优化Redis去重逻辑,减少开销
This commit is contained in:
parent
8e10b5afdd
commit
3a7b4051c1
250
Pixiv.py
250
Pixiv.py
@ -1,6 +1,6 @@
|
||||
"""
|
||||
P站小爬虫 爬每日排行榜
|
||||
环境需求:Python3.6+ / Redis
|
||||
环境需求:Python3.8+ / Redis
|
||||
项目地址:https://github.com/nyaasuki/PixivSpider
|
||||
|
||||
"""
|
||||
@ -24,67 +24,165 @@ error_list = []
|
||||
|
||||
class PixivSpider(object):
|
||||
|
||||
def __init__(self):
|
||||
def __init__(self, db=0):
|
||||
self.ajax_url = 'https://www.pixiv.net/ajax/illust/{}/pages' # id
|
||||
self.top_url = 'https://www.pixiv.net/ranking.php'
|
||||
self.r = redis.Redis(host='localhost', port=6379, decode_responses=True)
|
||||
self.r = redis.Redis(host='localhost', port=6379, db=db, decode_responses=True)
|
||||
|
||||
def get_list(self, pid):
|
||||
"""
|
||||
:param pid: 插画ID
|
||||
获取作品所有页面的URL
|
||||
:param pid: 作品ID
|
||||
"""
|
||||
response = requests.get(self.ajax_url.format(pid), headers=self.headers, verify=False)
|
||||
json_data = response.json()
|
||||
list_temp = json_data['body']
|
||||
for l in list_temp:
|
||||
url_tamp = l['urls']['original']
|
||||
n = self.r.get(pid)
|
||||
if not n:
|
||||
why_not_do = self.get_img(url_tamp)
|
||||
# 判断是否返回异常 如果有异常则取消这个页面的爬取 等待下次
|
||||
try:
|
||||
# 检查Redis中是否已记录该作品已完全下载
|
||||
if self.r.get(f'downloaded:{pid}') == 'complete':
|
||||
print(f'作品ID:{pid}已在Redis中标记为完全下载,跳过')
|
||||
return None
|
||||
|
||||
# 发送请求获取作品的所有图片信息
|
||||
response = requests.get(self.ajax_url.format(pid), headers=self.headers, verify=False)
|
||||
json_data = response.json()
|
||||
|
||||
# 检查API返回是否有错误
|
||||
if json_data.get('error'):
|
||||
print(f'获取作品ID:{pid}失败:{json_data.get("message")}')
|
||||
return pid
|
||||
|
||||
# 从返回数据中获取图片列表
|
||||
images = json_data.get('body', [])
|
||||
if not images:
|
||||
print(f'作品ID:{pid}没有图片')
|
||||
return pid
|
||||
|
||||
# 获取Redis中已下载的页面记录
|
||||
downloaded_redis = set()
|
||||
for i in range(len(images)):
|
||||
if self.r.get(f'downloaded:{pid}_p{i}') == 'true':
|
||||
downloaded_redis.add(i)
|
||||
|
||||
# 检查本地已下载的文件并更新Redis记录
|
||||
if os.path.exists('./img'):
|
||||
for f in os.listdir('./img'):
|
||||
if f.startswith(f'{pid}_p'):
|
||||
page = int(re.search(r'_p(\d+)\.', f).group(1))
|
||||
if self.r.get(f'downloaded:{pid}_p{page}') != 'true':
|
||||
self.r.set(f'downloaded:{pid}_p{page}', 'true')
|
||||
print(f'发现本地文件并更新Redis记录:{f}')
|
||||
|
||||
# 使用Redis记录作为唯一来源
|
||||
downloaded = downloaded_redis
|
||||
|
||||
# 遍历所有图片进行下载
|
||||
for image in images:
|
||||
# 检查图片数据格式是否正确
|
||||
if 'urls' not in image or 'original' not in image['urls']:
|
||||
print(f'作品ID:{pid}的图片数据格式错误')
|
||||
continue
|
||||
|
||||
# 获取原图URL和页码信息
|
||||
original_url = image['urls']['original']
|
||||
page_num = int(re.search(r'_p(\d+)\.', original_url).group(1))
|
||||
|
||||
# 检查是否已下载过该页面(优先使用Redis记录)
|
||||
if page_num in downloaded:
|
||||
print(f'作品ID:{pid} 第{page_num}页在Redis中已标记为下载,跳过')
|
||||
continue
|
||||
|
||||
# 下载图片,如果下载失败返回作品ID以便后续处理
|
||||
why_not_do = self.get_img(original_url)
|
||||
if why_not_do == 1:
|
||||
return pid
|
||||
else:
|
||||
print(f'插画ID:{pid}已存在!')
