I'm scraping review from moocs likes this one
From there I'm getting all the course details, 5 items and another 6 items from each review itself.
This is the code I have for the course details:
def parse_reviews(self, response):
l = ItemLoader(item=MoocsItem(), response=response)
l.add_xpath('course_title', '//*[@class="course-header-ng__main-info__name__title"]//text()')
l.add_xpath('course_description', '//*[@class="course-info__description"]//p/text()')
l.add_xpath('course_instructors', '//*[@class="course-info__instructors__names"]//text()')
l.add_xpath('course_key_concepts', '//*[@class="key-concepts__labels"]//text()')
l.add_value('course_link', response.url)
return l.load_item()
Now I want to include the review details, another 5 items for each review. Since the course data is common for all the reviews I want to store it in a different file and use course name/id to relate the data afterward.
This is the code I have for the review's items:
for review in response.xpath('//*[@class="review-body"]'):
review_body = review.xpath('.//div[@class="review-body__content"]//text()').extract()
course_stage = review.xpath('.//*[@class="review-body-info__course-stage--completed"]//text()').extract()
user_name = review.xpath('.//*[@class="review-body__username"]//text()').extract()
review_date = review.xpath('.//*[@itemprop="datePublished"]/@datetime').extract()
score = review.xpath('.//*[@class="sr-only"]//text()').extract()
I tried to work with a temporary solution, returning all the items for each case but is not working either:
def parse_reviews(self, response):
#print response.body
l = ItemLoader(item=MoocsItem(), response=response)
#l = MyItemLoader(selector=response)
l.add_xpath('course_title', '//*[@class="course-header-ng__main-info__name__title"]//text()')
l.add_xpath('course_description', '//*[@class="course-info__description"]//p/text()')
l.add_xpath('course_instructors', '//*[@class="course-info__instructors__names"]//text()')
l.add_xpath('course_key_concepts', '//*[@class="key-concepts__labels"]//text()')
l.add_value('course_link', response.url)
for review in response.xpath('//*[@class="review-body"]'):
l.add_xpath('review_body', './/div[@class="review-body__content"]//text()')
l.add_xpath('course_stage', './/*[@class="review-body-info__course-stage--completed"]//text()')
l.add_xpath('user_name', './/*[@class="review-body__username"]//text()')
l.add_xpath('review_date', './/*[@itemprop="datePublished"]/@datetime')
l.add_xpath('score', './/*[@class="sr-only"]//text()')
yield l.load_item()
The output file for that script is corrupted, cells are displaced and the size of the fields is not correct.
EDIT: I want to have two files at the output:
The first one containing:
course_title,course_description,course_instructors,course_key_concepts,course_link
And the second one with:
course_title,review_body,course_stage,user_name,review_date,score
The issue is you are mixing everything up into a single item, which is not the right way to do it. You should create two items: MoocsItem and MoocsReviewItem.
And then update the code like below
def parse_reviews(self, response):
#print response.body
l = ItemLoader(item=MoocsItem(), response=response)
l.add_xpath('course_title', '//*[@class="course-header-ng__main-info__name__title"]//text()')
l.add_xpath('course_description', '//*[@class="course-info__description"]//p/text()')
l.add_xpath('course_instructors', '//*[@class="course-info__instructors__names"]//text()')
l.add_xpath('course_key_concepts', '//*[@class="key-concepts__labels"]//text()')
l.add_value('course_link', response.url)
item = l.load_item()
for review in response.xpath('//*[@class="review-body"]'):
r = ItemLoader(item=MoocsReviewItem(), response=response, selector=review)
r.add_value('course_title', item['course_title'])
r.add_xpath('review_body', './/div[@class="review-body__content"]//text()')
r.add_xpath('course_stage', './/*[@class="review-body-info__course-stage--completed"]//text()')
r.add_xpath('user_name', './/*[@class="review-body__username"]//text()')
r.add_xpath('review_date', './/*[@itemprop="datePublished"]/@datetime')
r.add_xpath('score', './/*[@class="sr-only"]//text()')
yield r.load_item()
yield item
Now what you want is that different item type goes in different csv files, which is what the below SO thread answers:
I have not tested the below, but the code will look something like this:
from scrapy.exporters import CsvItemExporter
from scrapy import signals
from scrapy.xlib.pydispatch import dispatcher
def item_type(item):
return type(item).__name__.replace('Item','').lower() # TeamItem => team
class MultiCSVItemPipeline(object):
SaveTypes = ['moocs','moocsreview']
def __init__(self):
dispatcher.connect(self.spider_opened, signal=signals.spider_opened)
dispatcher.connect(self.spider_closed, signal=signals.spider_closed)
def spider_opened(self, spider):
self.files = dict([ (name, open(CSVDir+name+'.csv','w+b')) for name in self.SaveTypes ])
self.exporters = dict([ (name,CsvItemExporter(self.files[name])) for name in self.SaveTypes])
[e.start_exporting() for e in self.exporters.values()]
def spider_closed(self, spider):
[e.finish_exporting() for e in self.exporters.values()]
[f.close() for f in self.files.values()]
def process_item(self, item, spider):
what = item_type(item)
if what in set(self.SaveTypes):
self.exporters[what].export_item(item)
return item
You need to make sure the ITEM_PIPELINES is updated to use this MultiCSVItemPipeline class
ITEM_PIPELINES = {
'mybot.pipelines.MultiCSVItemPipeline': 300,
}