LFGSpider WIP

lfg.py

# -*- coding: utf-8 -*-
import scrapy

class Comic(scrapy.Item):
    name              = scrapy.Field()
    volumes           = scrapy.Field()
    volumes_selectors = scrapy.Field()

class Volume(scrapy.Item):
    name           = scrapy.Field()
    chapters       = scrapy.Field()

class LFGSpider2(scrapy.Spider):
    name = 'lfg2'
    allowed_domains = ['lfg.co']
    start_urls = ['http://www.lfg.co/archives/']

    def parse(self, response):
        #comic = Comic()
        #self.parse_comic(response, comic)
        # How to get the volumes?
        #links = comic['volumes_selectors']
        #comic['volumes'] = dict()
        comic = 'Looking for a group'
        links = response.xpath("//a[@class='archive-thumb']/@href").extract()[:3]
        if len(links) > 0:
            url = response.urljoin(links[0])
            yield scrapy.Request(url, callback=self.parse_volume,
                meta={'comic': comic, 'links': links})
        yield {'Comic': comic}

    def parse_comic(self, response, comic):
        comic['name']              = 'Looking for a group'
        comic['volumes_selectors'] = response.xpath("//a[@class='archive-thumb']/@href").extract()[:3]

    def parse_volume(self, response):
        # Remove the first link
        response.meta['links'].pop(0)
        #volume = Volume()
        #volume['name']     = 'TO DO'
        #volume['chapters'] = 'TO DO'
        #response.meta['comic']['volumes']['TO DO'] = volume
        yield {'Volume': len(response.meta['links'])}
        if len(response.meta['links']) > 0:
            url = response.urljoin(response.meta['links'][0])
            yield scrapy.Request(url, callback=self.parse_volume,
                meta={'comic': response.meta['comic'], 'links': response.meta['links']})

Why call self.parse_comic(response, comic)? I would just do that code inline in parse