1249424 - remove nsAutoArrayPtr usage from dom/media/gmp/

Status: RESOLVED FIXED

(Core :: Audio/Video: GMP, defect)

Description

[Nathan Froyd [:froydnj]]

The lone use of nsAutoArrayPtr in dom/media/gmp/ comes from our scoped_array polyfill.  We want to remove nsAutoArrayPtr, though, so we need to come up with a plan for fixing this.  The two options I see are:

1) use UniquePtr instead;
2) stick the definition of nsAutoArrayPtr in that file, rename it to something that's clear you shouldn't actually be using it, and forget about it.

I obviously prefer doing #1, but I'm not clear on how much API surface we really need from nsAutoArrayPtr here.  Looking at the scoped_array use:

https://dxr.mozilla.org/mozilla-central/source/dom/media/gmp/rlz/win/lib/machine_id_win.cc#44

it looks like we use a very limited subset.  Can we implement just enough API surface to satisfy that use case, or do we need to implement something more complete to accommodate partner usage of this code?

ni? to jesup, since cpearce is on PTO.  Feel free to redirect.

Comment 1

[Randell Jesup [:jesup] (needinfo me)]

Need to redirect this to cpearce; that's deep in the EME side

Comment 2

[Chris Pearce [:cpearce (Not reading bugmail)]]

I'm not super-keen on using UniquePtr since we have to pass the raw pointer to OS functions, which breaks the "uniqueness" of the pointer.

We can replace the use of scoped_array<wchar_t> with a std::vector<wchar_t> in that file, since we're unlikely to be needing to pull in updates from upstream again.

Comment 3

[Chris Pearce [:cpearce (Not reading bugmail)]]

Attachment: MozReview Request: Bug 1249424 - Remove use of nsAutoArrayPtr from librlz. r?gerald

Review commit: https://reviewboard.mozilla.org/r/36175/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/36175/

Comment 4

[Gerald Squelart (he/him) (not at Mozilla since 2022-09-15)]

Comment on attachment 8722673 [details]
MozReview Request: Bug 1249424 - Remove use of nsAutoArrayPtr from librlz. r?gerald

https://reviewboard.mozilla.org/r/36175/#review32761

Comment 5

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/1ce56afd66f4

Comment 6

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/1ce56afd66f4