In D25546#16489, @valerio.bozzolan wrote:

Maybe we can add our dependency:

(new PhabricatorApplicationTransactionCommentView())
  ->setEngine($engine)
  ->render();

Write Phorge application extension which uses backticked strings containing spaces.

Did my previous comment sufficiently elaborate that this is not a change in behavior? If the comment did not, please speak up before I may merge this. :) Thanks!

(I took the liberty to link the specific change in Git in the commit message)

That's due to https://git.kernel.org/pub/scm/git/git.git/commit/revision.c?id=71a1e94821666909b7b2bd62a36244c601f8430e

I still think that the two Subscriber Rule value should be a list strings should for consistency be Add subscriber rule value should be a list respectively Remove subscriber rule value should be a list.

UI wise this seems to work as expected, and I think there's no way to move stuff on a workboard via Conduit. Just needs some smaller string adjustments.

Small capitalization nitpick, seems all other options use sentence case instead of title case:

git rebase master

Make lint happy

Thanks! Definitely way more correct :) Feel free to arc land

and I'm sorry that this is becoming so time expensive

I guess another question is whether to define an abstract public function getCommentFieldPlaceholderText() in parent PhabricatorEditEngine and then add to all and every child EditEngine a default public function getCommentFieldPlaceholderText($object_phid) { return '' };. Feels cleaner to me.

Abstract logic: Move application specific conditions to display a comment field placeholder text into the corresponding EditEngine of the object that the comment field is displayed in, by defining a public getCommentFieldPlaceholderText() function.

In D25546#15665, @valerio.bozzolan wrote:

I also don't get if in this part we have access to any thing related to Maniphest, ManiphestEditEngine or this kind of stuff, to just call $something->getCommentPanelPlaceholder() and have the business logic there in the specific application itself.

In D25505#14536, @avivey wrote:

Alternatively, maybe catch the existing exception and show a nice error message + the form?

+celerity map

If it's that easy

FTR I gave this last version another test locally, feeding an mbox file with no subject header (which is accepted now without complaints), and bugging input again (which prints a correct error message instead of misleadingly complaining about a missing subject line)

Simplify logic (thanks Mukunda!)

In D25568#16354, @valerio.bozzolan wrote:

In this way, do we skip that column creating a column shift?

Maybe we need to promote that NULL to empty string

In T15670#16064, @valerio.bozzolan wrote:

A root problem is that highlighted line number(s) should be a # fragment really, to do not multiply pages exponentially.

IIUC this adds Flags in Differential, Maniphest, etc list views, and adds the repository in Differential list view?

Remove another change which is unneeded now

Remove two changes from patch which got already fixed in the meantime; rebase

Abandoning per last comment

Abandoning per last comment

Argh, sorry!

Correct array initialization; keep line in parse().

and now for real, without #[\AllowDynamicProperties]

In D25563#16157, @avivey wrote:

Can we just define the used properties instead? Doesn't look like there's too many of them.

LGTM