I work at Uber ATC and we kinda like Phabricator.
- User Since
- Aug 25 2012, 1:10 AM (235 w, 8 h)
Thu, Feb 16
Fri, Feb 10
Wed, Feb 8
Yeah I clicked "Delete File" and can't see it anywhere now.
Edited to not give out my credit card number on the internet.
FWIW, this is the thing that I point someone at when I want to show them how to contribute changes to Phabricator, like just go through the workflow.
Tue, Feb 7
Jan 4 2017
This might break our stuff, test async diff before deploying upstream updates.
Dec 9 2016
At worst we can just add another API method to our install, like, maniphest.reallysearchtitle
I think we are doing this all downstream.
I assume everyone is fully aware of the caveats and just being lazy and not wanting to store the identifier they get back when they create the task? I will inquire further.
Is Phabricator not a suitable replacement for a Jiffy Lube inventory terminal?
Because people are filing tasks from external systems in some systematic way (consistent titles) not saving the ID, and then wanting to reference the task later.
Dec 5 2016
This seems like a good plan to me, we're just checking in that this isn't totally outside the realm of possibility while we plan some super secret changes to our instance.
Nov 23 2016
Nov 22 2016
Nov 16 2016
Nov 10 2016
Oct 13 2016
use array() instead of inheriting validation errors
remove class from conduit fields
Oct 11 2016
Oct 7 2016
This seems to work afaik. Still need to do custom fields and try this in conjunction.
- Fix display of quoted transaction values
- Remove unneeded empty checks for create transactions
- Remove getRequiredHandlePHIDs and transactionHasEffect methods
- Use generateOldValue instead of getOldValue when applying external effects
Oct 3 2016
I still have stylistic stuff to do throughout, I just focused on faithfully moving functionality out of the editor/transaction into the modular transactions here. There are a few things that don't work, haven't even looked at custom fields yet.
finishing the gruntwork
Sep 28 2016
Sep 26 2016
(not building locally anymore, eg, any time I want to do anything with a built binary at all, i'm actually doing it with a binary on a remote server because I can get the build results faster)
Well one facet of this is that remote servers tend to be significantly more powerful than developer machines, and so since I've implemented this downstream, I have basically taken to not building locally anymore.
Sep 24 2016
Don't yet presume that users will be lease owners
Removing all of the *.search stuff and leaving the rest to make it easier on future diff archaeologists
Our actual internal use case is as follows:
Sep 21 2016
@epriestley / @chad do you have pending comments on this? I can split it up into "obvious improvements", like bundling together search engine / dao / API methods per-dao. And then leave drydock.createlease and drydock.destroylease in their own diffs since I imagine they are the most contentious.
Sep 20 2016
Sep 19 2016
I may have misunderstood you, were you advocating for application capabilities instead of authorizations?
Edit engine is deceitfully magical, and you may have indeed done everything you needed to here.
Sep 18 2016
From my perspective, it makes sense for it to be the user requesting authorization, for our use cases a user (or a bot user in the case of other services consuming this) should be the "thing" that is requesting authorization.
Sep 17 2016
@epriestley couple thoughts:
fix lint problem
Sep 12 2016
The bash workflow is just loaded the same way all of the other ssh workflows are loaded, I think that the problem is one of security policy / segmentation.
Maybe make it so that we can host an arbitrary number of ssh daemons on different ports with different entry scripts? I suppose we can kinda do that already by just copying the existing script, minus the part where we have different groups of ssh keys, but could probably sub in passphrase as the backing for authentication there.
Oh nice, I was expecting the "this is awful" upstream response for this :) I can put together use cases and either @jcox or I can work on upstreaming it.
Below are two extension classes which allow our users to lease and execute commands on drydock machines. This is highly useful for scripting, and looks basically like:
Sep 2 2016
In theory we could make it so that steps resume correctly, but in practice I have very little control over the contents of the scripts that get run. People tend to wrap all of the stuff up that their build is supposed to do into a single script, and assume they're starting fresh each time it's executed.
use pretty numbers when displaying build count
I'm not sure how this wasn't reported earlier, I guess the module cache prevented everyone from encountering it?
Aug 30 2016
If you only knew how lucky you are to have Evan explain this to you rather than trolling through grep -Rn "PhutilClassMapQuery" * --include \*.php errday.
need some arc flow up in here
Live, armed, pre-commit A/B testing is unambiguously the most exciting part of our jobs.
Should we implement this as an internal search attachment in the meantime? Or is there a possible pathway for @jcox to get something like this working in ~3 days?