Sorry, we don't offer help with third-party application development!

Probably not. Because it's a scalability issue this does get some amount of internal priority, but we have a large amount of external prioritized work in queue ahead of it (see Prioritized).

Not yet, but soon. New code should use project.edit.

Did the project.create API deprecated?

T10423 was resolved indirectly (in connection with T11524) so we've sort of inched forward here.

Try using project.edit instead.

I'd like to sort out T9365 / T10635 (better aggregation and/or query options for unit/coverage data) first, since that likely impacts how we GC this data. Ideally, we'd leave an aggregated stub behind (e.g., pass/failure count) rather than nuking the data completely.

Specifically, my plan here is:

du -sh returns 1.3GB for that repository so it should be below that limit.

GHC appears to be running into this limitation. We would prefer to use anonymous HTTP access with our build bots; is there any hope that this will be fixed in the near-term?

We can build this first, then backport it to Doorkeeper and feed.http-hooks, so this doesn't need to be an enormous mess.

Curious, Did you follow the setup guide that includes the rewrite rules? Check your apache / nginx configuration

Yes that would do it.

If we just added an action to Herald like Queue Call to Webhook: http://whatever.derp/, would that cover this? I think that's often kind-of-a-bit-hacky, but probably makes a lot of these soft use cases easier.

I'm a little wary about this. Some of my concerns:

I think this is all generally reasonable to bring upstream, at least in spirit, but I want to wait for planned arc changes (overviewed in T11429) to get a little further along and build it on top of that.

I think you should already be able to add subscribers and projects (and take some other actions) with Herald, by writing a "Global" rule using the "Receiving email address" condition. Here's an example:

Going to merge this into v2 Dashboards so if that does ever move forward, we'll have your feedback.

Thanks for the suggestion. We have no current plans to pursue this. Please use the link provided to see the full queries results.

What do you mean that stuff? normalizeInput?

This stack overflow post suggests that with apache you could use something like:

Go blame that stuff and try to figure out why it's as early as it is.

I had some family stuff come up and was only able to spend about three hours on this before Ludum Dare ended, and it never turned into anything playable. I got as far as letting you jump between systems and zoom into planetary bodies:

This is almost certainly T4369. I would expect that this repository is larger than 2GB?

In T11517#192337, @rayzyar wrote:

In T11517#192332, @michaeljs1990 wrote:

I've just tried to reproduce again, seems my steps cannot reproduce it.
I couldn't find out a good way to reproduce this bug.
Here is what I found :
when I change my existing code (removing the comment of an exported variable or method), arc lint at root git directory doesn't trigger the ADVICE
however arc lint xxx/xxx.go could trigger the lint ADVICE

And I tried to copy the same code to try to reproduce this, it works perfectly with arc lint at root git directory.
BTW, I am using some extension, https://github.com/zerodiff/traphic

Thank you a lot for looking into this

In T11517#192332, @michaeljs1990 wrote:

I've just tried to reproduce again, seems my steps cannot reproduce it.
I couldn't find out a good way to reproduce this bug.
Here is what I found :
when I change my existing code (removing the comment of an exported variable or method), arc lint at root git directory doesn't trigger the ADVICE
however arc lint xxx/xxx.go could trigger the lint ADVICE

You are going to have to post more information. I am able to correctly see the error you posted following these steps.

Can you post your arclint file

See Planning for timelines on how we work. If this issue is urgent and blocking your company, see Support Resources for prioritization options.

@chad could u take a look again

I put up a diff that fixes this however I'm not sure if it's the correct solution.

In T11535#191978, @epriestley wrote:

Why are you tracking branches you don't want to land onto?

this is absolutely not a bug in libphutil it is with lighttpd. I will post a small example above showing you why.