_2=rem_1.split("/",1)[1]
-                if not os.path.exists(f'{name}/{dir_1}/{dir_2}'):
-                       os.makedirs(f'{name}/{dir_1}/{dir_2}')                
-                sf=rem_2.split(".",1)[1]
-                pf=rem_2.split(".",1)[0]
-                f_name=f'{dir_1}/{dir2}/{pf}.{sf}'
-            
-            else:
-                sf=rem_1.split(".",1)[1]
-                pf=rem_1.split(".",1)[0]
-                f_name=f'{dir_1}/{pf}.{sf}'
-            print(f_name)
-            '''
+        os.makedirs(name)
+
+    for file_path in file_list:
+        # Handle nested directories
+        if "/" in file_path:
+            # Split into directory and remainder
+            dir_part, file_part = file_path.split("/", 1)
+            # Ensure directory exists
+            dir_path = os.path.join(name, dir_part)
+            if not os.path.exists(dir_path):
+                os.makedirs(dir_path)
+            # Recursively handle subdirectories if needed
+            if "/" in file_part:
+                processed_files.extend(get_hf_files(repo, name, dir_part))
+                continue
+        
+        # Safely split filename into prefix and extension
+        filename = os.path.basename(file_path)
+        if "." in filename:
+            pf, sf = filename.rsplit(".", 1)  # Safely split on last period
+            f_name = f"{pf}.{sf}"
         else:
-    
-            sf=d_app.split(".",1)[1]
-            pf=d_app.split(".",1)[0]
-            f_name=f'{pf}.{sf}'            
-            pass
-        
-        r = requests.get(f'https://huggingface.co/spaces/{repo}/{name}/raw/main/{d_app}')
-        print(d_app)
-        #print (r.text)
-        uid = uuid.uuid4()
-        file = open(f'{name}/{f_name}','wb')
-        file.write(r.content)
-        file.close()
-               
-        file_list.append({"path":f'{d_app}'})
-    print(file_list)
-    return file_list
-
-def get_repo_contents(url):
-    """Parse URL and fetch repository contents."""
-    try:
-        if "huggingface.co" in url:
-            parts = url.rstrip('/').split('/')
-            owner, repo = parts[-2], parts[-1]
-            # Fallback approach: manually fetch known files or use a simpler file list
-            # For now, assume a flat structure and fetch known files directly
-            # This is a workaround until a proper API token or endpoint is confirmed
-            known_files = [
-                {'path': 'app.py', 'type': 'file'},
-                {'path': 'README.md', 'type': 'file'}
-                # Add more known paths or implement HTML scraping if needed
-            ]
-            #files = get_all_files(owner, repo, "", True) or known_files
-            files = get_hf_files(owner, repo)
-            return owner, repo, files, True
-        else:  # Assume GitHub URL
-            parts = url.rstrip('/').split('/')
-            owner, repo = parts[-2], parts[-1]
-            files = get_all_files(owner, repo, "", False)
-            if files is None:
-                raise Exception("Failed to fetch GitHub repository contents")
-            return owner, repo, files, False
-    except Exception as 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}"
-        else:
-            file_url = f"{GITHUB_API}{owner}/{repo}/contents/{file_path}"
-        
-        response = requests.get(file_url)
-        response.raise_for_status()
+            pf = filename
+            sf = ""
+            f_name = pf
         
-        if is_hf:
-            content_raw = response.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"
-        else:  # GitHub
-            data = response.json()
-            if 'content' in data:
-                content_raw = base64.b64decode(data['content'])
-                size = data['size']
-                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"
-            else:
-                content = f"### File: {file_path}\n[No content available]\n\n"
-    except Exception as e:
-        content = f"### File: {file_path}\n[Error fetching file content: {str(e)}]\n\n"
-    
-    return content
+        # Construct full local path
+        local_path = os.path.join(name, file_path)
+        # Ensure subdirectory exists for nested files
+        os.makedirs(os.path.dirname(local_path), exist_ok=True)
 
-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()  # Read file content into memory
-        size = len(content_raw)    # Compute size in bytes
-        
-        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']
+        # Download file content
+        r = requests.get(f'https://huggingface.co/spaces/{repo}/{name}/raw/main/{file_path}')
+        print(f"Downloading: {file_path}")
+        with open(local_path, 'wb') as file:
+            file.write(r.content)
         
-        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:  # Handle uploaded files
-        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():
-    # Ensure consistent response structure
-    response_data = {'markdown': '', 'html': '', 'filename': '', 'error': None}
-    
-    if 'files[]' in request.files:
-        files = request.files.getlist('files[]')
-        if not 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')
-        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:
-            response_data['error'] = markdown_content  # Error message from get_repo_contents
-            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')
-    
-    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'
-    )
-with open("html_template.html", "r") as f:
-    html_template=f.read()
-f.close()
-
-
-if not os.path.exists('templates'):
-    os.makedirs('templates')
-with open('templates/index.html', 'w') as f:
-    f.write(html_template)
+        processed_files.append({"path": file_path})
 
-if __name__ == '__main__':
-    app.run(host="0.0.0.0", port=7860, debug=True)
+    print(f"Processed files: {processed_files}")
+    return processed_files