Document CloseSpider extension better

[redapple]

See #2748 (comment)

@kmike 's details:

When CLOSERSPIDER_ITEMCOUNT limit is reached, Scrapy stops the spider gracefully; it means requests which are currently in the downloader queue (up to CONCURRENT_REQUEST requests) are still processed. This is the expected behavior.

[redapple]

Closed via #2777

[OfirD1]

See #2748 (comment)

@kmike 's details:

When CLOSERSPIDER_ITEMCOUNT limit is reached, Scrapy stops the spider gracefully; it means requests which are currently in the downloader queue (up to CONCURRENT_REQUEST requests) are still processed. This is the expected behavior.

I think it's worth adding the same documentation for CLOSESPIDER_PAGECOUNT and CLOSESPIDER_ERRORCOUNT. If it were there, I wouldn't have to test it myself to verify :)

[eLRuLL]

@OfirD1 yeah I think we should put this part:

Requests which are currently in the downloader queue (up to :setting: CONCURRENT_REQUESTS requests) are still processed.

in a more generic part, as all closing behaviors still continue processing pending requests.

[Gallaecio]

@OfirD1 Could you please open a new ticket about it?

[OfirD1]

@Gallaecio, you mean open a new issue? or just edit it here and hit "propose changes"? I'm not sure what "open a new ticket" is, but I want to learn!

[Gallaecio]

Sorry, I meant a new issue.

You can of course create a pull request the way you describe if you are willing to handle the change yourself. But if you do it that way, make sure you replace 2.3 with master in the URL.

[OfirD1]

Done.

#4836