def jdholder2(queue, redis_client, JDBase, outlog, cookie=None, start=10, end=40):
if cookie:
JDBase.set_cookie(cookie)
# JDBase.set_succeed_log_path(succeedlog)
# JDBase.set_failed_log_path(failedlog)
JDBase.set_redis_client(redis_client)
JDBase.set_result_save_path(outlog)
JDBase.set_useragent(fetch_util.get_pc_useragent())
task_id = queue.get()[0]
task_id = fetch_util.byte_to_str(task_id)
JDBase.set_request_path(task_id)
JDBase.execute()
fetch_util.print_log('process ' + JDBase.getshowlog() + '\t\n')
# 获取结果是否成功
issucceed = JDBase.get_result()
if issucceed:
# 保存成功flag
JDBase.save_succeed_log(task_id)
else:
# 保存失败flag
JDBase.save_failed_log(task_id)
# 睡眠
JDBase.sleep(start, end)
def init_task2(alltaskreadlines, succeedfile, failedfile):
beforetasks = []
if os.path.exists(succeedfile) and os.path.getsize(succeedfile) > 0:
slls = fetch_util.read(succeedfile)
while True:
try:
sll = slls.__next__()
beforetasks = fetch_util.remove_duplicate(sll, alltaskreadlines)
except StopIteration:
break
else:
beforetasks = alltaskreadlines
nowtasks = []
if os.path.exists(failedfile) and os.path.getsize(failedfile) > 0:
flls = fetch_util.read(failedfile)
while True:
try:
fll = flls.__next__()
nowtasks = fetch_util.remove_duplicate(fll, beforetasks)
except StopIteration:
break
else:
nowtasks = beforetasks
fetch_util.print_log("failed file size = 0")
return nowtasks
def jdholder(tasks, JDBase, succeedlog, failedlog, outlog, cookie=None, start=10, end=40):
jd = JDBase
# if cookie:
jd.set_cookie(cookie)
jd.set_succeed_log_path(succeedlog)
jd.set_failed_log_path(failedlog)
jd.set_result_save_path(outlog)
taskslen = len(tasks)
count = 0
for task in tasks:
count += 1
jd.set_useragent(fetch_util.get_pc_useragent())
jd.set_request_path(task)
jd.execute()
fetch_util.print_log('process [' + str(count) + '/' + str(taskslen) + '] ' + ' ' + jd.getshowlog() + '\t\n')
# 获取结果是否成功
issucceed = jd.get_result()
if issucceed:
# 保存成功flag
jd.save_succeed_log(task)
else:
# 保存失败flag
jd.save_failed_log(task)
# 睡眠
jd.sleep(start, end)
def func(lists, maincategory, outfile):
if lists and len(lists) > 0:
alllen = len(lists)
count = 0
for currenturl in lists:
count += 1
pageinfo = PageInfo(currenturl)
appname = pageinfo.getappname()
category = pageinfo.getcategory()
if not category:
category = 'unknow'
tag = pageinfo.gettag()
if not tag:
tag = 'unknow'
# 打印日志
# captureutil.printlog(currenturl + '\t' + appname)
fetch_util.print_log('[' + str(count) + '/' + str(alllen) + '] ' + currenturl)
outinfo = currenturl + '\t' + maincategory + '>' + appname + '\tc:' + category + '\tt:' + tag
# 写入结果
fetch_util.write(outinfo, outfile)
# 随机休眠几秒
sleep(fetch_util.randnum(10, 40))
def match_type1(self, html, requesturl, sources):
divtext = sources.get_text()
currentresult = requesturl + '\t' + fetch_util.replace_some_string(
divtext, '>', '>')
itmenamehtml = html.find("div", {'id': 'name'})
if itmenamehtml:
h1itemname = itmenamehtml.find('h1')
itemname = h1itemname.contents[0]
if itemname:
currentresult += '>' + itemname
# 保存请求成功的path
price = get_price(requesturl)
self.save_result(currentresult + price)
fetch_util.print_log(currentresult + price)
self.set_print_log(requesturl + ' succeed')
self.set_result(True)
def main():
# 开始时间
starttime = time()
# 日志输出目录
dirpath = jdconfig.jd_out_dir
fetch_util.mkdirs(dirpath)
# 输出结果文件
outlog = dirpath + jdconfig.jd_out
redis_pool = redis_util.get_redis_pool_connection()
redis_client = redis.Redis(connection_pool=redis_pool)
# all_task_size = task_dispatch.get_all_task_size(redis_client, 'jd_20161024')
task_iter = task_dispatch.get_all_task_iter(redis_client, 'jd_20161024')
queue = Queue()
task_dispatch.get_task_queue(queue, task_iter)
task_size = queue.qsize()
if task_size == 0:
fetch_util.print_log("任务总数为0,结束任务")
return
fetch_util.print_log("任务总数: " + str(task_size))
