• Audit is missing the "attach, drop-voided, upgrade-to-draft-text" code that Differential has when submitting "isEditing" comments.
• Audit is missing the entire submit-editing workflow in general.

• When you load the page and there are multiple inlines which we're pushing into an edit state, we focus and select them in sequence (and the overall order is basically random). The focus + select behavior should be disabled for this edit pathway.

• When we force saves of inlines, we probably (?) save the last saved state, not the draft state.

• When you click a line, type text, get a draft, reload the page, and cancel the inline: we cancel into an empty state.

• When you click a line, type text, then go to the list view: the revision draft state has not re-synchronized based on an empty inline comment with only a draft, because saving a draft doesn't trigger a re-sync.

• In the preview, editing inline comments don't reflect draft states correctly.

• The preview section underneath the comment area doesn't mark "editing" inlines in any special way.
• The "view" links in the preview don't jump to editing comments correctly.
• The "delete" links in the preview don't put editing comments into the right state (delete is normally impossible for an editing comment).

As of D21188, inlines have these remaining weird behaviors:

In the DOM, inline comments are currently structured as a "comment row", followed by zero or more related rows (which may contain the editor UI or the undo UI).

OKAY SO

• Collapsing paths has moved forward to T13519.
• Table of Contents updates have moved forward to T13520.
• FormationView issues have moved forward to T13521.

I believe D21167 fixed this.

The request sequence here is:

I'm not aware of any current customer interest in improving this behavior, and it's likely a lot of work to improve what is largely a moot case (there's usually no value in reviewing individual file-level changes in a "create a new branch X" commit) that only affects Subversion.

Pretty much ready to ship:

I get the sense that most of this discussion is around diffs, but we've been having issues with the default size of 8 showing in code blocks in Phriction causing more wrapping than we would like.

"FutureGraph" is dead. Long live "HardpointEngine".

Generators can't "return" until PHP 7:

In T11968#251717, @epriestley wrote:

XHPAST currently can't build an AST for $result = yield ..., even though this is a valid construct. This is probably a straightforward fix.

XHPAST currently can't build an AST for $result = yield ..., even though this is a valid construct. This is probably a straightforward fix.

Does using --max-pack-size to reduce the maximum packfile size really let Git "checkpoint" after each packfile, so the process is effectively resumable?

PHI1655 identifies a specific case where enormous packfiles may create problems:

I have some code which runs and looks plausible (i.e., not covered in piles of callback garbage), at least:

Here's an actual example of loadHardpoints($objects, $hardpoint):

In the specific case of the Hardpoints, we currently often have code which loads objects but doesn't do anything with them. For example, most Query classes use didFilterResults() to fill things-that-sure-look-like-hardpoints, but few do anything with the results.