fix deprecated calls to scrapy.utils.request.request_fingerprint
#50
Conversation
| strategy: | ||
| matrix: | ||
| python-version: [3.5, 3.6, 3.7, 3.8, 3.9] | ||
| python-version: ["3.7", "3.8", "3.9", "3.10", "3.11", "3.12", "3.13"] |
There was a problem hiding this comment.
3.5 and 3.6 don't seem to work anymore? at least in my fork
| from scrapy.http import Request | ||
| from scrapy.item import Item | ||
| from scrapy.utils.request import request_fingerprint | ||
| from scrapy.utils.request import fingerprint |
There was a problem hiding this comment.
On Scrapy 2.7+, the right approach is using Crawler.fingerprinter.fingerprint().
There was a problem hiding this comment.
@Gallaecio wouldn't using this require storing an instance of (and instantiating) a Crawler object as an instance variable?
from the messages in scrapy 2.11.2 (https://github.com/scrapy/scrapy/blob/e8cb5a03b382b98f2c8945355076390f708b918d/scrapy/utils/request.py#L86-L136) it seems to suggest getting the crawler during instantiation with DeltaFetch.from_crawler if I am reading it right.
but what if the DeltaFetch object isn't instantiated that way? what should happen in DeltaFetch._get_key?
There was a problem hiding this comment.
DeltaFetch is a spider middleware, instantiated by Scrapy.
Most Scrapy components are instantiated with the create_instance (Scrapy 2.11-) / build_from_crawler (Scrapy 2.12+) functions, which call from_crawler if defined. Spider middlewares are definitely one of those components. So __init__ should never be called without first calling from_crawler.
It does happen in tests, that use self.mwcls(), and would need to change to either use from_crawler or, better yet, use create_instance / build_from_crawler.
2 participants
Currently this plugin is broken on Scrapy 2.12
scrapy.utils.request.request_fingerprintwas removed in Scrapy 2.12 (https://docs.scrapy.org/en/2.12/news.html#deprecation-removals)If accepted, a version bump would be helpful 🙂