def __enter__(self):
logInfo("before--lock %s" % self.name)
self.lock.acquire()
logInfo("lock %s" % self.name)
return self
def __exit__(self, exc_type, exc_val, exc_tb):
self.lock.release()
logInfo("unlock %s" % self.name)
try:
pro = psutil.Process(pid)
if not createTime or pro.create_time() == float(createTime):
for proc in pro.children(recursive=True):
killone(proc)
parent = pro.parent()
killone(pro)
if killParent:
killone(parent)
else:
logError("error:createTime=%s,proTime=%s" %
(createTime, pro.create_time()))
except psutil.NoSuchProcess, e:
info = "\nthe process is killed already-%s" % pid
logInfo(info)
killInfo.append(info)
except Exception:
logException()
return killInfo
def showProcess(qinfo='python'):
"""
获取进程的进程号和cmdline
:param qinfo:python!!job.id=!!
:return:
cd /opt/work/spiderman/superbase/;sudo git pull;cd ..;python jobManager/manage/node.py pkill getTYCDetail;exit
"""
import psutil
# 遍历所有运行的进程的进程号,然后通过进程号获取每个进程的cmdline
def runProcess(cmd,
outInfo=None,
maxOutInfoNum=1000,
debug=False,
redirect=False,
exitInfo=None):
"""
运行多进程
:param cmd:
:param outInfo: 输出的console信息list
:param log: 可定制的logger
:param maxOutInfoNum: 最多输出的console 信息行数
:param debug: debug模式只是输出命令行
:param redirectFile: 是否用重定向文件模式
:param ,exitInfo: 遇到该消息退出
:return:
"""
# cmd += "\n" #what the hell use it?
from superbase.utility.logUtil import logInfo
try:
if redirect:
idx = cmd.rfind(">")
if idx > 0: # 判断是否需要重定向,重定向必须是绝对路径
outfile = cmd[idx + 1:].strip()
outfile = os.path.abspath(outfile)
logInfo("redirect-file=%s" % outfile)
dir1 = os.path.dirname(outfile)
from superbase.utility.ioUtil import mkdir
mkdir(dir1)
redirectFile = open(outfile, "w")
cmd = cmd[:idx]
else:
redirectFile = None
logDebug("\n%s the cmd is %s\n" % (timeUtil.getCurTime(), cmd))
if debug:
return
p = subprocess.Popen(cmd,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT,
shell=True)
lineNum = 0
while True:
line = p.stdout.readline()
if not line:
break
if exitInfo and line.find(exitInfo) >= 0:
break
# log.debug(line)
if (outInfo != None):
outInfo.append(line)
lineNum += 1
if maxOutInfoNum > 0 and lineNum > maxOutInfoNum:
del outInfo[:-1]
lineNum = 0
if redirectFile:
redirectFile.flush()
if redirectFile:
redirectFile.write(line)
if redirectFile:
redirectFile.close()
logDebug("process-done:%s" % cmd)
except Exception:
from superbase.utility.logUtil import logException
logException()
return outInfo
def post(self, userId='9650668145'):
"""
获取帖子 9650668145 1761234358
{"count":20,"statuses":[...],"total":474,"page":2,"maxPage":24}
https://xueqiu.com/v4/statuses/user_timeline.json?user_id=9650668145&page=12&type=&_=1508802861537
https://xueqiu.com/v4/statuses/user_timeline.json?user_id=1761234358&page=12&type=&_=1508802861537
:param userId:
:return:
"""
#每个用户作为一个索引
newIndex = gConfig.get(CFG_DOWN_INDEX) + "/%s" % userId
gConfig.set(CFG_DOWN_INDEX, newIndex)
urlTemp = "https://xueqiu.com/v4/statuses/user_timeline.json?user_id={userId}&page={pageIdx}&type=&_={ts}"
syncPoint = SyncPoint() #第一条数据作为同步点，因为第一条数据的帖子最新
#先获得同步点信息:
sp = syncPoint.getSyncPoint()
oldSyncInfo = sp.get("syncInfo", None) if sp else None
syncInfo = syncPoint.getNewSyncInfoByDesc(oldSyncInfo)
saveHandler = HourlySaveLocalHandler(syncPoint=syncPoint) #
def save(result):
posts = result["statuses"]
if len(posts) == 0:
syncInfo[CFG_DOWN_SYNCEND] = syncInfo[CFG_DOWN_SYNCCURR]
syncPoint.saveLastSyncInfo({CFG_DOWN_SYNCINFO: syncInfo})
return 0
saveNum = 0
for post in posts:
result = {
"xid":
post["id"],
"userId":
post["user_id"],
"commentId":
post.get("commentId", 0),
"topicId":
post["retweet_status_id"],
"topicUser":
post["retweeted_status"]["user_id"]
if post["retweet_status_id"] else post["user_id"],
"replyCount":
post["reply_count"],
"favCount":
post["fav_count"],
"likeCount":
post["like_count"],
"retweetCount":
post["retweet_count"],
"viewCount":
post["view_count"],
"inTime":
post["created_at"],
}
test = True #测试，内容不做保存
if not test:
text = dehtml(post["text"])
idx = text.find("//@")
if idx > 0:
text = text[:idx]
wordCount = len(text)
if wordCount > gConfig.get("xueqiu.postSaveMinWord",
200): #超过200字的保存
result["info"] = text,
result["wordCount"] = wordCount
if post["reward_count"] > 0:
reward = {
"count": post["reward_count"],
"users": post["reward_user_count"],
"amount": post["reward_amount"],
}
result["reward"] = json.dumps(reward)
if int(result["inTime"]) < int(syncInfo[CFG_DOWN_SYNCBEGIN]):
syncInfo[CFG_DOWN_SYNCCURR] = result["inTime"]
result[CFG_DOWN_SYNCINFO] = syncInfo
saveHandler.handle(result)
saveNum += 1
return saveNum
downNum = 0
curPage = 0 # if gConfig.get(CFG_DOWN_INCMODE,1) else hasDowned/20+1 #该参数可用做断点使用
while int(syncInfo[CFG_DOWN_SYNCCURR]) > int(
syncInfo[CFG_DOWN_SYNCEND]):
url = urlTemp.format(userId=userId,
pageIdx=curPage,
ts=int(time.time()))
result = self.downOne(url, None)
downNum += save(result)
curPage += 1
logInfo("user=%s,curPage=%s,downNum=%s" %
(userId, curPage, downNum))
self.jobDone()
def testAsyncJob(self):
cmd2 = "python spiderx/testx.py testJob"
TAsyncJob(cmd2, delay=10)
logInfo("testAsyncJob done!")
