局部优化与整体效果 新增时间>节省时间 权衡利弊

原代码

from selenium import webdriver
import requests,time
url_l=[]
with open('DISTINCT_url.txt', 'r', encoding='utf-8') as fo:
    for i in fo:
        url = '%s%s' % ('http://', i.replace('
', ''))
        url_l.append(url)
le,c=len(url_l),0
# browser = webdriver.Firefox()
# browser = webdriver.Chrome()

firefox_profile = webdriver.FirefoxProfile()
firefox_profile.set_preference('permissions.default.image', 2)#某些firefox只需要这个
firefox_profile.set_preference('browser.migration.version', 9001)#部分需要加上这个
#禁用css
firefox_profile.set_preference('permissions.default.stylesheet', 2)
#禁用flash
firefox_profile.set_preference('dom.ipc.plugins.enabled.libflashplayer.so', 'false')
# #禁用js
# firefox_profile.set_preference('javascript.enabled', 'false')

browser = webdriver.Firefox(firefox_profile=firefox_profile)

# browser = webdriver.Firefox()
import random


for i in url_l:
    c+=1
    with open('DISTINCT_url.break.log', 'r', encoding='utf-8') as fc:
        for breaker in fc:
            break
    if c<int(breaker):
        continue
    if c%50==0:
        time.sleep(random.randint(0,3))
    with open('DISTINCT_url.break.log', 'w', encoding='utf-8') as flog:
        flog.write(str(c))

    print(c,'/',le,'---',i)
    try:
        # browser = webdriver.Firefox()
      #  browser.minimize_window()
       # browser.set_window_size(10,10)
        browser.get(i)
        page_source = browser.page_source
        # browser.quit()
        if 'us.com/adunion.js' not  in  page_source:
            s='%s%s' % (i.replace('http://',''),'
')
            print(s)
            with open('DISTINCT_url.404.txt', 'a', encoding='utf-8') as fr:
                fr.write(s)
            print('NOT-IN','---',i)
        else:
            print('OK')
        #browser.close()
    except Exception as e:
        time.sleep(1)
        with open('DISTINCT_url.404.ex.txt', 'a', encoding='utf-8') as fex:
            es='%s%s' %(str(e),'
')
            fex.write(es)
            print(es)


优化后
优化原因：上述代码在404情况下执行时间相对

requests.get(i)
                if req.status_code !
获取返回码时间长

但是404的url在总url池中的比例相对非404但需要检测html广告代码的部分少，而该部分必须调动浏览器执行；由此相当与对该部分增加了requests的时间消耗，而其代价已经超过了获取404url的节省的时间
新增时间>节省时间
弊大于利

from selenium import webdriver
import requests, time

url_l = []
with open('DISTINCT_url.txt', 'r', encoding='utf-8') as fo:
    for i in fo:
        url = '%s%s' % ('http://', i.replace('
', ''))
        url_l.append(url)
le, c = len(url_l), 0
# browser = webdriver.Firefox()
# browser = webdriver.Chrome()

firefox_profile = webdriver.FirefoxProfile()
firefox_profile.set_preference('permissions.default.image', 2)  # 某些firefox只需要这个
firefox_profile.set_preference('browser.migration.version', 9001)  # 部分需要加上这个
# 禁用css
firefox_profile.set_preference('permissions.default.stylesheet', 2)
# 禁用flash
firefox_profile.set_preference('dom.ipc.plugins.enabled.libflashplayer.so', 'false')
# #禁用js
firefox_profile.set_preference('javascript.enabled', 'false')

browser = webdriver.Firefox(firefox_profile=firefox_profile)

# browser = webdriver.Firefox()
import random

while True:
    for i in url_l:
        c += 1
        with open('DISTINCT_url.break.log', 'r', encoding='utf-8') as fc:
            for breaker in fc:
                break
        if c < int(breaker):
            continue
        if c % 50 == 0:
            time.sleep(random.randint(0, 3))
        with open('DISTINCT_url.break.log', 'w', encoding='utf-8') as flog:
            flog.write(str(c))

        print(c, '/', le, '---', i)
        try:
            try:
                req = requests.get(i)
                if req.status_code != 404:
                    try:
                        browser.get(i)
                        page_source = browser.page_source
                        if 'us.com/adunion.js' not in page_source:
                            s = '%s%s' % (i.replace('http://', ''), '
')
                            with open('DISTINCT_url.404.txt', 'a', encoding='utf-8') as fr:
                                fr.write(s)
                            print('NOT-IN', '---', i)
                        else:
                            print('OK', '---', i)
                    except Exception as e:
                        time.sleep(1)
                        with open('DISTINCT_url.404.ex.txt', 'a', encoding='utf-8') as fex:
                            es = '%s%s' % (str(e), '
')
                            fex.write(es)
                else:
                    with open('DISTINCT_url.404.txt', 'a', encoding='utf-8') as fr:
                        s = '%s%s' % (i.replace('http://', ''), '
')
                        print('404---', s)
                        fr.write(s)
                req.close()
            except:
                continue
        except:
            continue