20.multi_case07
时间:2022-07-25
本文章向大家介绍20.multi_case07,主要内容包括其使用实例、应用技巧、基本知识点总结和需要注意事项,具有一定的参考价值,需要的朋友可以参考一下。
# coding:utf-8
import re
import ssl
import csv
import json
import time
import random
import asyncio
import aiohttp
import requests
from lxml import etree
from asyncio.queues import Queue
from aiosocksy import Socks5Auth
from aiosocksy.connector import ProxyConnector, ProxyClientRequest
class Common():
task_queue = Queue()
result_queue = Queue()
market_cap_all = 0
currency_rate = 0
# 线上内网
socks5_address_prod = [
'socks5://10.1.100.253:1235',
'socks5://10.1.100.51:1235',
'socks5://10.1.100.70:1235',
'socks5://10.1.100.205:1235',
'socks5://10.1.100.73:1235'
]
# 办公网
socks5_address_dev = [
'socks5://18.208.81.123:1235',
'socks5://34.197.217.25:1235',
'socks5://52.20.255.43:1235',
'socks5://34.237.163.87:1235',
'socks5://18.208.81.123:1235',
'socks5://52.0.114.155:1235'
]
DEPLOY_MODE = "dev"
async def session_get(session, url, socks):
auth = Socks5Auth(login='...', password='...')
headers = {'User-Agent': 'Mozilla/4.0 (compatible; MSIE 5.5; Windows NT)'}
timeout = aiohttp.ClientTimeout(total=20)
response = await session.get(
url,
proxy=socks,
proxy_auth=auth,
timeout=timeout,
headers=headers,
ssl=ssl.SSLContext()
)
return await response.text(), response.status
async def download(url):
connector = ProxyConnector()
if DEPLOY_MODE == "dev":
socks = None
elif DEPLOY_MODE == "Prod":
socks = random.choice(socks5_address_prod)
async with aiohttp.ClientSession(
connector=connector,
request_class=ProxyClientRequest
) as session:
ret, status = await session_get(session, url, socks)
if 'window.kk' in ret and len(ret) < 1000:
url = ret.split("window.kk='")[1].split("'")[0]
ret, status = await session_get(session, url, socks)
return ret, status
async def parse_html(cid, url, response):
coin_info = {}
coin_value = {}
coin_info['url'] = url
coin_info['cid'] = cid
coin_info['time'] = int(time.time())
tree = etree.HTML(response)
try:
price_usd = tree.xpath(
'//div[@class="priceInfo"]/div[@class="sub"]/span[1]/text()'
)[0].strip().replace('$', '')
if '?' not in price_usd:
coin_value['price'] = float(price_usd)
except BaseException:
pass
try:
updown = tree.xpath(
'//div[@class="priceInfo"]/div[@class="sub smallfont"]/span[1]/text()'
)[0].strip().replace('%', '')
coin_value['updown'] = float(updown)
except BaseException:
pass
try:
volume_24h_rmb = tree.xpath(
'//div[@class="info"]/div[@class="charCell"][2]/div[2]/span/text()'
)[0].strip().replace('¥', '').replace(',', '')
coin_value['volume_24h'] = int(
float(volume_24h_rmb) / Common.currency_rate)
except BaseException:
pass
try:
circulating_supply = tree.xpath(
'//div[@class="info"]//div[@class="charCell"][1]/div[@class="val"]/text()'
)[0].strip().replace(',', '')
if '?' not in circulating_supply:
circulating_supply = re.match(
r'^(d+)(w+)$', circulating_supply).group(1)
coin_value['circulating_supply'] = int(circulating_supply)
except BaseException:
pass
try:
if coin_value['price'] and coin_value['circulating_supply']:
market_cap = coin_value['price'] * coin_value['circulating_supply']
coin_value['market_cap'] = market_cap
except BaseException:
pass
try:
if coin_value['market_cap']:
global_share = coin_value['market_cap'] / Common.market_cap_all
if global_share < 0.001:
coin_value['global_share'] = '<0.1%'
else:
coin_value['global_share'] = str(
(global_share * 100).__round__(2)) + '%'
except BaseException:
pass
try:
circulation_rate = tree.xpath(
'//div[@class="info"]//div[@class="charbox"][1]/div[@class="val"]/text()'
)[0].strip()
if '?' not in circulation_rate:
coin_value['circulation_rate'] = circulation_rate
except BaseException:
pass
try:
turnover_rate = tree.xpath(
'//div[@class="info"]//div[@class="charbox"][1]/div[@class="val"]/text()'
)[1].strip()
if '?' not in turnover_rate:
coin_value['turnover_rate'] = turnover_rate
except BaseException:
pass
try:
issue_time = tree.xpath(
'//div[@class="infoList"]/div[1]/div[1]/span[2]/text()'
)[0].strip()
if issue_time != '-':
coin_value['issue_time'] = issue_time
except BaseException:
pass
try:
exchange_num = tree.xpath(
'//div[@class="infoList"]/div[3]/div[1]/span[2]/text()'
)[0].strip().replace('家', '')
coin_value['exchange_num'] = int(exchange_num)
except BaseException:
pass
try:
total_circulation = tree.xpath(
'//div[@class="infoList"]/div[2]/div[2]/span[2]/text()'
)[0].strip().replace(',', '')
coin_value['total_circulation'] = int(total_circulation)
