问题
I write a small spider, when I run and it can't call pipeline.
After debug for a while, I find the bug code area.
The logic of the spider is that I crawl the first url to fetch cookie, then I crawl the second url to download the code picture with cookie, and I post some data I prepare to the third url. And If the text I get from the picture wrong then I download again to post the third url repeatedly, until I got the right text.
Let me show you the code:
# -*- coding: gbk -*-
import scrapy
from scrapy.http import FormRequest
import json
import os
from datetime import datetime
from scrapy.selector import Selector
from teacherCourse.handlePic import handle
from teacherCourse.items import DetailProfItem
from teacherCourse.items import DetailProfCourseItem
from teacherCourse.items import containItem
class GetTeacherCourseSpider(scrapy.Spider):
name = 'TeacherCourse'
# custom_settings = {
# 'ITEM_PIPELINES': {
# 'teacherCourse.pipelines.TeacherCoursePipeline': 300,
# }
# }
def __init__(self, selXNXQ='', titleCode=''):
self.getUrl = 'http://jwxt.dgut.edu.cn/jwweb/ZNPK/TeacherKBFB.aspx' # first
self.vcodeUrl = 'http://jwxt.dgut.edu.cn/jwweb/sys/ValidateCode.aspx' # second
self.postUrl = 'http://jwxt.dgut.edu.cn/jwweb/ZNPK/TeacherKBFB_rpt.aspx' # third
self.findSessionId = None # to save the cookies
self.XNXQ = selXNXQ
self.titleCode = titleCode
def start_requests(self):
request = scrapy.Request(self.getUrl,
callback = self.downloadPic)
yield request
def downloadPic(self, response):
# download the picture
# find the session id
self.findSessionId = response.headers.getlist('Set-Cookie')[0].decode().split(";")[0].split("=")
request = scrapy.Request(self.vcodeUrl,
cookies= {self.findSessionId[0]: self.findSessionId[1]},
callback = self.getAndHandleYzm)
yield request
def getAndHandleYzm(self, response):
yzm = handle(response.body)
yield FormRequest(self.postUrl,
formdata={'Sel_XNXQ': '20151',
'sel_zc': '011',
'txt_yzm': yzm,
'type': '2'},
headers={
'Referer': 'http://jwxt.dgut.edu.cn/jwweb/ZNPK/TeacherKBFB.aspx',
'Cookie': self.findSessionId[0] + '=' + self.findSessionId[1],
},
callback=self.parse)
def parse(self, response):
body = response.body.decode('gbk')
num = body.find('alert')
if num != -1:
# means CAPTCHA validation fails, need to re-request the CAPTCHA
yield scrapy.Request(self.vcodeUrl+'?t='+'%.f' % (datetime.now().microsecond / 1000),
headers={
'Referer': 'http://jwxt.dgut.edu.cn/jwweb/ZNPK/TeacherKBFB.aspx',
'Cookie': self.findSessionId[0]+'='+self.findSessionId[1]
},
callback=self.getAndHandleYzm)
else:
# parse data
self.parseData(body)
# item = containItem()
# item['first'] = len(body)
# return item
# the parse data part is a little bit long, but it doesn't matter.
# At the last line, I did yield a item
def parseData(self, body):
# parse body data
sel = Selector(text=body)
# get all the note text data
noteTables = sel.xpath('//table[@style="border:0px;"]').extract()
noteList = [] # to store all the note text
for noteTable in noteTables:
if '<b>' in noteTable:
sele = Selector(text = noteTable)
note = (sele.xpath('//table/tr/td/b/text()').extract())
noteText = (sele.xpath('//table/tr/td/text()').extract())
# combine note and noteText
if not noteText:
noteText.append('')
noteText.append('')
else:
if len(noteText) == 1:
noteText.append('')
noteList.append(noteText)
# get all the course data
courseTables = sel.xpath('//table[@class="page_table"]/tbody').extract()
AllDetailCourse = [] # all the teachers' course
for table in courseTables:
everyTeacherC = [] # every teacher's course
s = Selector(text = table)
trs = s.xpath('//tr').extract()
for tr in trs:
sel = Selector(text = tr)
snum = (sel.xpath('//td[1]/text()').extract())
course = (sel.xpath('//td[2]/text()').extract())
