import streamlit as st # Must be the first Streamlit command st.set_page_config(page_title="Advanced File Downloader", layout="wide") import os import subprocess import sys def install_playwright_dependencies(): try: # Use apt-get directly without sudo os.system('apt-get update -y') # Install required dependencies dependencies = [ 'libnss3', 'libnspr4', 'libatk1.0-0', 'libatk-bridge2.0-0', 'libcups2', 'libxcomposite1', 'libxdamage1', 'libatspi2.0-0' ] dependency_command = f"apt-get install -y {' '.join(dependencies)}" os.system(dependency_command) # Install playwright and its browsers os.system('playwright install') os.system('python -m playwright install') st.success("Successfully installed Playwright dependencies") except Exception as e: st.error(f"Error installing dependencies: {e}") st.error("Please run these commands manually in your terminal:") st.code("apt-get update") st.code(f"apt-get install -y {' '.join(dependencies)}") st.code("playwright install") # Import other required packages import spacy import spacy.cli @st.cache_resource def load_models(): try: # Try to load spaCy model try: nlp = spacy.load("en_core_web_sm") except OSError: st.info("Downloading spaCy model...") spacy.cli.download("en_core_web_sm") nlp = spacy.load("en_core_web_sm") # Load other models from sentence_transformers import SentenceTransformer semantic_model = SentenceTransformer('all-MiniLM-L6-v2') from transformers import pipeline summarizer = pipeline("summarization") return nlp, semantic_model, summarizer except Exception as e: st.error(f"Error loading models: {e}") return None, None, None # Initialize dependencies and models with st.spinner("Setting up dependencies..."): install_playwright_dependencies() with st.spinner("Loading models..."): nlp_model, semantic_model, summarizer = load_models() if not all([nlp_model, semantic_model, summarizer]): st.error("Failed to load required models. Please check the error messages above.") st.stop() # Rest of your imports and code here... # Rest of your code... import os import subprocess from playwright.async_api import async_playwright, TimeoutError as PlaywrightTimeoutError import asyncio import logging from urllib.parse import urlparse import re from pathlib import Path from io import BytesIO import random from bs4 import BeautifulSoup from PyPDF2 import PdfReader import zipfile import tempfile import mimetypes import requests # -------------------- spaCy Model Setup -------------------- import spacy import spacy.cli from spacy.language import Language @Language.factory("spacy-curated-transformers_RobertaTransformer_v1") def dummy_roberta_transformer(nlp, name): def dummy(doc): return doc return dummy @st.cache_resource def load_nlp_model(): try: nlp_model = spacy.load("en_core_web_sm") except OSError: st.write("Model en_core_web_sm not found. Downloading it now...") spacy.cli.download("en_core_web_sm") nlp_model = spacy.load("en_core_web_sm") return nlp_model nlp_model = load_nlp_model() # Also load SentenceTransformer for semantic re-ranking. from sentence_transformers import SentenceTransformer, util @st.cache_resource def load_semantic_model(): return SentenceTransformer('all-MiniLM-L6-v2') semantic_model = load_semantic_model() # -------------------- Transformers Summarization Setup -------------------- from transformers import pipeline @st.cache_resource def load_summarizer(): return pipeline("summarization") summarizer = load_summarizer() def summarize_pdf_url(pdf_url): try: with st.spinner("Downloading and processing PDF..."): response = requests.get(pdf_url, stream=True) temp_pdf = tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") with open(temp_pdf.name, "wb") as f: f.write(response.content) reader = PdfReader(temp_pdf.name) text = " ".join([page.extract_text() or "" for page in reader.pages]) os.remove(temp_pdf.name) limited_text = text[:3000] summary = summarizer(limited_text, max_length=200, min_length=50, do_sample=False) return