repo_to_md / app.py
broadfield-dev's picture
Update app.py
d6d177a verified
raw
history blame
17.3 kB
from flask import Flask, render_template, request, jsonify, send_file
import requests
import base64
import markdown
import json
import mimetypes
import os
import io
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."""
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="", 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, headers={"Accept": "application/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))
return files
except Exception as e:
return None
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
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()
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
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']
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_content
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'
)
html_template = """
<!DOCTYPE html>
<html>
<head>
<title>Repo & Files to Markdown</title>
<style>
body {
font-family: Arial, sans-serif;
margin: 20px;
max-width: 1200px;
margin: 0 auto;
}
.container {
padding: 20px;
}
textarea {
width: 100%;
height: 400px;
margin-top: 20px;
font-family: monospace;
}
button {
padding: 10px 20px;
background-color: #4CAF50;
color: white;
border: none;
cursor: pointer;
margin: 5px;
}
button:hover {
background-color: #45a049;
}
#output {
margin-top: 20px;
border: 1px solid #ddd;
padding: 20px;
background-color: #f9f9f9;
}
.spinner {
display: none;
border: 4px solid #f3f3f3;
border-top: 4px solid #3498db;
border-radius: 50%;
width: 30px;
height: 30px;
animation: spin 1s linear infinite;
margin: 20px auto;
}
@keyframes spin {
0% { transform: rotate(0deg); }
100% { transform: rotate(360deg); }
}
</style>
</head>
<body>
<div class="container">
<h1>Repository & Files to Markdown Converter</h1>
<p>Enter a GitHub/Hugging Face Space URL (e.g., https://huggingface.co/spaces/username/space)</p>
<input type="text" id="repoUrl" style="width: 100%; padding: 8px;" placeholder="Enter GitHub or Hugging Face Space URL">
<p>OR upload files (select multiple files or a folder - folder upload supported in Chrome)</p>
<input type="file" id="fileInput" multiple webkitdirectory style="margin: 10px 0;">
<br>
<button onclick="processRepo()">Convert URL</button>
<button onclick="processFiles()">Convert Files</button>
<button id="downloadBtn" style="display: none;" onclick="downloadMarkdown()">Download .md</button>
<div id="spinner" class="spinner"></div>
<h2>Markdown Output:</h2>
<textarea id="markdownOutput" readonly></textarea>
<h2>Preview:</h2>
<div id="output"></div>
</div>
<script>
let currentMarkdown = '';
let currentFilename = '';
async function processRepo() {
const repoUrl = document.getElementById('repoUrl').value;
await processContent('/process', { repo_url: repoUrl });
}
async function processFiles() {
const files = document.getElementById('fileInput').files;
if (files.length === 0) {
alert('Please select at least one file or folder');
return;
}
const formData = new FormData();
for (let file of files) {
formData.append('files[]', file);
}
await processContent('/process', formData, false);
}
async function processContent(url, data, isJson = true) {
const spinner = document.getElementById('spinner');
const buttons = document.querySelectorAll('button');
spinner.style.display = 'block';
buttons.forEach(btn => btn.disabled = true);
try {
const options = {
method: 'POST',
...(isJson ? {
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify(data)
} : { body: data })
};
const response = await fetch(url, options);
const result = await response.json();
if (result.error) {
alert(result.error);
return;
}
currentMarkdown = result.markdown;
currentFilename = result.filename;
document.getElementById('markdownOutput').value = result.markdown;
document.getElementById('output').innerHTML = result.html;
document.getElementById('downloadBtn').style.display = 'inline-block';
} catch (error) {
alert('An error occurred: ' + error.message);
} finally {
spinner.style.display = 'none';
buttons.forEach(btn => btn.disabled = false);
}
}
async function downloadMarkdown() {
try {
const response = await fetch('/download', {
method: 'POST',
headers: {
'Content-Type': 'application/json',
},
body: JSON.stringify({
markdown: currentMarkdown,
filename: currentFilename
})
});
const blob = await response.blob();
const url = window.URL.createObjectURL(blob);
const a = document.createElement('a');
a.href = url;
a.download = currentFilename;
document.body.appendChild(a);
a.click();
a.remove();
window.URL.revokeObjectURL(url);
} catch (error) {
alert('Error downloading file: ' + error.message);
}
}
</script>
</body>
</html>
"""
if not os.path.exists('templates'):
os.makedirs('templates')
with open('templates/index.html', 'w') as f:
f.write(html_template)
if __name__ == '__main__':
app.run(host="0.0.0.0", port=7860, debug=True)