|
||||
break
|
||||
|
||||
# with open('pixiv.json', 'a', encoding='utf-8') as f:
|
||||
# f.write(url_tamp + '\n')
|
||||
# 导出
|
||||
|
||||
except requests.exceptions.RequestException as e:
|
||||
print(f'获取作品ID:{pid}时发生网络错误:{str(e)}')
|
||||
return pid
|
||||
except Exception as e:
|
||||
print(f'处理作品ID:{pid}时发生错误:{str(e)}')
|
||||
return pid
|
||||
|
||||
def get_img(self, url):
|
||||
"""
|
||||
|
||||
:param url: 作品页URL
|
||||
:return:
|
||||
下载单个图片
|
||||
:param url: 原图URL,格式如:https://i.pximg.net/img-original/img/2024/12/14/20/00/36/125183562_p0.jpg
|
||||
:return: 0表示下载成功,1表示下载失败
|
||||
"""
|
||||
# 确保下载目录存在
|
||||
if not os.path.isdir('./img'):
|
||||
os.makedirs('./img')
|
||||
file_name = re.findall('/\d+/\d+/\d+/\d+/\d+/\d+/(.*)', url)[0]
|
||||
if os.path.isfile(f'./img/{file_name}'):
|
||||
print(f'文件:{file_name}已存在,跳过')
|
||||
# 单个文件存在并不能判断是否爬取过
|
||||
|
||||
# 从URL提取作品ID、页码和文件扩展名
|
||||
match = re.search(r'/(\d+)_p(\d+)\.([a-z]+)$', url)
|
||||
if not match:
|
||||
print(f'无效的URL格式: {url}')
|
||||
return 1
|
||||
|
||||
# 解析URL信息并构建文件名
|
||||
illust_id, page_num, extension = match.groups()
|
||||
file_name = f"{illust_id}_p{page_num}.{extension}"
|
||||
|
||||
# 检查Redis中是否已记录为下载
|
||||
if self.r.get(f'downloaded:{illust_id}_p{page_num}') == 'true':
|
||||
print(f'Redis记录:{file_name}已下载,跳过')
|
||||
return 0
|
||||
print(f'开始下载:{file_name}')
|
||||
t = 0
|
||||
|
||||
# 作为备份检查,验证文件是否存在
|
||||
if os.path.isfile(f'./img/{file_name}'):
|
||||
# 如果文件存在但Redis没有记录,更新Redis记录
|
||||
self.r.set(f'downloaded:{illust_id}_p{page_num}', 'true')
|
||||
print(f'文件已存在但Redis未记录,已更新Redis:{file_name}')
|
||||
return 0
|
||||
|
||||
# 开始下载流程
|
||||
print(f'开始下载:{file_name} (第{int(page_num)+1}张)')
|
||||
t = 0 # 重试计数器
|
||||
# 最多重试3次
|
||||
while t < 3:
|
||||
try:
|
||||
# 下载图片,设置15秒超时
|
||||
img_temp = requests.get(url, headers=self.headers, timeout=15, verify=False)
|
||||
break
|
||||
except requests.exceptions.RequestException:
|
||||
print('连接异常!正在重试!')
|
||||
if img_temp.status_code == 200:
|
||||
break
|
||||
print(f'下载失败,状态码:{img_temp.status_code}')
|
||||
t += 1
|
||||
except requests.exceptions.RequestException as e:
|
||||
print(f'连接异常:{str(e)}')
|
||||
t += 1
|
||||
|
||||
# 如果重试3次都失败,放弃下载
|
||||
if t == 3:
|
||||
# 返回异常 取消此次爬取 等待下次
|
||||
print(f'下载失败次数过多,跳过该图片')
|
||||
return 1
|
||||
|
||||
# 将图片内容写入文件
|
||||
with open(f'./img/{file_name}', 'wb') as fp:
|
||||
fp.write(img_temp.content)
|
||||
|
||||
# 下载成功后在Redis中记录
|
||||
self.r.set(f'downloaded:{illust_id}_p{page_num}', 'true')
|
||||
# 获取作品总页数并检查是否已下载所有页面
|
||||
page_count = self.r.get(f'total_pages:{illust_id}')
|
||||
if not page_count:
|
||||
# 当前页号+1可能是总页数(保守估计)
|
||||
self.r.set(f'total_pages:{illust_id}', str(int(page_num) + 1))
|
||||
elif int(page_num) + 1 == int(page_count):
|
||||
# 如果当前是最后一页,检查是否所有页面都已下载
|
||||
all_downloaded = all(
|
||||
self.r.get(f'downloaded:{illust_id}_p{i}') == 'true'
|
||||
for i in range(int(page_count))
|
||||
)
|
||||
if all_downloaded:
|
||||
self.r.set(f'downloaded:{illust_id}', 'complete')
|
||||
print(f'作品ID:{illust_id}已完全下载')
|
||||
|
||||
print(f'下载完成并已记录到Redis:{file_name}')
|
||||
return 0
|
||||
|
||||
def get_top_url(self, num):
|
||||
"""
|
||||
|
||||
:param num: 页码
|
||||
:return:
|
||||
获取每日排行榜的特定页码数据