summary[0]["summary_text"] except Exception as e: return f"Error summarizing PDF: {e}" # -------------------- Google API Setup -------------------- GOOGLE_OAUTH_CONFIG = { "web": { "client_id": "your_client_id", "project_id": "your_project_id", "auth_uri": "https://accounts.google.com/o/oauth2/auth", "token_uri": "https://oauth2.googleapis.com/token", "auth_provider_x509_cert_url": "https://www.googleapis.com/oauth2/v1/certs", "client_secret": "your_client_secret", "redirect_uris": ["your_redirect_uri"] } } import google_auth_oauthlib.flow import googleapiclient.discovery import google.auth.transport.requests def get_google_auth_url(): client_config = GOOGLE_OAUTH_CONFIG["web"] flow = google_auth_oauthlib.flow.Flow.from_client_config( {"web": client_config}, scopes=["https://www.googleapis.com/auth/drive.file"] ) flow.redirect_uri = client_config["redirect_uris"][0] authorization_url, _ = flow.authorization_url( access_type="offline", include_granted_scopes="true", prompt="consent" ) return authorization_url def exchange_code_for_credentials(auth_code): if not auth_code.strip(): return None, "No code provided." try: client_config = GOOGLE_OAUTH_CONFIG["web"] flow = google_auth_oauthlib.flow.Flow.from_client_config( {"web": client_config}, scopes=["https://www.googleapis.com/auth/drive.file"] ) flow.redirect_uri = client_config["redirect_uris"][0] flow.fetch_token(code=auth_code.strip()) creds = flow.credentials if not creds or not creds.valid: return None, "Could not validate credentials. Check code and try again." return creds, "Google Sign-In successful!" except Exception as e: return None, f"Error during token exchange: {e}" # -------------------- Playwright Setup -------------------- def install_playwright_dependencies(): os.environ['PLAYWRIGHT_BROWSERS_PATH'] = os.path.expanduser("~/.cache/ms-playwright") os.environ['LD_LIBRARY_PATH'] = '/usr/lib/playwright:/usr/lib/x86_64-linux-gnu' try: subprocess.run(['python3', '-m', 'playwright', 'install', 'chromium'], check=True) except Exception as e: st.error(f"Error installing Playwright: {e}") # Initialize Playwright dependencies install_playwright_dependencies() # -------------------- Logging Setup -------------------- logging.basicConfig( filename='advanced_download_log.txt', level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s' ) logger = logging.getLogger() # -------------------- Shared Utils -------------------- USER_AGENTS = [ 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36', 'Mozilla/5.0 (Macintosh; Intel Mac OS X 12_6_3) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.4 Safari/605.1.15', 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36', 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:115.0) Gecko/20100101 Firefox/115.0', ] def get_random_user_agent(): return random.choice(USER_AGENTS) def sizeof_fmt(num, suffix='B'): for unit in ['', 'K', 'M', 'G', 'T', 'P', 'E', 'Z']: if abs(num) < 1024.0: return f"{num:3.1f}{unit}{suffix}" num /= 1024.0 return f"{num:.1f}Y{suffix}" # ---------- Human-like Interactions ------------- async def human_like_scroll(page): scroll_height = await page.evaluate('document.body.scrollHeight') viewport_height = await page.evaluate('window.innerHeight') current_scroll = 0 while current_scroll < scroll_height: await page.evaluate(f'window.scrollTo(0, {current_scroll})') await asyncio.sleep(random.uniform(0.5, 1.5)) current_scroll += viewport_height * random.uniform(0.5, 1.5) scroll_height = await page.evaluate('document.body.scrollHeight') async def human_like_interactions(page): await page.mouse.move(random.randint(0, 1000), random.randint(0, 1000)) await asyncio.sleep(random.uniform(0.5, 1.5)) await page.mouse.click(random.randint(0, 1000), random.randint(0, 1000)) await asyncio.sleep(random.uniform(0.5, 1.5)) await page.evaluate("window.scrollBy(0, window.innerHeight / 2)") await asyncio.sleep(random.uniform(0.5, 1.5)) # ---------- NLP Helpers ------------- def nlp_preprocess(query: