search

dmclain / scrapy-heroku

BSD 3-Clause "New" or "Revised" License
68 stars 36 forks source link

readme

Scrapy-Heroku

A package to assist with running scrapy on heroku. This is accomplished by providing a custom application configuration at scrapy_heroku.app.application that launches the scrapyd web service using the PORT environment variable and a multi-process work queue implemented on a Postgres database specified by the DATABASE_URL environment variable.

Configuration

Create a git repo that has a scrapy project at the root (scrapy.cfg should be at the top level). Edit your scrapy.cfg to include the following::

[scrapyd]
application = scrapy_heroku.app.application

[deploy]
url = http://<YOUR_HEROKU_APP_NAME>.herokuapp.com:80/
project = <YOUR_PROJECT_NAME>
username = <A_USER_NAME>
password = <A_PASSWORD>

Add a requirements.txt file that includes scrapy, scrapy-heroku, and scrapyd. It is strongly recommended that you version pin scrapy-heroku as well as the version of scrapy that your project is developed against (pip freeze > requirements.txt).

For Example:

# requirements.txt
Scrapy==0.24.4
scrapyd==1.0.1
scrapy-heroku==0.7.1

Finally create a Procfile that consists of::

web: scrapyd

Make sure you have a postgres database with the DATABASE_URL env parameter set.