except BaseException:
pass
coin_info['value'] = coin_value
return coin_info
async def down_and_parse_task(queue):
while True:
try:
cid, url = queue.get_nowait()[:2]
except BaseException:
return
for retry_cnt in range(3):
try:
html, status = await download(url)
if status != 200:
html, status = await download(url)
if '访问控制拒绝了你的请求' in html:
html, status = await download(url)
html_parse_result = await parse_html(cid, url, html)
print(html_parse_result)
await Common.result_queue.put(html_parse_result)
break
except BaseException:
await asyncio.sleep(0.2)
continue
async def push(data):
url = 'http://127.0.0.1:8000/aaa'
error = None
for retry_cnt in range(3):
try:
async with aiohttp.ClientSession() as session:
async with session.post(
url,
data=json.dumps(data)
) as response:
pass
response.raise_for_status()
except Exception as e:
await asyncio.sleep(0.2)
print(e)
async def speed_monitor():
while Common.task_queue.qsize() != 0:
old_queue_len = Common.task_queue.qsize()
await asyncio.sleep(5)
new_queue_count = Common.task_queue.qsize()
print('=================')
print('speed = ', (old_queue_len - new_queue_count) / 5)
async def monitor_finish():
while len(asyncio.Task.all_tasks()) > 3:
await asyncio.sleep(1)
await asyncio.sleep(5)
raise SystemExit()
async def push_results():
temp_q = []
while True:
try:
await asyncio.sleep(3)
for _ in range(Common.result_queue.qsize()):
temp_q.append(await Common.result_queue.get())
if len(temp_q) > 0:
await push(temp_q)
temp_q.clear()
except BaseException:
import traceback
print(traceback.format_exc())
async def get_marketcap():
url = 'https://dncapi.feixiaohao.com/api/home/global?webp=0'
response = requests.get(url)
response_json = json.loads(response.text)
marketcap = response_json['data']['marketcapvol']
Common.market_cap_all = int(marketcap)
async def get_currency_rate():
url_rate = 'https://dncapi.feixiaohao.com/api/coin/web-rate/'
response = requests.get(url_rate)
currency_rate = json.loads(response.text)[11]['cny']
Common.currency_rate = currency_rate
# 300秒抓取时间上限
async def time_limit():
await asyncio.sleep(280)
raise SystemExit()
async def main():
# loop = asyncio.get_event_loop()
csv_reader = csv.reader(
open(
'feixiaohao_mapping_data.csv',
encoding='utf-8'))
for row in csv_reader:
try:
if row[1].startswith('https'):
await Common.task_queue.put(row)
except BaseException:
pass
print(Common.task_queue)
await get_marketcap()
print('总市值', Common.market_cap_all)
await get_currency_rate()
print('汇率', Common.currency_rate)
for _ in range(10):
loop.create_task(down_and_parse_task(Common.task_queue))
loop.create_task(monitor_finish())
loop.create_task(speed_monitor())
loop.create_task(push_results())
loop.create_task(time_limit())
if __name__ == '__main__':
loop = asyncio.get_event_loop()
loop.create_task(main())
loop.run_forever()
- linux:手动校准系统时间和硬件CMOS时间
- CSS3魔法堂:说说Multi-column Layout
- 数据可视化-EChart2.0使用总结2
- rpc框架之 thrift 学习 2 - 基本概念
- rpc框架之avro 学习 1 - hello world
- 探讨Android中的内置浏览器和Chrome
- java并发编程学习: 阻塞队列 使用 及 实现原理
- CSS魔法堂:说说Float那个被埋没的志向
- Netbeans配置Xdebug
- rpc框架: thrift/avro/protobuf 之maven插件生成java类
- WebComponent魔法堂:深究Custom Element 之 从过去看现在
- 数据可视化-EChart2.0使用总结1
- JavaScript事件概览
- gradle项目与maven项目相互转化
- JavaScript 教程
- JavaScript 编辑工具
- JavaScript 与HTML
- JavaScript 与Java
- JavaScript 数据结构
- JavaScript 基本数据类型
- JavaScript 特殊数据类型
- JavaScript 运算符
- JavaScript typeof 运算符
- JavaScript 表达式
- JavaScript 类型转换
- JavaScript 基本语法
- JavaScript 注释
- Javascript 基本处理流程
- Javascript 选择结构
- Javascript if 语句
- Javascript if 语句的嵌套
- Javascript switch 语句
- Javascript 循环结构
- Javascript 循环结构实例
- Javascript 跳转语句
- Javascript 控制语句总结
- Javascript 函数介绍
- Javascript 函数的定义
- Javascript 函数调用
- Javascript 几种特殊的函数
- JavaScript 内置函数简介
- Javascript eval() 函数
- Javascript isFinite() 函数
- Javascript isNaN() 函数
- parseInt() 与 parseFloat()
- escape() 与 unescape()
- Javascript 字符串介绍
- Javascript length属性
- javascript 字符串函数
- Javascript 日期对象简介
- Javascript 日期对象用途
- Date 对象属性和方法
- Javascript 数组是什么
- Javascript 创建数组
- Javascript 数组赋值与取值
- Javascript 数组属性和方法
- 第十六章 分布式爬虫--准备工作
- go 搭建并行处理管道
- 新一代基于大数据的管理信息系统(MIS)报表需求开发
- 3. docker-compose实战--ghost app
- 2.1 Kubernetes--Pod
- 3. Kubernetes集群安装
- macOS VirtualBox 桥接模式 设置静态ip 且能和联网
- 重新初始化k8s master节点
- 5.k8s基本命令汇总
- 6. k8s + jenkins 实现持续集成(完)
- 7. 复制k8s Node节点 并重新初始化k8s-nodes2节点 (k8s连载)
- 8.k8s连载--重新生成k8s token(kubeadm join报错及解决)
- 3. dcoker容器的命令
- 4. 镜像的原理
- 5.docker容器数据卷