#!/usr/bin/env python3 """ Instagram Reels Scraper ======================= This script scrapes view statistics from the latest Reels of a public Instagram profile. It relies exclusively on browser automation using Selenium and does **not** call any official or unofficial APIs. The script scrolls through the account's Reels page, opens each reel in a new tab, extracts the number of views, the publication date, and the unique slug/ID from the URL, then writes the collected data to a JSON or CSV file. A running total of views across the inspected reels is also reported. Usage (from the command line):: python scrape_instagram_reels.py --username sinkmate.de --reels 20 --output output.json --profile-id 687e41cb8cfee375657ad010 --token your_token Options: --username Instagram username to scrape (required) --reels Number of latest reels to process (default: 20) --output Output filename, .json or .csv (required) --profile-id GoLogin profile ID (required) --token GoLogin API token (required) --delay Delay in seconds between actions to mimic human behaviour (optional) Requirements: - Python 3.8+ - Selenium >= 4.0 - GoLogin Python SDK - webdriver-manager Notes: * Uses GoLogin profiles for browser fingerprint management * No Instagram account is required to run this script; it uses only publicly available data. * Web scraping can break if Instagram updates their HTML or CSS. """ from __future__ import annotations import argparse import contextlib import csv import json import logging import re import sys import time from dataclasses import dataclass from datetime import datetime from pathlib import Path from typing import List, Optional from gologin import GoLogin from selenium import webdriver from selenium.webdriver.common.by import By from selenium.webdriver.common.keys import Keys from selenium.webdriver.chrome.service import Service from selenium.webdriver.support.ui import WebDriverWait from selenium.webdriver.support import expected_conditions as EC from selenium.common.exceptions import NoSuchElementException, TimeoutException from selenium.webdriver import ActionChains from webdriver_manager.chrome import ChromeDriverManager from pyvirtualdisplay import Display logging.basicConfig( level=logging.INFO, format="[%(levelname)s] %(message)s", ) @dataclass class ReelInfo: """Container for information about a single Instagram reel.""" url: str views: int date: Optional[str] slug: str def to_dict(self) -> dict: return { "url": self.url, "views": self.views, "date": self.date, "id": self.slug, } def parse_view_count(text: str) -> int: """Convert a view count string (e.g. '12,3�Tsd. Aufrufe', '1.2M views') to an integer. The view counts on Instagram can be displayed in different locales and with abbreviations. This function normalises the number into an integer. Parameters ---------- text: str Raw text containing the view count and possibly a label ("Aufrufe", "views"). Returns ------- int Parsed integer view count. Returns 0 if parsing fails. """ # Remove non-breaking spaces and commas used as thousand separators. cleaned = text.replace('\u202f', ' ').replace('\xa0', ' ').strip() # Extract the numeric part and the unit (if any) match = re.search(r"([\d.,]+)\s*([KMk]|Tsd\.|M|Mn)?", cleaned) if not match: return 0 number_str, unit = match.groups() # Replace locale-specific decimal separators # In German, '.' is thousand separator and ',' is decimal separator. number_str = number_str.replace('.', '').replace(',', '.') try: value = float(number_str) except ValueError: return 0 multiplier = 1 if unit: unit = unit.lower() if unit in ('k', 'tsd'): multiplier = 1_000 elif unit in ('m', 'mn'): multiplier = 1_000_000 return int(value * multiplier) def handle_cookies_and_login_prompt(driver: webdriver.Remote, delay: float = 0.0) -> None: """Attempt to close cookie banner and login pop-up if present. Parameters ---------- driver: webdriver.Remote An instance of the Selenium WebDriver. delay: float Optional sleep delay to slow down actions. """ # Handle cookie banner (various languages). We'll search for common patterns. cookie_selectors = [ (By.XPATH, "//button[contains(translate(text(),'ABCDEFGHIJKLMNOPQRSTUVWXYZ���','abcdefghijklmnopqrstuvwxyz���'), 'allow') and