import os import requests from concurrent.futures import ThreadPoolExecutor from tqdm import tqdm local = 0 endpoint = "http://127.0.0.1:7860" if local else "https://q6-p.hf.space" img_base = 'https://i.pximg.net/img-original/img/' os.chdir(os.path.dirname(os.path.abspath(__file__))) valid = [f for f in os.listdir() if f.endswith(".txt")] for idx, file in enumerate(valid): print(f"{idx + 1}: {file}") inputs = input("Enter the index of the file: ").split() indexes = [] for inp in inputs: if "-" in inp: start, end = map(int, inp.split("-")) indexes.extend(range(start - 1, end)) elif inp.isdigit(): indexes.append(int(inp) - 1) def download_image(args): idx, url, group_name = args r = requests.get(url, headers={"Referer": "https://www.pixiv.net/", "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:126.0) Gecko/20100101 Firefox/126.0"}) with open(f"images/{group_name}/{idx}.png", "wb") as f: f.write(r.content) for i in indexes: group_name = valid[i].rsplit(".", 1)[0] os.makedirs(f"images/{group_name}", exist_ok=True) with open(valid[i], "r") as f: post_ids = [line.strip() for line in f if line.strip()] if not post_ids: continue params = [("post_ids", pid) for pid in post_ids] + [("only_first", "1")] data = requests.get(f"{endpoint}/allimages", params=params).json() image_urls = data.get("image_urls", []) if not image_urls: continue to_download = [(idx, url, group_name) for idx, url in enumerate(image_urls)] with ThreadPoolExecutor(max_workers=20) as executor: list(tqdm(executor.map(download_image, to_download), total=len(to_download), desc=f"Downloading {group_name}")) print(f"Saved {len(image_urls)} images to images/{group_name}")