str) -> str: doc = nlp_model(query) tokens = [token.lemma_.lower() for token in doc if not token.is_stop and token.is_alpha] processed = " ".join(tokens) return processed if processed.strip() else query def nlp_extract_entities(text: str): doc = nlp_model(text) return [(ent.text, ent.label_) for ent in doc.ents] # ---------- AI-enhanced Query Preprocessing ------------- def ai_preprocess_query(query: str) -> str: return query class DownloadManager: def __init__(self, use_proxy=False, proxy=None, query=None, num_results=5): self.use_proxy = use_proxy self.proxy = proxy self.query = query self.num_results = num_results self.playwright = None self.browser = None self.context = None self.page = None async def __aenter__(self): self.playwright = await async_playwright().start() opts = {"headless": True} if self.use_proxy and self.proxy: opts["proxy"] = {"server": self.proxy} self.browser = await self.playwright.chromium.launch(**opts) self.context = await self.browser.new_context(user_agent=get_random_user_agent()) self.page = await self.context.new_page() await self.page.set_extra_http_headers({ 'Accept-Language': 'en-US,en;q=0.9', 'Accept-Encoding': 'gzip, deflate, br', 'Referer': 'https://www.bing.com/' }) return self async def __aexit__(self, exc_type, exc_val, exc_tb): if self.browser: await self.browser.close() if self.playwright: await self.playwright.stop() async def get_file_size(self, url): try: async with self.context.new_page() as page: response = await page.request.head(url, timeout=15000) length = response.headers.get('Content-Length', None) if length: return sizeof_fmt(int(length)) else: return "Unknown Size" except Exception: return "Unknown Size" async def get_pdf_metadata(self, url): try: async with self.context.new_page() as page: resp = await page.request.get(url, timeout=15000) if resp.ok: content = await resp.body() pdf = BytesIO(content) reader = PdfReader(pdf) return { 'Title': reader.metadata.get('/Title', 'N/A') if reader.metadata else 'N/A', 'Author': reader.metadata.get('/Author', 'N/A') if reader.metadata else 'N/A', 'Pages': len(reader.pages), } else: return {} except Exception: return {} async def extract_real_download_url(self, url): try: async with self.context.new_page() as page: response = await page.goto(url, wait_until='networkidle', timeout=30000) # Check if the response is a redirect if response and response.headers.get('location'): return response.headers['location'] # Check if response is a file content_type = response.headers.get('content-type', '') if 'text/html' not in content_type.lower(): return url # Look for meta refresh content = await page.content() soup = BeautifulSoup(content, 'html.parser') meta_refresh = soup.find('meta', {'http-equiv': 'refresh'}) if meta_refresh: content = meta_refresh.get('content', '') if 'url=' in content.lower(): return content.split('url=')[-1].strip() return page.url except Exception as e: logger.error(f"Error extracting real download URL: {e}") return url async def extract_downloadable_files(self, url, custom_ext_list): found_files = [] try: # First try to load the page response = await self.page.goto(url, timeout=30000, wait_until='networkidle') if not response: return [] final_url = self.page.url # Handle redirects and download scripts if '.php' in final_url or 'download' in final_url or 'get' in final_url: real_url = await self.extract_real_download_url(final_url) if real_url != final_url: content_type = (await self.page.request.head(real_url)).headers.get('content-type', '') if content_type and 'text/html' not in content_type.lower(): found_files.append({ 'url': real_url, 'filename': os.path.basename(urlparse(real_url).path) or 'downloaded_file', 'size': await self.get_file_size(real_url), 'metadata': {} }) return found_files await self.page.wait_for_load_state('networkidle', timeout=30000) await human_like_interactions(self.page) content = await self.page.content() soup = BeautifulSoup(content, 'html.parser') # Define