1461830 - (libdweb-dns-sd) [experimental api] dns-sd APIs for IoT and p2p extensions

Status: REOPENED

(WebExtensions :: Experiments, enhancement, P5)

Description

[Dietrich Ayala (:dietrich)]

Create experimental WebExtension APIs for mDNS for p2p and IoT experimentation purposes. The WoT project and also our dweb explorations will utilize these.

This is distinct from bug 1342724, which is about implementing Chrome's apps API for mDNS.

Comment 1

[Caitlin Neiman [:caitmuenster]]

Hi Dietrich, this has been added to the agenda for the WebExtensions APIs triage on May 29, 2018. Would you be able to join us? 

Here’s a quick overview of what to expect at the triage: 

* We normally spend 5 minutes per bug
* The more information in the bug, the better
* The goal of the triage is to give a general thumbs up or thumbs down on a proposal; we won't be going deep into implementation details

Relevant Links: 

* Wiki for the meeting: https://wiki.mozilla.org/WebExtensions/Triage#Next_Meeting
* Meeting agenda: https://docs.google.com/document/d/1Y_oYPldTT_kQOOouyJbC-8y3ASIizScLKFRhQfsDQWI/edit#heading=h.v63i1wz9d1pc
* Vision doc for WebExtensions: https://wiki.mozilla.org/WebExtensions/Vision

Comment 2

[Caitlin Neiman [:caitmuenster]]

Hey dveditz, could you help us out with a security review for this proposal? 

If it looks like the code will be maintained and dveditz is ok with the proposal, we will approve the design.

Comment 3

[Shane Caraveo (:mixedpuppy)]

It looks like mdns has not been touched since 2016, and was landed and maintained solely for the flyweb project.  IMO this places the api into unmaintained mode.  

Either there are some platform plans (jduell? any idea) or there are product plans (mconca? could you find out) that could indicate the mdns support has legs, or we'd have to potentially take on maintenance to use it as an api.

Comment 4

[Shane Caraveo (:mixedpuppy)]

Another note on this:

I just re-read comment 0 and it mentions creating an experimental api for this.  There is no reason that cannot be done.  Dietrich, can you shed some light?  If experimental APIs are fine enough for now, no design-decision is necessary.

Comment 5

[Dietrich Ayala (:dietrich)]

I wish I'd gotten a ni? for the meeting. I definitely would've joined if I'd seen that this conversation was happening!

Shane, your last comment is correcto:

* Irakli and and I are going to be prototyping out a set of experimental WebExtension APIs for a set of primitives that enable a set of p2p/dweb/IoT use-cases.

* The purpose of these experiments are not to put them on a track to being adopted as WebExtension APIs. We're not ruling that out at some point in the future, but it's not a goal at this time.

I'm filing the bug so that the request is officially logged, and so that we can point people asking about it at these requests.

Feel free to close all these bugs if they're interfering with your product development process - we can point people at the closed bugs easily enough.

I'm canceling the ni?s since they're not really needed at this point.

Comment 6

[David Durst [:ddurst]]

Marking this as P5 incomplete until we have clear directive.

Comment 7

[Irakli Gozalishvili [:irakli] [:gozala] [@gozala]]

As per https://github.com/mozilla/libdweb/issues/98

Looks like following change mozilla/gecko-dev@b2f8565 removed nsMulticastDNSModule.cpp and there for '@mozilla.org/toolkit/components/mdnsresponder/dns-sd;1' component is no longer available.

---

Chatted with :kmag about corresponding changes and it appears that it was not intentional change. From what I can tell change was converting all component registrations to be static. However since JS implementatios is used on non mac:

https://searchfox.org/mozilla-central/source/netwerk/dns/mdns/libmdns/components.conf#20

That must have lead to missed native implementation registration. And there seems to be no test that would have caught the regression.

Comment 8

[BugBot [:suhaib / :marco/ :calixte]]

The bug assignee is inactive on Bugzilla, so the assignee is being reset.