This is a scrapy project I did for learning. The spider downloads and
scrap Brazilian Lottery's results. Each lottery results is published 
as HTML files inside a .zip file. This project will teach you how to
use scrapy to process .zip files, how to organize multiples pipelines
and how to produce JSON output from Items()

The URLs of each lottery and the spider code itself is defined in 
    hacksena/spiders/hacksena_spider.py

The Spider yields a FiledownloadItem(Item) defined in
    hacksena/items.py

Then 3 different pipelines are executed in a given sequence, 
check hacksena/settings.py to see the definitions. 

The sequence is:

1) HacksenaPipeline(object):
    This pipeline extract the zip file content (the HTML file),
    returning Item() with file_data = Selector()

2) ResultsPipeline(object):
    This pipeline grab the HTML extracted previously, use the Selector() 
    to extratec the relevant parts and returns ResultItem() 
    in preparition to write down a JSON file

3) JsonWriterPipeline(object):
    This pipeline effective write down the JSON output                                                                                                                                               

I hope you enjoy scrapy too as I am enjoying it :)

Carlos.