contains(text(), 'Alle')]"), (By.XPATH, "//button[contains(translate(text(),'ABCDEFGHIJKLMNOPQRSTUVWXYZ���','abcdefghijklmnopqrstuvwxyz���'), 'zustimmen')]"), (By.XPATH, "//button[contains(translate(text(),'ABCDEFGHIJKLMNOPQRSTUVWXYZ���','abcdefghijklmnopqrstuvwxyz���'), 'akzeptieren')]"), (By.XPATH, "//button[contains(translate(text(),'ABCDEFGHIJKLMNOPQRSTUVWXYZ���','abcdefghijklmnopqrstuvwxyz���'), 'optionale cookies ablehnen')]"), (By.XPATH, "//button[contains(text(), 'Allow all cookies')]"), ] for by, selector in cookie_selectors: try: button = WebDriverWait(driver, 2).until(EC.element_to_be_clickable((by, selector))) button.click() if delay: time.sleep(delay) break except (TimeoutException, NoSuchElementException): continue # Close login pop-up if it appears or attempt login try: # Try to find login form and login if credentials are available username_field = driver.find_element(By.XPATH, '//*[@id="loginForm"]/div[1]/div[1]/div/label/input') password_field = driver.find_element(By.XPATH, '//*[@id="loginForm"]/div[1]/div[2]/div/label/input') login_btn = driver.find_element(By.XPATH, '//*[@id="loginForm"]/div[1]/div[3]/button') username_field.send_keys("sinkmate.dev") password_field.send_keys("B9!n4g64") driver.execute_script("arguments[0].click();", login_btn) if delay: time.sleep(delay) except (TimeoutException, NoSuchElementException): # If login form not found, try to close any overlay try: driver.find_element(By.TAG_NAME, "body").send_keys(Keys.ESCAPE) if delay: time.sleep(delay) except: pass def scroll_to_load_reels(driver: webdriver.Remote, reels_count: int, delay: float = 0.0) -> List[ReelInfo]: """Scrolls and collects reels with URLs and view counts directly from the grid.""" collected: List[ReelInfo] = [] seen_urls = set() attempts = 0 max_attempts = 30 while len(collected) < reels_count and attempts < max_attempts: reels = driver.find_elements(By.XPATH, "//a[contains(@href, '/reel/')]") for a in reels: url = a.get_attribute("href") if not url or url in seen_urls: continue seen_urls.add(url) try: # Neuen XPATH verwenden, der gezielt das span-Element mit der View-Anzahl sucht # Es befindet sich innerhalb eines div, das wiederum im -Tag liegt. span = a.find_element(By.XPATH, ".//div/div/div/span/span[contains(@class, 'x1vvkbs') and contains(@class, 'xdj266r') and contains(@class, 'x14z9mp')]") views_text = span.text views = parse_view_count(views_text) if views == 0: logging.warning(f"Views-Element f�r Reel {url} konnte nicht korrekt ausgewertet werden. Text: '{views_text}'") driver.save_screenshot(f"error_screenshot_{attempts}.png") time.sleep(1) span = a.find_element(By.XPATH, ".//div/div/div/span/span[contains(@class, 'x1vvkbs') and contains(@class, 'xdj266r') and contains(@class, 'x14z9mp')]") views_text = span.text views = parse_view_count(views_text) logging.warning(f"Nächster Versuch: '{views_text}'") except NoSuchElementException: logging.warning(f"Views-Element nicht f�r Reel {url} gefunden. Vermutlich noch nicht geladen.") views = 0 slug = url.rstrip('/').split('/')[-1] collected.append(ReelInfo(url=url, views=views, date=None, slug=slug)) if len(collected) >= reels_count: break # Scroll nach unten driver.execute_script("window.scrollBy(0, window.innerHeight);") time.sleep(3) attempts += 1 return collected[:reels_count] def save_results(results: List[ReelInfo], total_views: int, output_path: Path) -> None: """Save the list of reel information to JSON or CSV. Parameters ---------- results: List[ReelInfo] List of extracted reel information. total_views: int Sum of all views across the processed reels. output_path: Path Destination file. The extension determines the format. """ if output_path.suffix.lower() == ".json": data = [r.to_dict() for r in results] payload = { "reels": data, "total_views": total_views, } with output_path.open("w", encoding="utf-8") as f: json.dump(payload, f, ensure_ascii=False, indent=2) elif output_path.suffix.lower() == ".csv": with output_path.open("w", newline="", encoding="utf-8") as csvfile: writer = csv.DictWriter(csvfile, fieldnames=["url", "views", "date", "id"]) writer.writeheader() for r in results: writer.writerow(r.to_dict()) # Append total views as a final row with empty url and date