(I think that title edit is accurate, let me know if I misunderstood.)

This probably isn't directly related to PHP7.

See Also: T9640

Can you show me the output of this command on your system?

Yep, that works. I didn't realize that null was a possible return value.

This isn't a bug, there are just no fields available.

I didn't purge the cache here, but here's an updated version of the original diff without the rendering error:

I wouldn't have expected the issue to be in xhpast at all. Thanks for the quick resolution!

Thanks, that test case was very helpful.

This one's slightly trickier than T10570 because leading whitespace is semantic in lists.

What version is libphutil at? This function was introduced in 821cad81.

Interesting. I just went and tried it again and different times its failing in different ways. I'm going to give up and assume this is something wonky in my browser until I prove otherwise.

1. Go to a project, e.g., https://secure.phabricator.com/project/view/1408/
2. Click manage - see https://secure.phabricator.com/project/manage/1408/
3. Click "edit details"
4. Select "visible too"
5. [ note that the current policy is "custom policy" ]
6. Select "custom policy"

Yeah, persistent details would be helpful. My best guess at repro steps from the screenshot is:

If you don't mind, can you provide specific steps to reproduce and attach actual photos to the task? External links can change/etc and we'd prefer to have a complete record. Thanks!

I'm going to close this since there's no pathway forward that I can see. We might just not understand your request.

Hrm, I'm going to do some followup with our DBAs as well. The performance numbers above were copied from the DBA analysis; however, when I make the change in production, rather than in staging, some of the behavior does not directly replicate.

I'm going to write a CurtainExtension for it (I just missed it in the original round) but you might have to fix my bad stylin'.

Not sure offhand how to fix this.

I'll leave this open for now since the behavior still feels a bit spooky to me, let me know how things go once you get the key in place? If they're smooth, we can chalk this up to be being bad at MySQL.

I suppose the originally proposed key may be better than I think. I was imagining that it couldn't use the third part of the key (on changeType) but I suppose it actually can. I'm still confused by the UNION ALL behavior, but if the key's working on your dataset I think it's fine to add it -- it takes <1 second to add to our dataset, so even for large datasets it shouldn't be wildly unreasonable, even if we do find a better key later.

I'm seeing a replay of the (current?) rules about this message?

That's fine -- it looks like it's picking up the keys, I'm just not sure why it's then examining 303 million rows when it has the keys available.

to clarify: I removed the key I originally added

KEY `commitID` (`commitID`,`isDirect`,`changeType`)

was I not supposed to do that?

Hrrm, let me see if I can reproduce anything like that, that's totally bizarre to me.

KEY `cidid` (`commitID`,`isDirect`),
KEY `cidct` (`commitID`,`changeType`)

are the keys created

Hmm, weird. Can you show me the EXPLAIN SELECT ... for that?

That variant of the query is slow as well.

It's definitely a little surprising that "Outbox > Delivery" is describing a sort of hypothetical scenario that assumes you were a recipient in the first place.

Maybe we could add a "you weren't a recipient in the first place" reason which has priority over the other reasons, and make the text on the page more clear. Or just remove "Outbox" completely, I'm not sure it's really useful.

That is, "Delivery" is "delivery to you, the viewer". Maybe that's what's unclear?

The "Delivery" tab is specific to the viewer, not the mail. There's no way to see see why mail was or wasn't delivered to other users.

223663 is you adding a comment to Conpherence.
184911 is you moving a task on a workboard.
219192 is you attaching a task to a revision.

If I understand the mail visibility, you should be able to see 229037
(PHID-MTAM-qfsqnqkebrkuaohcacsg), because it's "to" you?

My last comment added 3 messages in my outbox:

I can't actually see them, let me go pull up the console.

I think all of those are not self-mail - they are all have "TO" as other users than myself.

In all of them, I'm the actor, but they specify "Self Mail" as the (a) reason for voiding.
I do have "self mail" disabled, so it might be taking that?

That key seems like it's probably not great for the OR changeType part of the query. If you still have the table handy and playing with it doesn't take an unreasonably huge amount of time, is this better?

getEnableCoverage() returns three values:

to clarify

KEY `commitID` (`commitID`,`isDirect`,`changeType`)

Awesome, thanks! Updated to latest and confirmed.

Currently I don't believe we have any client-side form validation, at least none that I can think of.

This sounds consistent and expected to me. Either we have JavaScript validation of fields, and therefore notify you of errors immediately so they can be resolved before submission or we don't.

To clarify, you're expecting the error message to disappear immediately, before you submit the form?

Thanks for the report! Resolved at HEAD.

D15454 changes the wording to:

We should probably change the text on "placed up for grabs" to "unassigned from X". I don't think we use that term anywhere else nowadays, either.

Yeah, cut-off would also be acceptable (although I like wrapping slightly better). It might motivate users to pick shorter names but there's not even a lot of space for reasonable names when combined with milestones/subprojects like:

Tags over 320px are likely cut off in all mobile devices anywhere you use them, as well. There aren't too many options if you use really really long names. GitHub cuts it off in like, 20 characters.

We might not use tags over there, not clear yet.

This is particularly problematic in the UI when you need to revert the first change to a task since creation - there doesn't appear to be any way to find the initial assignee of https://phabricator.wikimedia.org/T129128#2105237 for example