1182357 - Add API to allow explicit creation of nsExpandedPrincipals from JS

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Bill McCloskey [inactive unless it's an emergency] (:billm)]

+++ This bug was initially created as a clone of Bug #1180921 +++
This is a follow-up:

> One thing I think we're missing is a way to create a sandbox with expanded
> principals that uses OriginAttributes. We could just copy the addonId
> argument from the Sandbox constructor to the OriginAttributes of the
> expanded principals. Or we could make a constructor for expanded principals.

Yes, we need this. Please file a followup bug and needinfo me and I'll do it.

Comment 1

[Gabor Krizsanits (INACTIVE)]

(In reply to Bill McCloskey (:billm) from comment #0)
> > One thing I think we're missing is a way to create a sandbox with expanded
> > principals that uses OriginAttributes. We could just copy the addonId

I think there are two cases where we need OriginAttributes, on parent/chrome side with a regular principal (some resource URI or later maybe an addon protocol) and on child/content side with an expanded principal. Can/should we throw for anything else? I'm particularly worried that someone creates a codebase principal in the content process with OriginAttributes. Since we ignore OriginAttribute in the subsumes code that could easily lead to a leak of privileges.

Comment 2

[Bobby Holley (:bholley)]

(In reply to Gabor Krizsanits [:krizsa :gabor] from comment #1)
> Since we ignore OriginAttribute in the subsumes code

We don't ignore it at all: https://hg.mozilla.org/mozilla-central/annotate/e7e69cc8c07b/caps/nsPrincipal.cpp#l169

That's the whole point - each origin attributes makes an orthogonal dimension of origin-ness.

Comment 3

[Gabor Krizsanits (INACTIVE)]

(In reply to Bobby Holley (:bholley) from comment #2)
> (In reply to Gabor Krizsanits [:krizsa :gabor] from comment #1)
> https://hg.mozilla.org/mozilla-central/annotate/e7e69cc8c07b/caps/
> nsPrincipal.cpp#l169

Bah, I was looking for this exact line but somehow managed to overlook it... I show myself out.

Comment 4

[Bobby Holley (:bholley)]

Attachment: Part 1 - Implement support for optional size_is for arrays passed from JS. v1

The fact that the caller needs to pass this is just an artifact of the clunky
XPIDL type system. This should let us make nicer APIs.

Comment 5

[Bobby Holley (:bholley)]

Attachment: Part 2 - Add an API to mint nsExpandedPrincipals. v1

Comment 6

[Bobby Holley (:bholley)]

https://treeherder.mozilla.org/#/jobs?repo=try&revision=6901a28b38b4

Comment 7

[Bobby Holley (:bholley)]

https://treeherder.mozilla.org/#/jobs?repo=try&revision=d86fa5e4653c

Comment 8

[Bobby Holley (:bholley)]

(In reply to Bobby Holley (:bholley) from comment #7)
> https://treeherder.mozilla.org/#/jobs?repo=try&revision=d86fa5e4653c

Looks green. re-running with ASAN just in case.

https://treeherder.mozilla.org/#/jobs?repo=try&revision=928a699ce497

Comment 9

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/86d4a584905c
https://hg.mozilla.org/integration/mozilla-inbound/rev/71069116ee28

Comment 10

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/86d4a584905c
https://hg.mozilla.org/mozilla-central/rev/71069116ee28