共计 3798 个字符，预计需要花费 10 分钟才能阅读完成。
提醒：本文最后更新于2026-09-10 11:07，文中所关联的信息可能已发生改变，请知悉！
说明：P站很多人不陌生了，这里分享个爬虫脚本，基于Python3可直接下载排行榜、画师的原图。不需要登录，支持多线程下载，支持过滤已下载。绕过官方直接获取下载原图
截图
要求: python 3、网络能访问Pixiv站。
2、代码
from urllib import request
import threading
import queue
import re
import os
class pixiv:
def __init__(self):
self.folder = 'PixivImage'
self.web_coding = 'utf-8'
self.root = os.path.dirname(os.path.abspath(__file__))
self.DefaultHeader = {
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101 Firefox/52.0",
"Accept": "*/*",
"Accept-Language": "zh-CN,zh;q=0.8,en-US;q=0.5,en;q=0.3",
"Accept-Encoding": "",
"Connection": "keep-alive",
}
self.data_low = []
def _http(self, url, headers, Obj=False):
res = request.urlopen(request.Request(url, headers=headers, method='GET'))
if Obj:
return res
else:
return res.read().decode(self.web_coding, "ignore")
def data_image(self, url_id):
_header = self.DefaultHeader.copy()
_header["Referer"] = "https://www.pixiv.net/member_illust.php?mode=medium&illust_id={}".format(url_id)
_url_data = "https://www.pixiv.net/touch/ajax/illust/details?illust_id={}".format(url_id)
data_url = str(str(str(re.findall('"url_big":"[^"]*"', self._http(_url_data, _header))[0]).replace('\\', '')).split(':', 1)[-1]).strip('"')
return data_url, _header
def get_item(self, UserID=None):
if not UserID:
UserID = 'https://www.pixiv.net/ranking.php?mode=male'
if '://' in str(UserID):
Mode_ID = 0
else:
Mode_ID = 1
if Mode_ID:
_url = "https://www.pixiv.net/ajax/user/{}/profile/all".format(str(UserID))
page = self._http(_url, self.DefaultHeader, True)
if page.code != 200:
raise Exception("Pixiv Page:", page.code)
_data = re.findall('"[0-9]+":null', page.read().decode(self.web_coding, "ignore"))
self.data_low = [str(str(item).split(":")[0]).strip('"') for item in _data if ':null' in str(item)]
else:
page = self._http(UserID, self.DefaultHeader, True)
if page.code != 200:
raise Exception("Pixiv Page:", page.code)
_data = re.findall('data-src="[^"]*"', page.read().decode(self.web_coding, "ignore"))
self.data_low = [str(str(str(str(str(item).split("=", 1)[-1]).strip('"')).rsplit('/', 1)[-1]).split('_')[0]) for item in _data if '/img-master/img/' in str(item)]
self.fliter_item()
def fliter_item(self):
folder = os.path.join(self.root, self.folder)
if not os.path.exists(folder):
return None
_split = "_"
_exist = [str(str(item).split(_split)[0]) for item in os.listdir(folder) if _split in item]
print("Exist Item:", len(_exist))
for _item in self.data_low.copy():
if _item in _exist:
self.data_low.remove(_item)
def get_data(self, data_list=None):
if not data_list:
data_list = self.data_low
for item in data_list:
self.get_data_by_item(item)
def get_data_by_item(self, item):
data = self.data_image(item)
image = self._http(data[0], data[1], True)
if image.code != 200:
raise Exception("Pixiv Image: [{} | {}]".format(image.code, data[0]))
self.write(data[0], image.read())
def write(self, url, data):
folder = os.path.join(self.root, self.folder)
if not os.path.exists(folder):
os.mkdir(folder)
file = os.path.join(folder, str(str(url).rsplit('/', 1)[-1]).strip())
fp = open(file, 'wb')
fp.write(data)
fp.close()
print("Pixiv Image: [ OK | {} ]".format(file))
def add_queue(self, _queue, data_list=None):
for item in data_list:
_item = str(item).strip()
if item and _item:
_queue.put(_item)
def multi_data(self, data_list=None, max=25):
if not data_list:
data_list = self.data_low
print("New Item:", len(data_list))
_threads = []
_queue = queue.Queue(maxsize=max)
task_main = threading.Thread(target=self.add_queue, args=(_queue, data_list))
task_main.setName("TaskMain")
task_main.setDaemon(True)
task_main.start()
while _queue.qsize() > 0:
if len(_threads) >= max:
for _item in _threads.copy():
if not _item.is_alive():
_threads.remove(_item)
continue
item = _queue.get()
task = threading.Thread(target=self.get_data_by_item, args=(item, ))
task.setDaemon(True)
task.start()
_threads.append(task)
for _task in _threads:
_task.join()
if __name__ == '__main__':
try:
task = os.sys.argv[1]
except:
task = None
p = pixiv()
p.get_item(task)
p.multi_data()脚本默认是25线程，想改的话，直接编辑脚本，搜索25，改成相应数字就可以了。
3、使用脚本
下载画师原图，使用画师UID
python3 pixiv.py “画师UID”
下载排行榜原图，使用排行榜地址即可
python3 pixiv.py “https://www.pixiv.net/ranking.php?mode=male”
正文完