Uh interesting - good to know for the future. So at least the button Mark All Read here in F3271029 should be visible for your case - as very minimal mitigation sub-task

The current workaround is to wait for a real notification, to then finally be able to click the Mark All Read.

It happened to me but only in Wikimedia Phabricator, there I had a notification counter in the top-bar, but with no unread messages in the related list

Thaanks - please wait 7 days + 1 hour + 1 second + 42 milliseconds before landing to attract more Pòkèmòn reviewers

(M is short for Mockup probably)

In T15450#16936, @Iniquity wrote:

I agree that an additional status is needed for closing when creating a task. Current statuses are not obvious

Uh! That's magic. Thanks avivey. Created here: H29 Phorge: add comment to remember "celerity map" - T15209

Ah, it also works on my computer but really not relevant, just the patch description was in line with my business expectations

This is EXACTLY the kind of serious-business patch description that should be mandatory in every Phorge patch to speedup code review. Thanks my friend. asd

Given the fact that the original comment from the kind @speck seems now fully satisfied, I think there are no problems to land this :) Thanks again to speck for that proposal ( https://we.phorge.it/D25022#768 ) - so, I flag that as "done" very boldly.

P.S. - useful snippet shared by Daimona from Wikimedia:

Increasing a bit the priority since this is feasible, simple to reproduce, and really frustrating in 2025.

P.S. welcome in the family of Trusted Contributors and feel free to escalate this question as a new task under Bug Reports and by marking this question as obsolete in case - thanks again

Thanks for the question. I will follow answers. BTW I've found the nuked link:

Thanksss

In the meanwhile I paste here a proposed alternative that may better reflect the current situation (no need to update the patch - since I guess this phrase will attract more changes):

I agree on discussing the removal of this phrase. I will wait the task, but my opinion is that it's a legacy phrase from a company that had better to do than fixing weird workflows based on Prototypes (and it had sense). But now it's a community and we already work on best effort on everything, including prototypes. Moreover that phrase does not reflect the current situation, since we triaged and fixed 6 bugs on the Calendar prototype Calendar (even if it should be un-prototyped one day) - and probably more evident already-reviewed patches.

That's a good question. In general in Phorge/Phabricator any archived thing is editable. So I would say no, that archiving a chat still allows to write a message. But maybe it should not generate notifications.

@Cigaryno does it work also applying this D25535 ? Thanks Cigaryno

(Premising that I'm affected by T15985 lol - unrelated)

I think this is the upstream bug report lol

In D25859#24307, @mainframe98 wrote:

Shall I just land this then?

Thanks

Probably no need for a warning. Thanks

Since using E_STRICT is deprecated, I thought the simple fact of using it from whatever Arcanist CLI (e.g. arc anoid) was triggering this new warning

What do you think about @mturdus?

You know better than me the context. Is there a possibility that an installation is already without that column?

OK lol no, this has nothing to do with D25898. Need another task for the "Missing input descriptions"

In D25898#24165, @avivey wrote:

• There's already a "description" field on these, which I think was supposed to be displayed; We might have lost it at some point, which is concerning.

have an approved Revision", but this too can be tricked by a sufficiently competent user

Hackerman meme

clarify how to run the command (last change promise)

Aldo I may need extra help from @waldyrious asdlol

aaand I go to bed :3

OK my friend, I've worked a lot to make it shorter... but still more dense of useful info, reducing jokes at minimum.

Dear friend @mainframe98 - I suggest to click the button Hide "Done" inlines in the top-right corner to easily do another reading

You mean a shorthand for git revert + arc diff? Sounds reasonable

Potentially related to Herald. Adding Herald to raise more attention.

Tried. I Shall Not Pass :O

@avivey what do you think about? Maybe we can test with some powerful clicks

I confirm that ./bin/celerity map does nothing after this change

This makes me and probably other people mad lol

Aaand last 48 lines checked. Differential is crashing in my laptop lol

(Review 3 of ?)

P.S. maybe nice to introduce in AphrontView a method like getUserRequired() or something similar to implement exactly this logic - so we avoid that scattered logic everywhere

(epic review part 1 of ?)

and reviews love you too! (lol)

In D25872#23861, @nib wrote:

@aklapper Would it make sense to move the branch-naming lines into a function in class ArcanistGitWorkEngine, since it the branch-naming is git-specific? E.g.

public function buildBranchName($task_ref) {
  $task_name = $task_ref->getName();
  // regex, preg_replace, trim, etc
  return $branch_name;
}

I don't know if anybody is supposed to run

Small question. Is it supposed to be used with null+null?

Maybe also just declaring $path = '' as default in 209

I've tested this and I'm impressed. You probably have changed my life.

Gotcha. Any opinion on this?

Maybe duplicate patch, look for same title (I'm on mobile now)

Finally something that can be tested using arc anoid - I guess?

Test plan works! Thanks :3

Hi @nib - still thanks and you are welcome in updating this patch using something like this:

last cosmetic change on a title with extra "==="

Question: in an ideal world would you love even more the same button, that "just" prefills the textarea? That could be also probably done server-side, without JavaScript. That would help T15991 even more I think.

add missing comma

implement tips, simplify (thanks)