scrapy-poet is the web-poet Page Object pattern implementation for Scrapy. scrapy-poet allows to write spiders where extraction logic is separated from the crawling one. With scrapy-poet is possible to make a single spider that supports many sites with different layouts.

Read the documentation for more information.

License is BSD 3-clause.

• Documentation: https://scrapy-poet.readthedocs.io
• Source code: https://github.com/scrapinghub/scrapy-poet
• Issue tracker: https://github.com/scrapinghub/scrapy-poet/issues

pip install scrapy-poet

Requires Python 3.8+ and Scrapy >= 2.6.0.

Add the following inside Scrapy's settings.py file:

DOWNLOADER_MIDDLEWARES = {
    "scrapy_poet.InjectionMiddleware": 543,
    "scrapy.downloadermiddlewares.stats.DownloaderStats": None,
    "scrapy_poet.DownloaderStatsMiddleware": 850,
}
SPIDER_MIDDLEWARES = {
    "scrapy_poet.RetryMiddleware": 275,
}
REQUEST_FINGERPRINTER_CLASS = "scrapy_poet.ScrapyPoetRequestFingerprinter"

Setup your local Python environment via:

1. pip install -r requirements-dev.txt
2. pre-commit install

Now everytime you perform a git commit, these tools will run against the staged files:

• black
• isort
• flake8

You can also directly invoke pre-commit run --all-files or tox -e linters to run them without performing a commit.