[BUG] PGP key with multiple identities fails to import

[msfjarvis]

Describe the bug

A GPG key with more than one identity fails to import with the PGPainless backend

Steps to reproduce

1. Get test keys from https://msfjarvis.dev/apsg/pass-test/406de9b7f0fc
2. Attempt to import private key with PGPainless backend

Expected behavior

Key imports successfully

Screenshots

N/A

Device information

• Device: Pixel 4a
• OS: Android 12
• App version: 3939003

Additional context

No response

[vanitasvitae]

Let me know if I can help you out with anything :)

[msfjarvis]

Let me know if I can help you out with anything :)

PGPainless seems to be doing its job correctly, the failure is in the formatting code here which appears to overflow a Long. Still investigating why that happens.

[moppman]

Continued from here:
I deleted the second identity from the key, but I'm still getting the same error on import, which was surprising to me at first, but then I read that removed identities aren't actually "deleted" but only revoked via special signature.
So the overflow bug would still make sense here. 😵‍💫

[vanitasvitae]

I'm still a bit confused... Why does an additional user-id have an impact on key-id calculation? 🤔

Edit: Looking at the code, why are you hex-encoding the key-id here, and then immediately decoding it again to calculate the key-id? Shouldn't just passing keyRing.publicKey.keyID to the KeyId constructor be sufficient?

Still, there might be a bug in the encode/decode function?

I'm not fluent in Kotlin, so maybe I'm missing something?

[msfjarvis]

Edit: Looking at the code, why are you hex-encoding the key-id here, and then immediately decoding it again to calculate the key-id? Shouldn't just passing keyRing.publicKey.keyID to the KeyId constructor be sufficient?

goddamnit 🤦🏼

I wish I could tell what was going on in my head when I wrote that. 1956b66 fixes the failing test for me, once #1741 lands and I get snapshots rolling again this should be fixed.