Mikhail Korobov

In other words, why does crawler need a logger and stats as soon as it is initialized? Can it do anything useful before `.crawl()` call?

> One thing which looks important to me - setup logging before Spider.**init** `Spider.__init__` is called in `Crawler.crawl`, so if we move logging setup to `Crawler.crawl` then `Spider.__init__` can still...

> Another option is to do better job in separating "core" settings which must not be changed by spider from extensions/middlewares/pipelines/addons settings that can be changed by spider. I like...

hey @gekco! this is almost fixed in https://github.com/scrapy/scrapy/pull/3111, the remaining issue is extra tests which are executed unintentionally.

Could you please explain, what's the motivation for using stalebot? I never understood it :) For me closing stale PRs and issues automatically looks like an anti-pattern. It feels hostile...

Hey! As discussed with @VMRuiz on a call, it seems we should have an API proposal for the complete solution, before we can merge a PR like this.

> convert README.rst to README.md Hey! I'm not sure about converting readme to Markdown, because all other Scrapy documentation uses ReStructuredText. > when a link is clicked it should direct...

hey! I'm fine with having me on a list, with Scrapinghub Inc. affiliation if that's not too much work. Thanks!

My 2c: measuring contribution by LoC or other "objective", automated metrics is misleading. Imagine your employer to implement such evaluation against you. For me it is obvious that @noamraph, as...

FTR: this is how to filter out these messages ``` logging.getLogger("scrapy.spidermiddlewares.depth").setLevel(logging.INFO) ``` this should be called after the Scrapy logging is configured, e.g. in a spider.