@valerio.bozzolan: Thank you! I know this is super boring work, but it allows PHPStan to show other type/class issues which are currently hidden due to PHPStan barking earlier about that non-standard question mark when parsing those lines. :-/

Abandoning in favor of T15890 and D25774 per previous discussion

Abandoning in favor of T15890 and D25774 per previous discussion

I confirm with my hand on my heart that I've consumed my eyes inspecting every single line to be absolutely sure that this PHPDoc probably has a better meaning and will not cause nuclear implosions.

In T15848#19430, @valerio.bozzolan wrote:

I would like to better understand the root problem.

Why changing the current user agent should improve the situation?

OK thanks (I need some rest)

In D25758#21773, @valerio.bozzolan wrote:

Maybe better to discuss a little bit in T15848: newOAuth1Future doesn't set a User-Agent since I'm curious but also confused

Not exactly sure what the confusion is. The patch and task are clear to me and solve the issue.

Moreover traditionally, the user agent is the name of the software, not the origin. So this could also deserve another comment about why we should replace the specific software name with a fixed origin. Premising that I agree that the current user agent may be re-discussed, to include more information, like, "Phorge", and the version of Phorge.

Actually no, user agents for automated tooling often includes some sort of contact email or website or link to a page describing the bot/tool. I think this should be treated like that.

Also, in case, maybe probably nice to have a getUserAgent(__CLASS__) method in parent class PhutilAuthAdapter()

@valerio.bozzolan: thats not what Andre says. He says another area of the code uses a terrible user agent and we should change that to a better one at the same time.

Maybe better to discuss a little bit in T15848: newOAuth1Future doesn't set a User-Agent since I'm curious but also confused

I would like to better understand the root problem. I see that Cloudflare was showing a captcha to Phorge (what Phorge?). It seems it was because an user agent was missing. But aklapper said that we are already setting an user agent.

I just discovered that sometime we can spawn warnings above the comment box.

güncelleme

(In general thanks and please do whatever needed change in PHPDoc but let's reduce the commits on single files for different PHPDoc changes; so if you have lot of warnings on a single file, just do them all thanksl

This sounds like something that folks at Facebook were overriding / patching. But, the class is final so it's difficult to believe that really somebody is using this.

Leave out special case src/workflow/ArcanistWorkflow.php for now

I'd prefer not to increase the diff to upstream (as a side effect of this very task) but instead to pull a more recent upstream version - see T15940.

fix typos in documentation

add a bit of documentation - just because we can

Right, should have probably also covered that line in D25186 which is more or less the same...

In D25535#15480, @avivey wrote:

Maybe this is wild, but: should we consider removing the compiled xhpast.exe file from the repository, and host is externally?

Ty. I’m still confused how “limit” is required but doesn’t throw an error when not specified. Can address that later if needed.