CODEWITHSUNDEEP
scrapypipeline
GMBill
GMBill

Reputation: 338

Scrapy Pipeline loads but doesn't work

I have a Scrapy project that loads the pipelines but doesn't pass items to them. Any help is appreciated.

A stripped down version of the spider:

#imports
class MySpider(CrawlSpider):
  #RULES AND STUFF 

  def parse_item(self, response):
    '''Takes HTML response and turns it into an item ready for database.  I hope.
    '''
    #A LOT OF CODE
    return item

At this point printing out the item produces the anticipated result and settings.py is straightforward enough:

ITEM_PIPELINES = [
  'mySpider.pipelines.MySpiderPipeline',
  'mySpider.pipelines.PipeCleaner',
  'mySpider.pipelines.DBWriter',
]

and the pipeline seems correct (sans imports):

class MySpiderPipeline(object):
  def process_item(self, item, spider):
    print 'PIPELINE: got ', item['name']
    return item

class DBWriter(object):
  """Writes each item to a DB.  I hope.
  """
  def __init__(self):
    self.dbpool = adbapi.ConnectionPool('MySQLdb'
                                        , host=settings['HOST']
                                        , port=int(settings['PORT'])
                                        , user=settings['USER']
                                        , passwd=settings['PASS']
                                        , db=settings['BASE']
                                        , cursorclass=MySQLdb.cursors.DictCursor
                                        , charset='utf8'
                                        , use_unicode=True
                                        )
    print('init DBWriter')

  def process_item(self, item, spider):
    print 'DBWriter process_item'
    query = self.dbpool.runInteraction(self._insert, item)
    query.addErrback(self.handle_error)
    return item

  def _insert(self, tx, item):
    print 'DBWriter _insert'
    # A LOT OF UNRELATED CODE HERE
    return item

class PipeCleaner(object):
  def __init__(self):
    print 'Cleaning these pipes.'

  def process_item(self, item, spider):
    print item['name'], ' is cleeeeaaaaannn!!'
    return item

When I run the spider, I get this output at startup:

Cleaning these pipes.
init DBWriter
2012-10-23 15:30:04-0400 [scrapy] DEBUG: Enabled item pipelines: MySpiderPipeline, PipeCleaner, DBWriter

Unlike their init clauses that do print to the screen when the crawler is started, the process_item methods are not printing (or processing) anything. I'm crossing my fingers that I've forgotten something very simple.

Upvotes: 3

Views: 3388

Answers (2)

you_me_on
you_me_on

Reputation: 61

"Better late than never"

#imports
class MySpider(CrawlSpider):
  #RULES AND STUFF 

  def parse_item(self, response):
    '''Takes HTML response and turns it into an item ready for database.  I hope.
    '''
    #A LOT OF CODE
    yield item       <------- yield instead of return

Upvotes: 1

akhter wahab
akhter wahab

Reputation: 4085

2012-10-23 15:30:04-0400 [scrapy] DEBUG: Enabled item pipelines: MySpiderPipeline, PipeCleaner, DBWriter

this line shows that your pipeline are initializing and they are ok.

problem is is your crawler class , 

class MySpider(CrawlSpider):
  #RULES AND STUFF 

  def parse_item(self, response):
    '''Takes HTML response and turns it into an item ready for database.  I hope.
    '''
    #A LOT OF CODE
    # before returning item , print it 
    return item

i think you should print an item , before returning it from MySpider.

Upvotes: 1

Related Questions