Write Depends on D123 in the commit message.

If we are to support git push HEAD:review, I'd like users who are interested in this to explain exactly what they expect it to do, and why that's dramatically better than copy-and-paste.

Yep, it was the Apache restart. Sorry for poor testing, thanks for fixing the incidentally discovered bug, and hopefully the other part of my comment was less useless :)

master as of yesterday (Tue) approx. 2 pm PST. When I get home I'll check if the logs match up, but I took care to pull all three repos and ran a storage upgrade.

Not opening another bug report since Phame is not out of beta yet, but I thought I should still let you know that I stumbled upon an exception yesterday after updating from before the comments revamp:

• When viewing a Blog page (not a post itself): (Unable to resolve method 'isArchived'.)
• When submitting a New Blog form: >>> UNRECOVERABLE FATAL ERROR <<< Undefined class constant &#039;TYPE_STATUS&#039;

Works for me!

For reference, I think this is an excellent plan and I'm stoked to see internal comments being considered. (If I already have to get my users to sign up for a Phabricator account, it's good if they don't have to also log into Disqus.)

Maybe there can be a way to enable it as a menu item, but being disabled by default. Your call! Thanks for getting back to me despite the beta label (forgot that you don't take bug reports / features on it yet).

In our community, for regular members (not involved in board/admin tasks), I expect it to be the one most commonly used action (the other common action being commenting on other people's posts). Either way, I can work with providing a separate link in a dashboard box. However, for a global menu I would find it confusing to not find "Create" actions in the "Create New" menu.

@epriestley, I agree that in many situations you might not want to auto-subscribe to sub-pages. However, I'm not certain that having "Subscribers" appear on every sub-page is a bad thing - it allows customization if the inherited default is inappropriate for a given page.

@chad, I thought about that too, but this rule only comes into play after the page has been created. Since the view/edit permissions are inherited from the parent page, which has "Subscribers" access in this case, the form complains because without filled-in subscriber list, I "would no longer be able to view/edit the object". The form doesn't know about the Herald rule that makes this possible afterwards.

Possibly related, Edit Notes also don't show up in the Feed. At least for wiki page changes, they would make a lot of sense there. Please ignore this comment if it's not the same (or a similar) underlying technical issue.

Potentially interesting (for integration or inspiration): http://blog.developer.bazaarvoice.com/2015/05/01/automating-a-git-rebase-workflow/

Okay, so... I updated and that didn't help, but what did help was the realization that diffs are only included in (Herald rule) "Commit Hook: Commit Content" emails, never in "Commits" (i.e. push) emails even with the inlined-patches setting enabled. That means I'll have to send out multiple emails for pushes with multiple commits, but at least it works now. Thanks for helping out!

Restarting Phabricator hasn't made a difference. Will play around with it more tonight to see where things are still going wrong. Thanks for all your input, it's much appreciated :)

Hm, I just pushed a couple of small changes to master and the commit email still only showed the summary. I had metamta.diffusion.attach-patches set to true and and metameta.diffusion.inline-patches set to 5000 (which should be more than enough for the few lines that were changed). The ones for metamta.differential were also set accordingly, although I guess that shouldn't matter as it didn't go through a code review in this case.

Oh wait, I set it for Differential but not Diffusion. Let me verify that it works and I'll close the bug after. Thanks!

Good point, I actually seem to have set that setting to 1000 before. However, none of our commit emails ever seems to include patches even below that number of lines. Maybe this has been fixed since I last updated our Phabricator instance?