credit = (sel.xpath('//td[3]/text()').extract())
teachWay = (sel.xpath('//td[4]/text()').extract())
courseType = (sel.xpath('//td[5]/text()').extract())
classNum = (sel.xpath('//td[6]/text()').extract())
className = (sel.xpath('//td[7]/text()').extract())
stuNum = (sel.xpath('//td[8]/text()').extract())
week = (sel.xpath('//td[9]/text()').extract())
section = (sel.xpath('//td[10]/text()').extract())
location = (sel.xpath('//td[11]/text()').extract())
tmpList = []
tmpList.append(snum)
tmpList.append(course)
tmpList.append(credit)
tmpList.append(teachWay)
tmpList.append(courseType)
tmpList.append(classNum)
tmpList.append(className)
tmpList.append(stuNum)
tmpList.append(week)
tmpList.append(section)
tmpList.append(location)
# to know whether every variable is empty
detailCourse = []
for each in tmpList:
if not each:
each = ''
else:
each = each[0]
detailCourse.append(each)
everyTeacherC.append(detailCourse)
AllDetailCourse.append(everyTeacherC)
# get department, teacher, gender and title
sel = Selector(text = body)
temp1 = sel.xpath('//*[@group="group"]/table/tr/td/text()').extract()
# fill two tables, which will store in the database
i = 0
# every professor
for each in temp1:
tables = containItem() # all the data in every for loop to send to the pipeline
each = each.replace(u'\xa0', u' ')
each = each.split(' ')
depart = each[0].split('£º')
teacher = each[1].split('£º')
gender = each[2].split('£º')
title = each[3].split('£º')
# first table
profItem = DetailProfItem()
profItem['XNXQ'] = self.XNXQ
profItem['department'] = depart[1] # department
profItem['teacher'] = teacher[1] # teacher
profItem['gender'] = gender[1]
profItem['title'] = title[1]
profItem['note1'] = noteList[i][0]
profItem['note2'] = noteList[i][1]
tables['first'] = profItem # add the first table
# second table
# every professor's courses
profCourses = []
for j in range(len(AllDetailCourse[i])): # how many course for every professor
profCourseItem = DetailProfCourseItem() # every course for every professor
profCourseItem['snum'] = AllDetailCourse[i][j][0] # i means i-th professor, j means j-th course, third num means what position of the course
profCourseItem['course'] = AllDetailCourse[i][j][1]
profCourseItem['credit'] = AllDetailCourse[i][j][2]
profCourseItem['teachWay'] = AllDetailCourse[i][j][3]
profCourseItem['courseType'] = AllDetailCourse[i][j][4]
profCourseItem['classNum'] = AllDetailCourse[i][j][5]
profCourseItem['className'] = AllDetailCourse[i][j][6]
profCourseItem['stuNum'] = AllDetailCourse[i][j][7]
profCourseItem['week'] = AllDetailCourse[i][j][8]
profCourseItem['section'] = AllDetailCourse[i][j][9]
profCourseItem['location'] = AllDetailCourse[i][j][10]
profCourses.append(profCourseItem) # every professor's courses
tables['second'] = profCourseItem # add the second table
i += 1
yield tables
Any suggestions would be appreciate!
settings.py: (pipeline part)
# Configure item pipelines
# See http://scrapy.readthedocs.org/en/latest/topics/item-pipeline.html
ITEM_PIPELINES = {
'teacherCourse.pipelines.TeacherCoursePipeline': 300,
}
items.py: (I don't think it's matter)
# detail professor course message
class DetailProfCourseItem(scrapy.Item):
snum = scrapy.Field() # serial number
course = scrapy.Field()
credit = scrapy.Field()
teachWay = scrapy.Field()
courseType = scrapy.Field()
classNum = scrapy.Field()
className = scrapy.Field()
stuNum = scrapy.Field()
week = scrapy.Field()
section = scrapy.Field()
location = scrapy.Field()
# the third item which contain first and second item
class containItem(scrapy.Item):
first = scrapy.Field() # for fist table
second = scrapy.Field() # for second table
pipeline code:
class TeacherCoursePipeline(object):
def process_item(self, item, spider):
print('I am called!!!!!')
