Registration fixes for 2023

[rubdos]

The registration procedure has changed a few months ago, and the old API has been terminated. This is an attempt to expose the new registration API surface for Whisperfish. All the other clients I know are secondary-only, so this shouldn't affect anyone but Whisperfish.

[rubdos]

FWIW, Whisperfish diff: https://gitlab.com/whisperfish/whisperfish/-/merge_requests/481

[MarcusGrass]

Don't know if this is related to this specific PR, but it seems that https://github.com/whisperfish/libsignal-service-rs/blob/main/libsignal-service/src/provisioning/manager.rs#L173 is now a raw string Ex Got provisioning code: 'ac9e637e-16d1-50bd-b0aa-daf33b346cb0.1691236722275:7U2fte60zkioI3KLaIEZQT2skhYgFXLjIL2hgQqqLAQ=' Interestingly that worked yesterday

[rubdos]

That would indeed be related, thanks for reporting. I have not been able to test this yet, because I don't have access to my second SIM.

[MarcusGrass]

That would indeed be related, thanks for reporting. I have not been able to test this yet, because I don't have access to my second SIM.

Yeah it's not easy with only one number, threw up my local fix as a separate PR here #240, it's nice in that it's just skipping some validation and shouldn't mess with any compatibility if someone were to unexpectedly have some old signal-compatible server they were testing against.

[rubdos]

Thanks! I've cherry-picked your commit into this PR. Were you able to confirm a device with the other patches in here?

[MarcusGrass]

Thanks! I've cherry-picked your commit into this PR. Were you able to confirm a device with the other patches in here?

Nice, I can't test the registration flow at the moment because I also need another SIM, also, I'm using https://github.com/whisperfish/presage/ which breaks from the api-change here, would need to figure out how to rewire it first, do you know where the current sources for the java sdk is? All I can find are archived repos.

But the link-device flow works after that change on this branch!

I am getting this in the logs though:

[2023-08-05T13:48:47Z ERROR libsignal_service::websocket] SignalWebSocket: Websocket error: end of web request stream; socket closing
[2023-08-05T13:48:51Z ERROR libsignal_service::websocket] SignalWebSocket: Websocket error: SignalWebSocket: end of application request stream; socket closing

It seems that the socket is closed from Signal's side https://github.com/whisperfish/libsignal-service-rs/blob/main/libsignal-service/src/websocket.rs#L231, then the stream starts returning None, although I don't think that's necessarily unexpected in this case since the connection is created and dropped for this request.

[rubdos]

do you know where the current sources for the java sdk is? All I can find are archived repos.

Everything is in Signal-Android these days. @gferon usually patches presage within a few days after I push things in here.

It seems that the socket is closed from Signal's side https://github.com/whisperfish/libsignal-service-rs/blob/main/libsignal-service/src/websocket.rs#L231, then the stream starts returning None, although I don't think that's necessarily unexpected in this case since the connection is created and dropped for this request.

I think there's an issue about this. Because it doesn't necessarily harm the procedure, we didn't really bother at that point...

[MarcusGrass]

do you know where the current sources for the java sdk is? All I can find are archived repos.

Everything is in Signal-Android these days. @gferon usually patches presage within a few days after I push things in here.

Ah nice, thanks!

It seems that the socket is closed from Signal's side https://github.com/whisperfish/libsignal-service-rs/blob/main/libsignal-service/src/websocket.rs#L231, then the stream starts returning None, although I don't think that's necessarily unexpected in this case since the connection is created and dropped for this request.

I think there's an issue about this. Because it doesn't necessarily harm the procedure, we didn't really bother at that point...

Makes sense, it doesn't really seem problematic

[gferon]

That would indeed be related, thanks for reporting. I have not been able to test this yet, because I don't have access to my second SIM

FWIW, did you think of trying with your main SIM on staging servers? You could re-register infinitely (don't quote me on that) on the test servers.

[rubdos]

Mea culpa, the confirm code was not part of the registration procedure at all. Any way useful to get it in here, it's part of registration anyway.

[rubdos]

@direc85 confirms this works for Whisperfish.

[rubdos]

The registration session thingy would be really cool in a nice RegistrationSession state machine structure, but I'm a bit too lazy to actually implement that now... Updating the examples.

[rubdos]

@gferon this should be it. If you can throw some secondary eyes on it, feel free to hit merge!

[codecov]

Codecov Report

Merging #238 (43b2d4f) into main (6d2649d) will decrease coverage by 0.34%.
Report is 2 commits behind head on main.
The diff coverage is 0.00%.

@@           Coverage Diff            @@
##            main    #238      +/-   ##
========================================
- Coverage   4.16%   3.82%   -0.34%     
========================================
  Files         36      36              
  Lines       2523    2746     +223     
========================================
  Hits         105     105              
- Misses      2418    2641     +223     

Files Changed	Coverage Δ
libsignal-service-actix/src/push_service.rs	9.83% <0.00%> (-2.21%)	⬇️
libsignal-service-hyper/src/push_service.rs	16.31% <0.00%> (-2.55%)	⬇️
libsignal-service/src/provisioning/manager.rs	0.00% <ø> (ø)
libsignal-service/src/provisioning/mod.rs	0.00% <ø> (ø)
libsignal-service/src/push_service.rs	0.00% <0.00%> (ø)

[gferon]

Java :')

[gferon]

LGTM! I'll adapt presage immediately and check that it works as expected (although I read on the chat that somebody tried already.)