It's a similar looking problem. Indeed, it has a similar solution!

I'm thinking of T8221, which was May 19th. Might not be the same bug.

Updated to head—no dice.

It's not, but it's fairly recent. I'm on Chrome and I'll try an upgrade.

I have a vague recollection this was fixed some time ago. What browser are you using and is your Phabricator at HEAD?

also, you can just type :poop: for 💩

@fzhan try bin/cache purge --purge-remarkup first?

Any way to convert existing documentation from Markdown to Remarkup? Thanks.

This seems like a small issue, but since there's absolutely no workaround and it's impossible to post any documentation that makes use of non-trivial tables, I think this is a pretty big deal.

Phame is a prototype application. See this guide for more information:

When you have text selected in a Remarkup text area and click this button:

Sorry, I misunderstood and was using the " Quote menu item on a comment.

Oh! You mean the " symbol above the remarkup text box, right?

I can't reproduce this.

In T8565#121037, @epriestley wrote:

In T8565#121036, @epriestley wrote:

Test.

Test.

In T8565#121036, @epriestley wrote:

Test.

Test.

Hrm. Yeah, T6823 raises some difficulties of finding a good default for everyone I hadn't considered though. Maybe we could support environments with multiple and mixed languages (e.g. sql embedded in python or html/css embedded in php) by parsing language suggestions in-line? Something like:

D13277 touches on another case for this which I hadn't considered in T6823: code blocks in Diviner articles. That one feels pretty relatively strong/compelling to me.

My patch seems to be stuck. Oh well.

In T8430#118598, @chad wrote:

❤️!

See Planning.

Any traction on this task?

Please where you say "To add files, drag and drop them into the comment text area.". be sure to also say what to do if one cannot/does not know how to do that !!!!!!!!!!!!

See D10671 for an example of an inline comment that doesn't wrap.

make the fullscreen mode z-index some smaller value

Just chiming in to say I also have this issue as I use i3. I can drag and drop but it goes against the entire i3/tiling workflow.

Taking a step back here...

This is a z-index problem. (Perhaps duh.)

I do think we can get the scrolling to work well, even if it feels pretty bad with the naive approach, but would have to do something like this:

Thought maybe this is also a lot more work that I thought to even test out, up to @btrahan.

I think you're right on 1), that we may need to pursue a click to preview for alternative cases like mobile. In regards to a 2-up view only being useful in Phriction/Legalpad, I'd probably disagree. Mostly, I don't know when or where I might want to spend an hour typing something out (comment, new task, Pholio, etc) so having something available everywhere would be my preference. In the basic sense, even if I have to manually scroll each window, it's still seems like improvement over the current situation. Or at least, I'd like to feel it out before deciding it was really bad.

After thinking about this for a bit, I sort of question whether this is actually a very good solution. I think there are three root problems:

The only piece there is the "Document Preview" label is either going to be just "Preview" or something per object type, like "Comment Preview" or "Message Preview", etc. I figure that's NBD but I do plan to deviate there.

If the design is missing anything, let me know, though I think this is straightforward interaction-wise.

My attempt(s) yesterday focused around trying to stick the existing preview in the pertinent area via CSS.

Restarting PHP-fpm5 5 did the job.
I now feel extra dumb, not trying to restart PHP (I did it for nginx thought).

Does the error go away when you restart Apache / php-fpm?

Haha, sorry, figured I was doing slightly better than that. This should be easy to fix up.

Well, this is certainly fun drive with policies and edge cases...

D12793 might be missing some edge cases with unusual objects but should fix at least 95% of this. Let me know if you run into more cases after D12793 lands.

T7186 seems like the correct fix, that is, properly detect what you are dropping and do the right thing. This is regardless of drag and drop field or Remarkup Textarea, it's all the same JS. Up to @btrahan if we should pursue better error handling in conjunction.

P.S: Destination was the conpherence room icon / image drag&drop field, not the message entry area.

I posted this primarily, because there are some division by zero errors logged. I guess those could be fixed by improved error handling.