1609815 - Remove Web Replay

Status: RESOLVED FIXED

(Core Graveyard :: Web Replay, task)

Description

[Brian Hackett [Laid off!]]

I was laid off yesterday, and web replay seems like it should be removed from the tree now. I will not be around/able to maintain it, so unless someone else wants to make the effort to maintain it, pretty soon it will break and stay broken.

The attached patch removes all the replay related code I could find, and applies onto 893e9fda4bf3. I can't break this up and submit it for review because my phabricator account has been disabled, so someone else will need to get this landed.

Comment 1

[Brian Hackett [Laid off!]]

For anyone wondering, I will be forking Gecko and continuing to develop web replay in a new tree. I've started https://webreplay.org/ for this effort.

Comment 2

[Brian Hackett [Laid off!]]

Attachment: patch (893e9fda4bf3)

Comment 3

[Jason Laster [:jlast]]

I can try landing this.

I'm sorry to see you go Brian, but happy that users will have an opportunity to experience Web Replay in the near future.

Comment 4

[Brian Hackett [Laid off!]]

Attachment: patch (893e9fda4bf3)

Comment 5

[Jason Laster [:jlast]]

Attachment: Bug 1609815 - Remove Web Replay C++ implementation.

Comment 6

[Andrew McCreight [:mccr8]]

I'm sorry to hear that you are leaving, Brian. Good luck with your continuing work on Web Replay. It is really neat.

(In case you don't know, you can set Bugzilla to block needinfos of you.)

Comment 7

[:Harald Kirschner :digitarald]

Thanks for all your work on this Brian. Watching you bring this from a crazy tech idea to this blazing-fast user-ready experience has been an amazing journey and I am honored that I could work with you on it.

It's Web Replay's nature to never go away but to evolve, so thanks for pushing on with it and best of luck!

Comment 8

[Kelsey Gilbert [:jgilbert]]

r+ for my parts, but I'd like to see some management sign-off on ripping this out.

Comment 9

[Andrew McCreight [:mccr8]]

(In reply to Jeff Gilbert [:jgilbert] from comment #8)

r+ for my parts, but I'd like to see some management sign-off on ripping this out.

This sounds like a reasonable request. I'll see what Jason thinks about it.

Comment 10

[Jason Laster [:jlast]]

I think that is reasonable.

Comment 11

[Jan Honza Odvarko [:Honza] (always need-info? me)]

It's sad to see this removal but, I think it's the right way to go so, OK for me.
Andrew can you please confirm?

Honza

Comment 12

[Jason Laster [:jlast]]

Attachment: Bug 1609815 - Remove Web Replay UI. r=loganfsmyth

Comment 13

[Andrew Overholt [:overholt]]

Confirmed (and confirmed with a few others), thank you.

Comment 14

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/c41e5432b3c5
Remove Web Replay UI. r=loganfsmyth

Comment 15

[Andreea Pavel [:apavel]]

Backed out for build bustages at JSControl.cpp

Push with failures: https://treeherder.mozilla.org/#/jobs?repo=autoland&selectedJob=286213282&resultStatus=testfailed%2Cbusted%2Cexception&revision=c41e5432b3c509f98e79af9e0e234e43ce02d02d

Failure log: https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=286213282&repo=autoland&lineNumber=41011

Backout: https://hg.mozilla.org/integration/autoland/rev/1f891f357702b1d7016a61c8f4d84ff18cd46a6a

Comment 17

[Mark Banner (:standard8)]

Jason: Please note that bug 1252803 just landed, so you'll have a little bitrot on the .eslintignore file to update as well.

Comment 18

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/cfe2388a674e
Remove Web Replay UI. r=loganfsmyth

Comment 19

[Noemi Erli[:noemi_erli]]

Backed out changeset cfe2388a674e (Bug 1609815) for causing TypeError in browserContext CLOSED TREE

Push with failures: https://treeherder.mozilla.org/#/jobs?repo=autoland&selectedJob=288273391&resultStatus=pending%2Crunning%2Csuccess%2Ctestfailed%2Cbusted%2Cexception&searchStr=73d04a55ba6c705a6b4573608ed129afd31b7171&revision=cfe2388a674e024d1920245e867f9982087c8164

Failure log: https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=288267964&repo=autoland&lineNumber=13787

Backout: https://hg.mozilla.org/integration/autoland/rev/8e920c4afb281ad904a843d3fb3482fae273575c

Comment 20

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/e8342190e430
Remove Web Replay UI. r=loganfsmyth

Comment 21

[Bogdan Tara[:bogdan_tara | bogdant]]

https://hg.mozilla.org/mozilla-central/rev/e8342190e430

Comment 22

[Mark Banner (:standard8)]

Attachment: Bug 1609815 - Remove unused devtools/client/webreplay reference from .eslintignore. r?jlast

