Thanks :) tested

Any comment about the first proposal would be awesome. Probably it's not so difficult

Softaculous integrates with various shared hosting systems as an alternative to the native package system within each one.

Yeah, please integrate this kind of info in your answer - this is just a comment (same is true in StackOverflow)

The relevant part seems to be

As stated, some documentation is available for package authoring.

Yeah don't worry, but it's just to say that this is not actionable without any documentation. If you can, contact them, to publish that information at least.

My relation to Softaculous is its general availability to customers who purchase shared hosting, through cPanel or Plesk.

Also Open software can be commercial. For example, Nextcloud offers commercial assistance. Softaculous instead is "just" a proprietary software. So, this topic it's borderline as spam, since there is no margin of contribution there from anybody that does not work for Softaculus.

Are you using Phorge? Administrating it?
Are you related to Softaculous?

In D25021#10250, @vedranmiletic wrote:

So you would prefer system-ui to mean Roboto instead of, I presume, DejaVu Sans Condensed? Have you tried setting ~/.config/fontconfig/fonts.conf like:

<?xml version="1.0"?>
[…]

In D25021#10248, @kwisatz wrote:

My humble opinion is that the the version that uses system-ui does not look good (The kerning is off, some fonts look too heavy, etc).

In D25021#10246, @vedranmiletic wrote:

In any case, Phabricator now has roughly the same font list as Bootstrap 5. Does Bootstrap 5 look bad on Debian?

Worked

In D25021#10244, @avivey wrote:

Looks like the original request was because things looked extremely bad on Garuda Linux which was picking Segoe UI Symbol for everything. I guess it's also possible that it was a single bad install, because why would anyone do that?

Turns out the "Create Security Task" form doesn't allow users to create tasks, because they can't see the S2 space.

In D25021#10244, @avivey wrote:

I've tried sort of reverting this diff on my system, but not knowing how to make celerity regenerate what is finally being served to clients, had little to no success.

(To regen the celerity maps, you can use ./bin/celerity map if your code is new enough - I think it's 2-3 months old).

In D25021#10243, @kwisatz wrote:

...but it looks really bad to all of us here at the office (all Debian machines).
We all preferred the previous configuration which was ending up being Lato for us.

I've tried sort of reverting this diff on my system, but not knowing how to make celerity regenerate what is finally being served to clients, had little to no success.

In D25021#1763, @Cigaryno wrote:

The Ubuntu font is not used and looks to me terrible.

Please share the output of this:

I mark this question as resolved since it has at least one answer. Feel free to provide more information to help other people to add other answers later :)

Suggested meme: whatcouldgowrong

simplify check

remove unuseful newline - sorry

• remove unuseful comment

• use a single variable $classes
• remove unuseful comment
• recognize ./ and ../ as internal
• remove isURIStartingWith()

Created R7 arc-vscode, but it's currently Observing the GH one.

Feel free to provide at least some documentation in your question so to help more people in sharing some useful answers

I have created T15559 with extra information, now that D25342 was closed and the issue behavior changed somewhat.

I picked this up again recently. I’m stuck on getting mariadb valid certificates it uses for connections, for testing my Phorge changes.

PhutilURI is obviously a better type for href then string. It should not log anything in this case...

We don't allow cross-application table joins

We don't allow cross-application table joins

This is totally scarying but I'm quite interested in exploring the proposed solution :)

Created Semi Structured (Extension) and R6 Semi Structured:

I went to https://we.phorge.it/D25018 and that URL says:
Received unexpected type for href: PhutilURI. Please paste this log as comment in https://we.phorge.it/T15316

Probably, the "color change" should be handled as "expanded transaction" - whatever it means - so that it changes only after its related cause (status change to → archived) and not always as long as the current value is archived.

remove bool

Thanks for finding that! Yes, makes a lot of sense to reduce that scope

Make change less intrusive/dangerous by reducing CSS scope to "type=file" input

Thanks! Now the second step is, finding at least one file that is related to the generation of the above feeds / texts

This is partially related to Comments

I'm not able to provide a smart title for this Task :D Thanks for any help in this

We can continue the discussion in Task T15556 - thanks for reporting

Feel free to elevate this as a Task, since it's a nice bug report

+1