# 设置cookie
cookie = jdutil.jd_pc_cookie('beijing')
jd = JDPage()
while True:
threads = []
for i in range(jdconfig.thread_num):
# th = threading.Thread(target=jdutil.jdholder, args=(task, jd, succeedlog, failedlog, outlog, cookie, 0, 5))
th = threading.Thread(target=jdutil.jdholder2, args=(queue, redis_client, jd, outlog, cookie, 5, 10))
th.start()
threads.append(th)
for th in threads:
th.join()
# 结束时间
endtime = time()
fetch_util.print_log("网页数据获取结束,耗时: " + str(float(endtime - starttime) / 60) + "分钟")
# 写入耗时结果
costtimepath = dirpath + 'jd_' + fetch_util.get_time_yyyymmddhhmmss() + '.txt'
fetch_util.write(str(float(endtime - starttime) / 60), costtimepath, 'w')
pass
def init_task(urlsfile, succeedfile, failedfile):
exist = fetch_util.file_exist(urlsfile)
if not exist:
fetch_util.print_log(urlsfile + " not exist")
return []
with open(urlsfile, 'r') as alltaskopener:
alltaskreadlines = alltaskopener.readlines()
succeedfilereadlines = []
if os.path.exists(succeedfile):
with open(succeedfile, "r") as succeedfileopener:
succeedfilereadlines = succeedfileopener.readlines()
failedfilereadlines = []
if os.path.exists(failedfile):
with open(failedfile, "r") as failedfileopener:
failedfilereadlines = failedfileopener.readlines()
beforetasks = fetch_util.remove_duplicate(succeedfilereadlines, alltaskreadlines)
nowtasks = fetch_util.remove_duplicate(failedfilereadlines, beforetasks)
return nowtasks
def jdholder2(task, JDBase, succeedlog, failedlog, outlog, cookie=None, start=10, end=40):
if cookie:
JDBase.set_cookie(cookie)
JDBase.set_succeed_log_path(succeedlog)
JDBase.set_failed_log_path(failedlog)
JDBase.set_result_save_path(outlog)
JDBase.set_useragent(fetch_util.get_pc_useragent())
JDBase.set_request_path(task)
JDBase.execute()
fetch_util.print_log('process ' + JDBase.getshowlog() + '\t\n')
# 获取结果是否成功
issucceed = JDBase.get_result()
if issucceed:
# 保存成功flag
JDBase.save_succeed_log(task)
else:
# 保存失败flag
JDBase.save_failed_log(task)
# 睡眠
JDBase.sleep(start, end)
def save_result(path, request_url, bread_tag_name, price):
if bread_tag_name:
save_info = request_url + '\t' + bread_tag_name + '\t' + price
fetch_util.print_log(request_url + ' succeed')
fetch_util.print_log_debug(save_info)
if path:
fetch_util.write(save_info, path)
else:
fetch_util.print_log('****** 未设置结果存储路径!!! ******')
else:
fetch_util.print_log(request_url + ' failed')
def main():
# 指定大类别url
outfile = '/Users/Lan/TestDir/out/wandoujia.txt'
# 文件输出位置
specurls = ['http://www.wandoujia.com/category/396']
# outfile = wandoujiaconfig.outfile
# specurls = wandoujiaconfig.specurls
# specurls = ['http://www.wandoujia.com/category/382', 'http://www.wandoujia.com/category/388',
# 'http://www.wandoujia.com/category/402', 'http://www.wandoujia.com/category/392']
allurls = []
fetch_util.print_log('update request urls ...')
for specurl in specurls:
maincategoryurls = MainCategoryUrls(specurl)
url = maincategoryurls.geturls()
allurls.append(url)
urls = fetch_util.liststolist(allurls)
fetch_util.print_log('update request urls finished, len: ' + str(len(urls)))
for url in urls:
parentpage = ParentPage(url=url)
requesturls = parentpage.getpageurls()
if requesturls and len(requesturls) > 0:
for requesturl in requesturls:
parentpageurl = ParentPageUrl(requesturl)
# 当前主大类别
maincategory = parentpageurl.getcategory()
if not maincategory:
maincategory = 'unknow'
# 当前页面可请求urls
currenturls = parentpageurl.getcurpageurls()
if currenturls and len(currenturls) > 0:
tasks = fetch_util.task_dispatch(currenturls, 10)
threads = []
for task in tasks:
th = threading.Thread(target=func, args=(task, maincategory, outfile))
th.start()
threads.append(th)
pass
for th in threads:
th.join()
# 写入结果
fetch_util.write('\r\n------ i am line -----\r\n', outfile)
fetch_util.print_log("has finish: " + url)
sleep(fetch_util.randnum(10, 30))