extensions to look for default_exts = ['.pdf', '.docx', '.doc', '.zip', '.rar', '.mp3', '.mp4', '.avi', '.mkv', '.png', '.jpg', '.jpeg', '.gif', '.xlsx', '.xls', '.ppt', '.pptx', '.txt'] all_exts = set(default_exts + [ext.strip().lower() for ext in custom_ext_list if ext.strip()]) # Parse base URL for relative links parsed_base = urlparse(final_url) base_url = f"{parsed_base.scheme}://{parsed_base.netloc}" # Find all links for a in soup.find_all('a', href=True): href = a['href'].strip() # Skip empty or javascript links if not href or href.startswith('javascript:') or href == '#': continue # Handle special cases (PHP scripts, download handlers) if '.php' in href.lower() or 'download' in href.lower() or 'get' in href.lower(): full_url = href if href.startswith('http') else urljoin(base_url, href) real_url = await self.extract_real_download_url(full_url) if real_url and real_url != full_url: size_str = await self.get_file_size(real_url) found_files.append({ 'url': real_url, 'filename': os.path.basename(urlparse(real_url).path) or 'downloaded_file', 'size': size_str, 'metadata': {} }) continue # Handle direct file links if any(href.lower().endswith(ext) for ext in all_exts): file_url = href if href.startswith('http') else urljoin(base_url, href) size_str = await self.get_file_size(file_url) meta = {} if file_url.lower().endswith('.pdf'): meta = await self.get_pdf_metadata(file_url) found_files.append({ 'url': file_url, 'filename': os.path.basename(urlparse(file_url).path), 'size': size_str, 'metadata': meta }) # Handle Google Drive links elif any(x in href for x in ['drive.google.com', 'docs.google.com']): file_id = None for pattern in [r'/file/d/([^/]+)', r'id=([^&]+)', r'open\?id=([^&]+)']: match = re.search(pattern, href) if match: file_id = match.group(1) break if file_id: direct_url = f"https://drive.google.com/uc?export=download&id={file_id}" async with self.context.new_page() as page: try: response = await page.request.head(direct_url, timeout=15000) filename = file_id content_disposition = response.headers.get('content-disposition', '') if content_disposition: filename_match = re.findall('filename="(.+?)"', content_disposition) if filename_match: filename = filename_match[0] found_files.append({ 'url': direct_url, 'filename': filename, 'size': await self.get_file_size(direct_url), 'metadata': {} }) except Exception as e: logger.error(f"Error processing Google Drive link: {e}") # Make list unique based on URLs seen_urls = set() unique_files = [] for f in found_files: if f['url'] not in seen_urls: seen_urls.add(f['url']) unique_files.append(f) return unique_files except Exception as e: logger.error(f"Error extracting files from {url}: {e}") return [] async def download_file(self, file_info, save_dir, referer): file_url = file_info['url'] fname = file_info['filename'] path = os.path.join(save_dir, fname) # Handle duplicate filenames base, ext = os.path.splitext(fname) counter = 1 while os.path.exists(path): path = os.path.join(save_dir, f"{base}_{counter}{ext}") counter += 1 os.makedirs(save_dir, exist_ok=True) try: # Special handling for Google Drive if 'drive.google.com' in file_url: import gdown try: st.write(f"Downloading from Google Drive: {fname}") output = gdown.download(file_url, path, quiet=False) if output: return path return None except Exception as e: logger.error(f"Google Drive download error: {e}") return None # Handle normal downloads async with self.context.new_page() as page: st.write(f"Downloading: {fname}") headers = { 'Accept': '*/*', 'Accept-Encoding': 'gzip, deflate, br', 'Referer': referer } response = await page.request.get(file_url, headers=headers, timeout=30000) if response.status == 200: content = await response.body() with open(path, 'wb') as f: f.write(content) return path else: logger.error(f"Download failed with status {response.status}: {file_url}") return None except Exception as e: logger.error(f"Error downloading {file_url}: {e}") return None async def