1557645 - Basic support for OOP-iframes in print preview

Status: RESOLVED FIXED

(Core :: Printing: Setup, defect, P2)

Description

[Jonathan Watt [:jwatt]]

A logical milestone on the way to getting printing fully working with out-of-process iframes is to get print preview working with OOP-iframes. Once that is working, getting printing fully working will mostly "just" be about interleaving and positioning of draw recordings from the various processes.

Comment 1

[Chris Peterson [:cpeterson]]

M6 because printing support blocks enabling Fission in Nightly. Neha says print preview doesn't need to block dogfooding (M5).

Comment 2

[Jonathan Watt [:jwatt]]

This will likely involve replacing the current setup code with something based on ManagedEndpoint.

Comment 3

[Chris Peterson [:cpeterson]]

Tracking for Fission Nightly (M6c)

Comment 4

[Neha Kochar [:neha]]

Sean said fission printing work will most probably not be ready in time for nightly experiment and we shouldn't block on it. It is expected to be fixed in a couple of weeks following the experiment launch.

Comment 5

[Emilio Cobos Álvarez (:emilio)]

Attachment: Bug 1557645 - Clone OOP iframes in the right process. r=nika,mattwoodrow

This seems to work (as in, I correctly process-switch the iframe and hit
the printf in the right process). I didn't hook up the cloning just yet,
but it should be relatively doable.

Comment 6

[Emilio Cobos Álvarez (:emilio)]

Comment on attachment 9175870 [details]
Bug 1557645 - Clone OOP iframes in the right process. r=nika,mattwoodrow

Matt, Nika, I think you're familiar with the process-switching logic that I'm touching here. Does this look reasonable?

For actual printing, we may want to provide some clone completion callback or something of that sort, perhaps, but for print preview I think hooking up the cloning is the last bit missing here.

Comment 7

[Emilio Cobos Álvarez (:emilio)]

Attachment: Bug 1557645 - Document::CreateStaticClone shouldn't be virtual. r=jwatt

Just minor preliminar cleanup.

Comment 8

[Emilio Cobos Álvarez (:emilio)]

Attachment: Bug 1557645 - Unify in-process document cloning a bit more. r=jwatt

By passing the content viewer to CreateStaticClone we can do this fairly
easily. I'm going to add another entry-point to the clone process and
it'd be good if we avoid repeating the SetDocument dance and so on.

Depends on D90314

Comment 9

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/142bea7765f9
Document::CreateStaticClone shouldn't be virtual. r=jwatt

Comment 10

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/0d33b3a1a7db
Unify in-process document cloning a bit more. r=jwatt

Comment 11

[Alexandru Michis [:malexandru]]

https://hg.mozilla.org/mozilla-central/rev/142bea7765f9
https://hg.mozilla.org/mozilla-central/rev/0d33b3a1a7db

Comment 12

[Nika Layzell [:nika] (ni? for response)]

Comment on attachment 9175870 [details]
Bug 1557645 - Clone OOP iframes in the right process. r=nika,mattwoodrow

replied on the review

Comment 13

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/713a1bb30a6d
Clone OOP iframes in the right process. r=nika

Comment 14

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/df9897d94dad
Fix gcc warning.

Comment 15

[Atila Butkovits]

https://hg.mozilla.org/mozilla-central/rev/713a1bb30a6d
https://hg.mozilla.org/mozilla-central/rev/df9897d94dad