Exemple #1
0
def write_results(results):
    sent_list_f = open(app_env.get_app_root() + "/xunguagua/sent-list.csv",
                       encoding="UTF-8",
                       mode="w")
    urls_f = open(app_env.get_app_root() + "/xunguagua/urls.txt",
                  encoding="UTF-8",
                  mode="w")
    for result in results:
        sent_list_f.write(",".join(result) + "\n")
        urls_f.write(result[2] + "\n")
    sent_list_f.flush()
    sent_list_f.close()
    urls_f.flush()
    urls_f.close()
Exemple #2
0
    def get_instance(cls, logger_name):
        if logger_name not in cls.__loggers:
            logger = logging.getLogger(logger_name)
            logger.setLevel("INFO")

            log_file_path = app_env.get_app_root() + "/baidu_submit/logs"
            if not os.path.exists(log_file_path):
                os.makedirs(log_file_path)
            fh = RotatingFileHandler(log_file_path + "/" + logger_name,
                                     maxBytes=4096 * 1024,
                                     backupCount=10)
            fmt = logging.Formatter('%(message)s')
            fh.setFormatter(fmt)
            logger.addHandler(fh)
            cls.__loggers[logger_name] = logger
        return cls.__loggers[logger_name]
Exemple #3
0
def init_session():
    cookie = open(app_env.get_app_root() + "/xunguagua/cookie.txt",
                  encoding="GBK").read().strip()
    sess = requests.Session()
    sess.headers.update({
        "Accept": "*/*",
        "Accept-Encoding": "gzip, deflate, sdch",
        "Accept-Language": "zh-CN,zh;q=0.8",
        "Connection": "keep-alive",
        "Host": "yun.xunguagua.com",
        "Referer":
        "http://yun.xunguagua.com/member/?app=member&controller=list",
        "User-Agent":
        "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36 SE 2.X MetaSr 1.0",
        "X-Requested-With": "XMLHttpRequest",
        "Cookie": cookie
    })
    return sess
Exemple #4
0
import json
import os
import queue
import sys
import threading
import time

from baidu_submit.job import UrlSubmitJob

sys.path.append(os.path.split(os.path.abspath(os.path.dirname(__file__)))[0])
from baidu_submit.url_logger import UrlLogger
import app_env
import requests

_COOKIE_FILE = app_env.get_app_root() + "/baidu_submit/cookie.txt"
_COOKIE_FILE_INVALID = app_env.get_app_root(
) + "/baidu_submit/cookie-invalid.txt"
_COOKIE_EXPIRE_COUNT = 10
_THREAD_SIZE = 2
_PROXY_CONF = json.load(
    open(app_env.get_app_root() + "/baidu_submit/proxy.json",
         encoding="UTF-8"))
_PROXY_REPO = "http://www.xdaili.cn/ipagent/greatRecharge/getGreatIp?spiderId={}&orderno={}&returnType=2&count=1"


class BaiduSubmit:
    def __init__(self):
        self._refill_cookies()
        self._change_proxy()
        self._change_cookie()
        self._url_buffer = queue.Queue()
Exemple #5
0
import json
import os
import queue
import random
import sys
import threading
import time

from baidu_submit.job import UrlSubmitJob

sys.path.append(os.path.split(os.path.abspath(os.path.dirname(__file__)))[0])
from baidu_submit.url_logger import UrlLogger
import app_env
import requests

_COOKIE_FILE = app_env.get_app_root() + "/baidu_submit/cookie.txt"
_COOKIE_EXPIRE_COUNT = 10
_THREAD_SIZE = 2
_PROXY_CONF = json.load(
    open(app_env.get_app_root() + "/baidu_submit/proxy.json",
         encoding="UTF-8"))
_PROXY_REPO = "http://www.xdaili.cn/ipagent/greatRecharge/getGreatIp?spiderId={}&orderno={}&returnType=2&count=1"


class BaiduSubmit:
    def __init__(self):
        self._refill_cookies()
        self._change_proxy()
        self._change_cookie()
        self._url_buffer = queue.Queue()
        self.start_buffer_consumer()