search_bing(self): if not self.query: return [], [] search_query = self.query if "filetype:pdf" not in search_query.lower(): search_query += " filetype:pdf" search_url = f"https://www.bing.com/search?q={search_query}&count={self.num_results}" try: await self.page.goto(search_url, timeout=30000) await self.page.wait_for_selector('li.b_algo', timeout=30000) await human_like_scroll(self.page) results = [] elements = await self.page.query_selector_all('li.b_algo') for element in elements: link = await element.query_selector('h2 a') if link: url = await link.get_attribute('href') if url: results.append(url) return results[:self.num_results] except Exception as e: logger.error(f"Bing search error: {e}") return [] async def get_sublinks(self, url, limit=100): try: await self.page.goto(url, timeout=30000) content = await self.page.content() soup = BeautifulSoup(content, 'html.parser') parsed_base = urlparse(url) base_url = f"{parsed_base.scheme}://{parsed_base.netloc}" links = set() for a in soup.find_all('a', href=True): href = a['href'].strip() if href.startswith('http'): links.add(href) elif href.startswith('/'): links.add(f"{base_url}{href}") return list(links)[:limit] except Exception as e: logger.error(f"Error getting sublinks: {e}") return [] async def deep_search(self, url, custom_ext_list=None, sublink_limit=100): if not custom_ext_list: custom_ext_list = [] progress_text = st.empty() progress_bar = st.progress(0) try: # Search main page progress_text.text("Analyzing main page...") main_files = await self.extract_downloadable_files(url, custom_ext_list) # Get and search sublinks progress_text.text("Getting sublinks...") sublinks = await self.get_sublinks(url, sublink_limit) if not sublinks: progress_bar.progress(1.0) return main_files # Process sublinks all_files = main_files total_links = len(sublinks) for i, sublink in enumerate(sublinks, 1): progress_text.text(f"Processing sublink {i}/{total_links}: {sublink}") progress_bar.progress(i/total_links) sub_files = await self.extract_downloadable_files(sublink, custom_ext_list) all_files.extend(sub_files) # Make results unique seen_urls = set() unique_files = [] for f in all_files: if f['url'] not in seen_urls: seen_urls.add(f['url']) unique_files.append(f) progress_text.text(f"Found {len(unique_files)} unique files") progress_bar.progress(1.0) return unique_files except Exception as e: logger.error(f"Deep search error: {e}") return [] # ---------- Main Streamlit UI Implementation ------------- def main(): if 'initialized' not in st.session_state: st.session_state.initialized = True st.session_state.discovered_files = [] st.session_state.current_url = None st.session_state.google_creds = None st.title("Advanced File Downloader") # Sidebar for settings with st.sidebar: st.header("Settings") mode = st.radio("Select Mode", ["Manual URL", "Bing Search", "PDF Summarizer"]) with st.expander("Advanced Options"): custom_extensions = st.text_input( "Custom File Extensions", placeholder=".csv, .txt, .epub" ) max_concurrency = st.slider( "Max Concurrency", min_value=1, max_value=1000, value=200 ) use_proxy = st.checkbox("Use Proxy") proxy = st.text_input("Proxy URL", placeholder="http://proxy:port") # Google OAuth Section with st.expander("Google Drive Integration"): if st.button("Start Google Sign-In"): auth_url = get_google_auth_url() st.markdown(f"[Click here to authorize]({auth_url})") auth_code = st.text_input("Enter authorization code") if st.button("Complete Sign-In") and auth_code: creds, msg = exchange_code_for_credentials(auth_code) st.session_state.google_creds = creds st.write(msg) # Main content area if mode == "Manual URL": st.header("Manual URL Mode") url = st.text_input("Enter URL", placeholder="https://example.com") col1, col2 = st.columns(2) with col1: if st.button("Deep Search", use_container_width=True): if url: async def run_deep_search(): async with DownloadManager( use_proxy=use_proxy, proxy=proxy ) as dm: with st.spinner("Searching