Scrapy: Follow link to get additional Item data?

问题:

I don't have a specific code issue I'm just not sure how to approach the following problem logistically with the Scrapy framework:

The structure of the data I want to scrape is typically a table row for each item. Straightforward enough, right?

Ultimately I want to scrape the Title, Due Date, and Details for each row. Title and Due Date are immediately available on the page...

BUT the Details themselves aren't in the table -- but rather, a link to the page containing the details (if that doesn't make sense here's a table):

|-------------------------------------------------|
|             Title              |    Due Date    |
|-------------------------------------------------|
| Job Title (Clickable Link)     |    1/1/2012    |
| Other Job (Link)               |    3/2/2012    |
|--------------------------------|----------------|

I'm afraid I still don't know how to logistically pass the item around with callbacks and requests, even after reading through the CrawlSpider section of the Scrapy documentation.

回答1:

Please, first read the docs to understand what i say.

The answer:

To scrape additional fields which are on other pages, in a parse method extract URL of the page with additional info, create and return from that parse method a Request object with that URL and pass already extracted data via its meta parameter.

how do i merge results from target page to current page in scrapy?

回答2:

An example from scrapy documentation

def parse_page1(self, response):
    item = MyItem()
    item['main_url'] = response.url
    request = scrapy.Request("http://www.example.com/some_page.html",
                     callback=self.parse_page2)
    request.meta['item'] = item
    return request

def parse_page2(self, response):
    item = response.meta['item']
    item['other_url'] = response.url
    return item

回答3:

You can also use Python functools.partial to pass an item or any other serializable data via additional arguments to the next Scrapy callback.

Something like:

import functools

# Inside your Spider class:

def parse(self, response):
  # ...
  # Process the first response here, populate item and next_url.
  # ...
  callback = functools.partial(self.parse_next, item, someotherarg)
  return Request(next_url, callback=callback)

def parse_next(self, item, someotherarg, response):
  # ...
  # Process the second response here.
  # ...
  return item