writer.writerow({"url": "TOTAL", "views": total_views, "date": "", "id": ""}) else: raise ValueError(f"Unsupported output format: {output_path.suffix}") def run_scraper(username: str, count: int, output: str, delay: float = 0.0, profile_id: str = None, token: str = None) -> None: """Main entry point to run the reel scraper with GoLogin. Parameters ---------- username: str Instagram account username (without '@'). count: int Number of latest reels to process. output: str Path to output file (.json or .csv). delay: float Optional delay between actions to avoid detection. profile_id: str GoLogin profile ID (optional, uses hardcoded value if not provided). token: str GoLogin API token (optional, uses hardcoded value if not provided). """ # Use hardcoded values if not provided PROFILE_ID = profile_id or "685054d57cb66888bd4a4e6f" TOKEN = token or ( "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiI2ODQzMzQ5MjIwMDMwYjQ0OTdlZmVmN2YiLCJ0eXBlIjoiZGV2Iiwiand0aWQiOiI2ODQ0NDA0MDYyNjQ0YWU1NzQ5ODI2MWMifQ.vj-RJv3gm0KpGnb0fAtmipDS2pDnLAFlDik8QKpsc8A" ) # Initialize GoLogin gl = GoLogin({ "token": TOKEN, "profile_id": PROFILE_ID, "extra_params": [ "--no-sandbox", "--disable-dev-shm-usage", "--enable-unsafe-swiftshader", "--disable-gpu", "--disable-audio-output" ] }) display = Display(visible=0, size=(1920, 1080)) display.start() try: # Start GoLogin profile debugger_address = gl.start() chromium_version = gl.get_chromium_version() service = Service(ChromeDriverManager(driver_version=chromium_version).install()) # Configure Chrome options for GoLogin options = webdriver.ChromeOptions() options.add_experimental_option("debuggerAddress", debugger_address) options.add_argument("--lang=en-US,en;q=0.9") options.add_argument("--disable-blink-features=AutomationControlled") options.add_argument("--disable-notifications") # Create the driver driver = webdriver.Chrome(service=service, options=options) driver.set_window_size(1280, 900) # Wait for browser to be ready time.sleep(2) start_url = f"https://www.instagram.com/{username}/reels/" logging.info(f"Opening page: {start_url}") driver.get(start_url) # Accept cookies/login prompts on initial page ActionChains(driver).move_by_offset(5, 5).perform() #handle_cookies_and_login_prompt(driver, delay) # Wait until reel anchors load try: WebDriverWait(driver, 15).until( EC.presence_of_element_located((By.XPATH, "//a[contains(@href, '/reel/')]")) ) except TimeoutException: logging.error("No reels found on the page. Check if the username is correct or if Instagram requires login.") driver.quit() gl.stop() display.stop() return # Scroll and collect reel links logging.info(f"Collecting up to {count} reel links�") results = scroll_to_load_reels(driver, count, delay) total_views = sum(r.views for r in results) # Iterate through each reel link and extract details logging.info(f"Finished scraping {len(results)} reels. Total views: {total_views}") # Save results to file output_path = Path(output) save_results(results, total_views, output_path) logging.info(f"Results saved to {output_path.resolve()}") # Cleanup driver.quit() gl.stop() except Exception as e: logging.error(f"Error during scraping: {e}") try: gl.stop() except: pass raise def main(argv: List[str] | None = None) -> None: parser = argparse.ArgumentParser(description="Scrape the latest Instagram reels of a public profile using GoLogin profiles.") parser.add_argument("--username", required=True, help="Instagram username (without @)") parser.add_argument("--reels", type=int, default=20, help="Number of recent reels to scrape (default: 20)") parser.add_argument("--output", required=True, help="Output file path (.json or .csv)") parser.add_argument("--profile-id", help="GoLogin profile ID (optional, uses hardcoded value if not provided)") parser.add_argument("--token", help="GoLogin API token (optional, uses hardcoded value if not provided)") parser.add_argument("--delay", type=float, default=0.0, help="Optional delay (seconds) between actions to mimic a human user") args = parser.parse_args(argv) if args.reels <= 0: parser.error("--reels must be a positive integer") run_scraper(args.username, args.reels, args.output, args.delay, args.profile_id, args.token) if __name__ == "__main__": main()