for files..."): files = await dm.deep_search( url=url, custom_ext_list=custom_extensions.split(',') if custom_extensions else [], max_concurrency=max_concurrency ) st.session_state.discovered_files = files st.session_state.current_url = url return files files = asyncio.run(run_deep_search()) if files: st.success(f"Found {len(files)} files!") else: st.warning("No files found.") with col2: if st.button("Preview Page", use_container_width=True): if url: async def preview(): async with DownloadManager(use_proxy=use_proxy, proxy=proxy) as dm: with st.spinner("Loading preview..."): return await dm.preview_page(url) preview_html = asyncio.run(preview()) st.markdown(preview_html, unsafe_allow_html=True) # File selection and download section if st.session_state.discovered_files: with st.expander("Download Options", expanded=True): file_options = [f"{f['filename']} ({f['size']})" for f in st.session_state.discovered_files] selected_indices = st.multiselect( "Select files to download", range(len(file_options)), format_func=lambda x: file_options[x] ) if selected_indices: download_dir = st.text_input("Download Directory", value="./downloads") delete_after = st.checkbox("Delete after creating ZIP?") upload_drive = st.checkbox("Upload to Google Drive?") if st.button("Download Selected"): selected_files = [st.session_state.discovered_files[i] for i in selected_indices] async def download_files(): async with DownloadManager(use_proxy=use_proxy, proxy=proxy) as dm: paths = [] for file_info in selected_files: with st.spinner(f"Downloading {file_info['filename']}..."): path = await dm.download_file( file_info, download_dir, st.session_state.current_url ) if path: paths.append(path) return paths downloaded_paths = asyncio.run(download_files()) if downloaded_paths: st.success(f"Successfully downloaded {len(downloaded_paths)} files!") # Create ZIP if needed if len(downloaded_paths) > 1 or delete_after or upload_drive: with tempfile.NamedTemporaryFile(delete=False, suffix='.zip') as tmp: with zipfile.ZipFile(tmp.name, 'w') as zf: for p in downloaded_paths: zf.write(p, arcname=os.path.basename(p)) if upload_drive and st.session_state.google_creds: file_id = google_drive_upload(tmp.name, st.session_state.google_creds) if file_id and not isinstance(file_id, str): st.success(f"Uploaded to Google Drive! File ID: {file_id}") else: st.error("Failed to upload to Google Drive") if delete_after: for p in downloaded_paths: try: os.remove(p) except: pass elif mode == "Bing Search": st.header("Bing Search Mode") query = st.text_input("Enter search query") num_results = st.slider("Number of results", 1, 50, 5) if st.button("Search"): if query: async def run_search(): async with DownloadManager( use_proxy=use_proxy, proxy=proxy, query=query, num_results=num_results ) as dm: with st.spinner("Searching..."): return await dm.search_bing() urls, info = asyncio.run(run_search()) if urls: st.success(f"Found {len(urls)} results!") for i, (url, info) in enumerate(zip(urls, info), 1): with st.expander(f"Result {i}: {url}", expanded=i==1): st.write(f"Snippet: {info['snippet']}") if info['entities']: st.write("Entities:", ', '.join(f"{e[0]} ({e[1]})" for e in info['entities'])) if st.button(f"Deep Search This Result {i}"): st.session_state.current_url = url async def search_result(): async with DownloadManager(use_proxy=use_proxy, proxy=proxy) as dm: return await dm.deep_search( url=url, custom_ext_list=custom_extensions.split(',') if custom_extensions else [], max_concurrency=max_concurrency ) files = asyncio.run(search_result()) if files: st.session_state.discovered_files = files st.success(f"Found {len(files)} files!") else: st.warning("No files found.") else: st.warning("No results found.") else: # PDF Summarizer mode st.header("PDF Summarizer") pdf_url = st.text_input("Enter PDF URL") if st.button("Summarize"): if pdf_url: summary = summarize_pdf_url(pdf_url) st.write("Summary:") st.write(summary) if __name__ == "__main__": main()