print(item)
return item
And When I run spider scrapy crawl TeacherCourse
it output:
2016-07-19 17:39:18 [scrapy] INFO: Scrapy 1.1.0rc1 started (bot: teacherCourse)
2016-07-19 17:39:18 [scrapy] INFO: Overridden settings: {'BOT_NAME': 'teacherCourse', 'NEWSPIDER_MODULE': 'teacherCourse.spiders', 'ROBOTSTXT_OBEY': True, 'SPIDER_MODULES': ['teacherCourse.spiders']}
2016-07-19 17:39:18 [scrapy] INFO: Enabled extensions:
['scrapy.extensions.corestats.CoreStats', 'scrapy.extensions.logstats.LogStats']
2016-07-19 17:39:18 [scrapy] INFO: Enabled downloader middlewares:
['scrapy.downloadermiddlewares.robotstxt.RobotsTxtMiddleware',
'scrapy.downloadermiddlewares.httpauth.HttpAuthMiddleware',
'scrapy.downloadermiddlewares.downloadtimeout.DownloadTimeoutMiddleware',
'scrapy.downloadermiddlewares.useragent.UserAgentMiddleware',
'scrapy.downloadermiddlewares.retry.RetryMiddleware',
'scrapy.downloadermiddlewares.defaultheaders.DefaultHeadersMiddleware',
'scrapy.downloadermiddlewares.redirect.MetaRefreshMiddleware',
'scrapy.downloadermiddlewares.httpcompression.HttpCompressionMiddleware',
'scrapy.downloadermiddlewares.redirect.RedirectMiddleware',
'scrapy.downloadermiddlewares.cookies.CookiesMiddleware',
'scrapy.downloadermiddlewares.chunked.ChunkedTransferMiddleware',
'scrapy.downloadermiddlewares.stats.DownloaderStats']
2016-07-19 17:39:18 [scrapy] INFO: Enabled spider middlewares:
['scrapy.spidermiddlewares.httperror.HttpErrorMiddleware',
'scrapy.spidermiddlewares.offsite.OffsiteMiddleware',
'scrapy.spidermiddlewares.referer.RefererMiddleware',
'scrapy.spidermiddlewares.urllength.UrlLengthMiddleware',
'scrapy.spidermiddlewares.depth.DepthMiddleware']
2016-07-19 17:39:18 [scrapy] INFO: Enabled item pipelines:
['teacherCourse.pipelines.TeacherCoursePipeline']
2016-07-19 17:39:18 [scrapy] INFO: Spider opened
2016-07-19 17:39:18 [scrapy] INFO: Crawled 0 pages (at 0 pages/min), scraped 0 items (at 0 items/min)
2016-07-19 17:39:18 [scrapy] DEBUG: Crawled (404) <GET http://jwxt.dgut.edu.cn/robots.txt> (referer: None)
2016-07-19 17:39:18 [scrapy] DEBUG: Crawled (200) <GET http://jwxt.dgut.edu.cn/jwweb/ZNPK/TeacherKBFB.aspx> (referer: None)
2016-07-19 17:39:19 [scrapy] DEBUG: Crawled (200) <GET http://jwxt.dgut.edu.cn/jwweb/sys/ValidateCode.aspx> (referer: http://jwxt.dgut.edu.cn/jwweb/ZNPK/TeacherKBFB.aspx)
2016-07-19 17:39:19 [scrapy] DEBUG: Crawled (200) <POST http://jwxt.dgut.edu.cn/jwweb/ZNPK/TeacherKBFB_rpt.aspx> (referer: http://jwxt.dgut.edu.cn/jwweb/ZNPK/TeacherKBFB.aspx)
2016-07-19 17:39:19 [scrapy] INFO: Closing spider (finished)
2016-07-19 17:39:19 [scrapy] INFO: Dumping Scrapy stats:
{'downloader/request_bytes': 1330,
'downloader/request_count': 4,
'downloader/request_method_count/GET': 3,
'downloader/request_method_count/POST': 1,
'downloader/response_bytes': 230886,
'downloader/response_count': 4,
'downloader/response_status_count/200': 3,
'downloader/response_status_count/404': 1,
'finish_reason': 'finished',
'finish_time': datetime.datetime(2016, 7, 19, 9, 39, 19, 861620),
'log_count/DEBUG': 4,
'log_count/INFO': 7,
'request_depth_max': 2,
'response_received_count': 4,
'scheduler/dequeued': 3,
'scheduler/dequeued/memory': 3,
'scheduler/enqueued': 3,
'scheduler/enqueued/memory': 3,
'start_time': datetime.datetime(2016, 7, 19, 9, 39, 18, 774293)}
2016-07-19 17:39:19 [scrapy] INFO: Spider closed (finished)
回答1:
The problem seems to be that the parse method only yields scrapy.Request objects, never scrapy.Item instances.
The else: branch calls the generator parseData(body) but doesn't use the data it can produce (namely containItem objects).
One way to solve this is to loop on the generator results and yield them one-by-one:
def parse(self, response):
body = response.body.decode('gbk')
num = body.find('alert')
if num != -1:
# means CAPTCHA validation fails, need to re-request the CAPTCHA
yield scrapy.Request(self.vcodeUrl+'?t='+'%.f' % (datetime.now().microsecond / 1000),
headers={
'Referer': 'http://jwxt.dgut.edu.cn/jwweb/ZNPK/TeacherKBFB.aspx',
'Cookie': self.findSessionId[0]+'='+self.findSessionId[1]
},
callback=self.getAndHandleYzm)
else:
# parse data
for i in self.parseData(body):
yield i
# item = containItem()
# item['first'] = len(body)
# return item
来源：https://stackoverflow.com/questions/38432786/scrapy-a-weird-bug-code-that-cant-call-pipeline