def testExc(self):
logInfo("testing---")
a = 1 / 0
def testDb(self):
gConfig.set("env", "ONLINE")
name = gTop.get(CFG_DB_MONITOR).getOne("select name from job limit 1")
logInfo("testing-name-%s" % name)
from superbase.utility.aliyun import AliYun
AliYun().upFile(os.path.join(PROJECT_ROOT, "log/spiderx.log"))
def downCoupon(self, url):
"""
有三种页面:
1,旗舰店/专卖店:
淘宝:http://super.fanli.com/brand-5311?spm=super_home.pc.bid-5311
非淘宝:http://super.fanli.com/brand-32665?spm=super_home.pc.bid-32665
2,品牌普通店集合:http://super.fanli.com/brand-1181?spm=super_home.pc.bid-1181
3,http://super.fanli.com/brand-63591?pid=13276411402&spm=super_home.pc.pid-13276411402~bid-63591&lc=super_abtest_14071c
:param url:
:return:
"""
self._enableScroll(length=8000, num=3)
result = {}
content = self.http.get(url)
self.extractor.getResultByContent(
content, {
"groups":
ListElement(
"div.gather-wrap> div.container > div.gather-floor-content",
itemCssElement={
"url": CssElement("a.gather-link", "href")
})
}, result)
if result["groups"]: # mode2,brands
for item in result["groups"]:
logInfo("group mode:begin download %s" % item["url"])
self.downCoupon(item["url"])
else:
self.extractor.getResultByContent(
content, {"mode": CssElement("a.coupon-link>p.detail")},
result)
pageMode = 1 if result["mode"] else 0
def handleCoupon(result):
try:
patterns = self.patterns[pageMode]
if not result["url"]:
return logError("handleCoupon:url is null")
url = safeReg1(patterns["url"], result["url"], "url")
if "taobao.com" in url or "tmall.com" in url:
begin, end = result.get("time").split(",")
result2 = {
"beginTime": getTimestampBySec(float(begin)),
"endTime": getTimestampBySec(float(end)),
}
result2["info"] = result["val"]
result2["url"] = url
if pageMode == 0:
result2["sellerId"] = safeReg1(
patterns["sellerId"], result["url"],
"sellerId")
result2["couponId"] = safeReg1(
patterns["couponId"], result["url"],
"couponId")
result2["productId"] = safeReg1(
patterns["productId"], result["url"],
"productId")
result2["limit"], result2["discount"] = patterns[
"discount"].search(result["val"]).groups()
else:
result2["couponId"] = safeReg1(
patterns["couponId"], result["url"],
"couponId")
result2["productId"] = safeReg1(
patterns["productId"], result["url"],
"productId")
result2["discount"] = safeReg1(
patterns["discount"], result["url"],
"discount")
result2["sellerId"] = result2["limit"] = "",
return result2
else:
logError("the url is not taobao or tmall!")
except Exception:
logException(result.get("url", "no_url"))
class CouponHandler(HourlySaveLocalHandler):
def __init__(self):
HourlySaveLocalHandler.__init__(self)
def preProcess(self, result):
result = HourlySaveLocalHandler.preProcess(self, result)
return handleCoupon(result)
if pageMode == 1: # 非正常模式
listItemConf = {
"time": CssElement(None, "data-time"),
"url": CssElement("a.coupon-link", "href"),
"val": CssElement("a.coupon-link>p.coupon"),
}
else:
listItemConf = {
"time": CssElement(None, "data-time"),
"url": CssElement("a.item-coupon", "href"),
"val": CssElement("a.item-coupon"),
}
self.downOneList2(
url,
content,
listConf={TAG_LIST_ITEMS: "[class*=J_super_item]"},
listItemConf=listItemConf,
resultHandler=CouponHandler())