Comment 23

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/487a1fa09d0c
Remove unused devtools/client/webreplay reference from .eslintignore. r=mccr8

Comment 24

[Razvan Maries]

https://hg.mozilla.org/mozilla-central/rev/487a1fa09d0c

Comment 25

[Alexandru Ionescu (needinfo me) [:alexandrui]]

== Change summary for alert #24935 (as of Fri, 14 Feb 2020 04:51:24 GMT) ==

Improvements:

0.21% installer size osx-shippable opt nightly 80,459,534.92 -> 80,294,078.50

For up to date results, see: https://treeherder.mozilla.org/perf.html#/alerts?id=24935

Comment 26

[Alexandru Ionescu (needinfo me) [:alexandrui]]

(In reply to Pulsebot from comment #23)

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/487a1fa09d0c
Remove unused devtools/client/webreplay reference from .eslintignore. r=mccr8

Hi, is the below regression a won't fix?
== Change summary for alert #24917 (as of Wed, 12 Feb 2020 17:48:12 GMT) ==

Regressions:

0.27% installer size osx-shippable opt nightly 80,179,302.42 -> 80,393,913.33
0.21% installer size osx-shippable opt instrumented 113,427,749.58 -> 113,665,917.00

For up to date results, see: https://treeherder.mozilla.org/perf.html#/alerts?id=24917

Comment 27

[Mark Banner (:standard8)]

(In reply to Alexandru Ionescu :alexandrui (needinfo me) from comment #26)

(In reply to Pulsebot from comment #23)

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/487a1fa09d0c
Remove unused devtools/client/webreplay reference from .eslintignore. r=mccr8

Hi, is the below regression a won't fix?
== Change summary for alert #24917 (as of Wed, 12 Feb 2020 17:48:12 GMT) ==
...
For up to date results, see: https://treeherder.mozilla.org/perf.html#/alerts?id=24917

I think you have the wrong bug. The pushlog for that alert points to bug 1586915. The change I pushed only affected a configuration file, so there's no way that could have affected the build size.

Comment 28

[Alexandru Ionescu (needinfo me) [:alexandrui]]

You're right, sorry.

Comment 29

[Andreea Pavel [:apavel]]

Does this still need to have leave-open?

Comment 30

[Andrew McCreight [:mccr8]]

(In reply to Andreea Pavel [:apavel] from comment #29)

Does this still need to have leave-open?

There's still a lot of Web Replay to be removed. The patch that landed only removed the JS part of it.

Comment 31

[Andreea Pavel [:apavel]]

(In reply to Andrew McCreight [:mccr8] from comment #30)

(In reply to Andreea Pavel [:apavel] from comment #29)

Does this still need to have leave-open?

There's still a lot of Web Replay to be removed. The patch that landed only removed the JS part of it.

Oh okay, thanks.

Comment 32

[Haik Aftandilian [:haik]]

(In reply to Andrew McCreight [:mccr8] from comment #30)

(In reply to Andreea Pavel [:apavel] from comment #29)

Does this still need to have leave-open?

There's still a lot of Web Replay to be removed. The patch that landed only removed the JS part of it.

Is it OK to land something that breaks WebReplay or removes some WebReplay calls at this time?

Comment 33

[Jason Laster [:jlast]]

Yes

Comment 34

[Andrew McCreight [:mccr8]]

I've rebased the remainder of the patch.

Comment 35

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/d407a28318e6
Remove Web Replay C++ implementation. r=jgilbert,jandem,gbrown

Comment 36

[Andrei Ciure[:aciure]]

Backed out changeset d407a28318e6 (bug 1609815) for causing windows ming bustages
https://hg.mozilla.org/integration/autoland/rev/6ea49052598fbebed1ed389269ab58ad91b53d2f

push that caused the backout: https://treeherder.mozilla.org/#/jobs?repo=autoland&selectedJob=290711643&revision=d407a28318e68dd78af8365a1ea3f8f7384f92c5&failure_classification_id=2

Comment 37

[Mike Hommey [:glandium]]

netwerk/base/nsSocketTransportService2.h is missing a #include <mozilla/Maybe.h>. The Maybe was introduced in bug 1596799.

Comment 38

[Andrew McCreight [:mccr8]]

(In reply to Mike Hommey [:glandium] from comment #37)

netwerk/base/nsSocketTransportService2.h is missing a #include <mozilla/Maybe.h>. The Maybe was introduced in bug 1596799.

Thanks, Mike. Yeah, a few places were bootlegging Maybe<> from WebReplay headers. I'll see if a Windows build finishes, then try to land it again.

Comment 39

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/954d5a7490a3
Remove Web Replay C++ implementation. r=jgilbert,jandem,gbrown

Comment 40

[Arthur Iakab [arthur_iakab]]

https://hg.mozilla.org/mozilla-central/rev/954d5a7490a3