#scrapy

Hello :)

Can anyone explain why I'm getting an error AttributeError: 'NoneType' object has no attribute 'iterkeys' only when adding a json export to my crawl? ( -o test.json ) I do have custom pipelines to put everything into a DB

but also want it to output to .json but the file has nothing in it ( just some spaces and , ] )

woo fixed it, I had added another pipeline and forgot to return the item

:D

Can some one give me xpath to pick first href link from each row in the table in the html from pastebin shared above?

I use this - "//a[contains(text(),"Requisition Number")]/ancestor::tr/following-sibling::tr/td[1]/a"

but it is very risky. One change to the html structure and i have to rewrite it again