URLs in start_urls are not affected by this middleware

[gfrmin]

If a URL is in start_urls of the spider, it is never skipped by this middleware, i.e. every URL in start_urls is parsed by the parse method of the spider.

If this can't be fixed, it should be documented.

[eliorc]

I had an issue with this also.
Even though all of the start URL were Request objects containing the meta 'deltafetch_key' needed, the were crawled anyway

[feng1632009]

Yes, I also encountered this problem. Is there any good solution?

[gfrmin]

This is now documented under NOTE 1: https://support.scrapinghub.com/support/solutions/articles/22000221912-incremental-crawls-with-scrapy-and-deltafetch-in-scrapy-cloud

[feng1632009]

For the time being, my solution is to set a link that must be accessible at the start of start_url, and then the URL of the yield Request crawler url can be affected by the plugin.