加入 Gitee
与超过 1200万 开发者一起发现、参与优秀开源项目,私有仓库也完全免费 :)
免费加入
文件
克隆/下载
fetch_qiniu.py 2.10 KB
一键复制 编辑 原始数据 按行查看 历史
mktime 提交于 2015-06-23 18:20 . pick yellow picture
from qiniu import Auth, BucketManager
import pickle
import cPickle
import requests
import json
import os
import shutil
ACCESS_KEY = "APPKEY"
SECRET_KEY = "SECRETKEY"
BUCKET = "YELLOW"
def main():
items = []
auth = Auth(ACCESS_KEY, SECRET_KEY)
bucket = BucketManager(auth)
data = bucket.list(BUCKET)
if data[0].get("items"):
items += data[0]['items']
while data[0].get("marker") != None:
print "fetch again:", data[0]["marker"]
data = bucket.list(BUCKET, marker = data[0]["marker"])
items += data[0]["items"]
print "++++++++++++++++++++++++++++++++++++++++"
print len(items), type(items[0]), type(items[len(items)-1])
print "++++++++++++++++++++++++++++++++++++++++"
print items[0]
print "++++++++++ BEGING YELLOW ++++++++++"
#for item in items:
# print item["key"]
f = open("data.pk", "wb")
pickle.dump(items, f)
f.close()
def parse():
f = open("data.pk", "r+")
data = cPickle.load(f)
#print type(data), len(data), data[0]
domain = "http://7xjwhg.com1.z0.glb.clouddn.com"
i = 0
for item in data:
i += 1
if i % 500 == 0:
i = 0
print "COUNT:", i
url = domain + "/" + item["key"] + "?nrop"
try:
data = get_url_data(url)
except:
print "error occured:[%s]" % url
continue
if data["code"] != 0:
print "nrop error", data["code"]
else:
info = data["fileList"][0]
if info["label"] in (0,):
got_yellow(info, item["key"])
def got_yellow(info, name):
if name.find("/") != -1:
name = name[name.find("/")+1:]
path = "pictures/full/" + name[0:2] + "/" + name
if os.path.exists(path):
dst_path = "pictures/yellow/" + name
if os.path.exists(dst_path):
return
print dst_path
shutil.copy(path, dst_path)
else:
print "MISS:", path
def get_url_data(url):
ret = requests.get(url)
ret.close()
return json.loads(ret.text)
if __name__ == '__main__':
parse()
#main()
马建仓 AI 助手
尝试更多
代码解读
代码找茬
代码优化