Rust Conduit Client/R4 is the first project in Phactory!

I created Rust Conduit Client and (unnamed) R4.

Cool. Let's start with rRustConduit, because I think there's a bad joke in there about rusted pipes (12 seconds of google did not find a special name for it).

By not signing a CLA, you'd have to explicitly set a license for each contribution you make - i.e., every single revision would have to include an explicit license, or it could not be incorporated into the main code-base.
This is technically handled by item 5 in our license, but it's reasonable to assume nobody ever reads that.

In D25178#5455, @valerio.bozzolan wrote:

While waiting for a reviewer, feel free to enjoy and sign L1 Phorge Vision Statement

I think you can get there using herald:

A change without a test plan is just a bug report waiting to happen.

Why won't this software run with safe_mode enabled?

1. please add phlogs in setter (or whatever) about expected types
2. please remember this issue in all similar changes.

putting it in the setter is fine...

I'd argue that almost any use of phutil_nonempty_stringlike() is hiding a bug, because it hides an object we don't know the type of - which is by itself a problem.

a phlog would be useful, because anything other than string or PhutilUri would be very surprising, and possibly a bug. My desire is to expose bugs as much as possible, not hide them.

I'd like to know more about why this method even exists, and what other things might end up in href.

href can be an object - PhutilUri, and maybe some other things.
what is this method for anyway? ie what usecases is a PhuiTag not know what kind of tag it is?

@ton: Why are "they" not using arc patch to download the patch?

Fixed in rP74dfe6f971af.

In D25139#4421, @valerio.bozzolan wrote:

Even if probably it would be nice to have an isAbsoluteURI() method in PhutilURI to reduce the business logic here

lint

Thanks - looks like this might have been introduced in D25131.

Can you post the entire stack trace please?
(We use 7.4 on this server too).

(there used to be a "feature flags" application in Phabricator,

mm, no, this is a new break in latest master.

I think the unit tests will pass if you just rebase off of the current master.

I think the future() I linked for literally calls ./bin/repository update ..., and sets the timeout on that.
That invokes the future() you found, which calls hg pull, and adds another timeout.

I can no longer reproduce this, so I'm guessing it was something bad in my setup.

Looking at the diff, I think this timeout only applies to Drydock cloning the repository for its builds, not for "observing" repositories - and Drydock might not support Mercurial anyway.

I've added H22, but I think at this point we should start writing code for what exactly is allowed/not allowed for these two branches.
It's also kind of annoying that there's no way to use the same rule for both repos.

We've sort of reached a status-quo:

• Trusted Contributors is basically "community members" - with a low bar for entry
• Blessed Committers is the next level up, and can basically push whatever code they want
• admin/root/security is basically "core team".

In D25131#4070, @aklapper wrote:

Garr, this includes D25130. I am not a huge fan of arc's "commit everything or nothing" though I did use arc diff --create already.

T15011: Build a Phorge Developer Environment

I think we're more-or-less done here...

In D25125#3939, @goddenrich wrote:

do you think https://we.phorge.it/book/contrib/article/contributing_code/ needs updating too? I went there rather than to bug filing because I thought I would take a crack at raising a fix hence why I din't go down the bug report route

In T15249#6391, @ton wrote:

Ok, after spending a weekend looking at PHP I can say that I overestimated my level of enthusiasm.

You can configure remarkup.ignored-object-names (in /config) to just ignore specific words like S3.

You can configure remarkup.ignored-object-names (in /config) to just ignore specific words like S3.