Adrián Chaves

Yes, I believe that last behavior is also known, although opening a (separate) feature request to make it more smart is probably OK, and maybe a good first issue (and...

The original issue could be addressed either by making `--output-format` work with `-O`, by having the command fail when those are combined, or by making it so that the warning...

I don’t think it is a common problem, I think a specific solution is the best approach here.

I do wonder if maybe it makes sense to implement this along with the changes to cache handling that make use of this.

I’m unsure (+0) on whether or not it’s fine. I do wonder if merging this change would affect https://github.com/scrapy/scrapy/pull/5175 in any way (as in https://github.com/scrapy/scrapy/pull/5175 could be implemented differently if...

@elacuesta What would you think of extracting a subset of these changes into a separate pull request, specifically `Response.attributes` and its use in `replace`, to address https://github.com/scrapy/scrapy/issues/1877? I suspect we’ll...

It would be great if we can include tests which cover the missing parts from https://codecov.io/gh/scrapy/scrapy/compare/febe82a9076d37ccc0b09a0ad97c8ae6ee1bd1b1...cab1207ce6add6e87a58a392b1c9411322a40dbf/diff

Note to self: update those Sphinx version directives when preparing the release notes.

We should also make sure any new implementation still supports both [crawl orders](https://docs.scrapy.org/en/latest/faq.html#does-scrapy-crawl-in-breadth-first-or-depth-first-order).

Today, working on speeding up a broad crawl project, I found that applying @whalebot-helmsman’s downloader-aware priority queue to `start_requests` can make a difference, because unlike requests found while crawling, start...