|
||||
:param num: 页码数(1-10)
|
||||
:return: None
|
||||
"""
|
||||
params = {
|
||||
'mode': 'daily',
|
||||
@ -97,22 +195,61 @@ class PixivSpider(object):
|
||||
self.pixiv_spider_go(json_data['contents'])
|
||||
|
||||
def get_top_pic(self):
|
||||
"""
|
||||
从排行榜数据中提取作品ID和用户ID
|
||||
并将用户ID存入Redis数据库中
|
||||
:return: 生成器,返回作品ID
|
||||
"""
|
||||
for url in self.data:
|
||||
illust_id = url['illust_id']
|
||||
illust_user = url['user_id']
|
||||
yield illust_id # 生成PID
|
||||
self.r.set(illust_id, illust_user)
|
||||
illust_id = url['illust_id'] # 获取作品ID
|
||||
illust_user = url['user_id'] # 获取用户ID
|
||||
yield illust_id # 生成作品ID
|
||||
self.r.set(illust_id, illust_user) # 将PID保存到Redis中
|
||||
|
||||
@classmethod
|
||||
def pixiv_spider_go(cls, data):
|
||||
"""
|
||||
存储排行榜数据供后续处理
|
||||
:param data: 排行榜JSON数据中的contents部分
|
||||
"""
|
||||
cls.data = data
|
||||
|
||||
@classmethod
|
||||
def pixiv_main(cls):
|
||||
"""
|
||||
爬虫主函数
|
||||
1. 选择Redis数据库
|
||||
2. 获取或设置Cookie
|
||||
3. 配置请求头
|
||||
4. 遍历排行榜页面(1-10页)
|
||||
5. 下载每个作品的所有图片
|
||||
6. 处理下载失败的作品
|
||||
"""
|
||||
# 选择Redis数据库
|
||||
while True:
|
||||
try:
|
||||
print("\n可用的Redis数据库:")
|
||||
for i in range(6):
|
||||
print(f"{i}.DB{i}")
|
||||
db_choice = input("\n请选择Redis数据库 (0-5): ")
|
||||
db_num = int(db_choice)
|
||||
if 0 <= db_num <= 5:
|
||||
break
|
||||
print("错误:请输入0到5之间的数字")
|
||||
except ValueError:
|
||||
print("错误:请输入有效的数字")
|
||||
|
||||
global pixiv
|
||||
pixiv = PixivSpider(db_num)
|
||||
print(f"\n已选择 DB{db_num}")
|
||||
|
||||
# 从Redis获取Cookie,如果没有则要求用户输入
|
||||
cookie = pixiv.r.get('cookie')
|
||||
if not cookie:
|
||||
cookie = input('请输入一个cookie:')
|
||||
pixiv.r.set('cookie', cookie)
|
||||
|
||||
# 配置请求头,包含必要的HTTP头部信息
|
||||
cls.headers = {
|
||||
'accept': 'application/json',
|
||||
'accept-language': 'zh-CN,zh;q=0.9,zh-TW;q=0.8,en-US;q=0.7,en;q=0.6',
|
||||
@ -123,19 +260,34 @@ class PixivSpider(object):
|
||||
'sec-fetch-site': 'same-origin',
|
||||
'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/77.0.3865.75 Safari/537.36'
|
||||
}
|
||||
|
||||
print('开始抓取...')
|
||||
for i in range(1, 11, 1): # p站每日排行榜最多为500个
|
||||
pixiv.get_top_url(i)
|
||||
for j in pixiv.get_top_pic():
|
||||
k = pixiv.get_list(j) # 接口暂时不想写了 先这样凑合一下吧
|
||||
if k:
|
||||
# 遍历排行榜前10页
|
||||
for i in range(1, 11, 1): # p站每日排行榜最多为500个(50个/页 x 10页)
|
||||
pixiv.get_top_url(i) # 获取当前页的排行榜数据
|
||||
for j in pixiv.get_top_pic(): # 遍历当前页的所有作品
|
||||
k = pixiv.get_list(j) # 下载作品的所有图片
|
||||
if k: # 如果下载失败,将作品ID添加到错误列表
|
||||
error_list.append(k)
|
||||
|
||||
# 清理下载失败的作品记录
|
||||
for k in error_list:
|
||||
pixiv.r.delete(k)
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
pixiv = PixivSpider()
|
||||
pixiv.pixiv_main()
|
||||
# for id_url in pixiv.get_list():
|
||||
# pixiv.get_img(id_url)
|
||||
try:
|
||||
print('正在启动Pixiv爬虫...')
|
||||
print('确保已安装并启动Redis服务')
|
||||
print('确保已准备好有效的Pixiv Cookie')
|
||||
|
||||
# 运行主程序
|
||||
PixivSpider.pixiv_main()
|
||||
|
||||
print('爬虫运行完成')
|
||||
except redis.exceptions.ConnectionError:
|
||||
print('错误:无法连接到Redis服务,请确保Redis服务正在运行')
|
||||
except KeyboardInterrupt:
|
||||
print('\n用户中断运行')
|
||||
except Exception as e:
|
||||
print(f'发生错误:{str(e)}')
|
||||
|
Loading…
Reference in New Issue
Block a user