feat(cardano): Message signing

[jaskp]

Adds support for arbitrary message signing as per CIP-8.

https://cips.cardano.org/cip/CIP-0008

[jaskp]

Made some changes after feedback from software wallet devs

[jaskp]

Heads-up: There are conflicts with main so I'm about to rebase

[jaskp]

Heads-up: Rebasing again to fix conflict

[matejcik]

you are adding new texts, so you will now need to convert to translations and update.

1. rebase on current main
2. find the newly added strings, convert them to TR.cardano__... symbols
3. add english texts for those symbols to en.json
4. run python core/translations/order.py
5. rebuild the firmware
6. regenerate fixtures
7. when that is ready, force-push here and add translations label
8. wait for CI results

[davidmisiak]

I've rebased on top of the current main. (neither update_fixtures.py ci nor update_fixtures.py ci --github does anything at this moment)

[davidmisiak]

Hi @matejcik, could we finish also this please?

[davidmisiak]

@matejcik any update please?

[prusnak]

Spamming GitHub with comments that add no value is exactly what you do not want to do as a community. It adds lots of noise into all developers' inboxes and we should prefer work on them on code, not reading useless comments.

Locking discussion to collaborators-only.

[gubatron]

eyes on this please, lots of users can't claim midnight drop because they can't sign messages with their trezor ada wallets

[matejcik]

Suggested change

	if not isinstance(response, m.CardanoMessageSignature):
	raise ValueError("Unexpected response")
	response = m.CardanoMessageSignature.ensure_isinstance(response)

[matejcik]

ACK from me, leaving the rest to @obrusvit

whoever ends up doing the final review, please specifically confirm that we're showing the whole payload to the user; there are remnats in the code for "first chunk" display

[ibz]

whoever ends up doing the final review, please specifically confirm that we're showing the whole payload to the user

I am taking this over.

• I can confirm that we are requesting the whole data in a single chunk as per https://github.com/trezor/trezor-firmware/pull/3509/files#diff-ff436e870f332b283aae9a85591a15b35cbbba59bb134c7b515502cd99783d9cR101 - even though the chunkification mechanism stays in place, it is not actually used
• the whole data is passed to the layout function, whose parameter I am renaming from payload_first_chunk to payload as you suggested to avoid future confusion
• HOWEVER if the data is too long it is truncated on display and ... is appended as per https://github.com/trezor/trezor-firmware/pull/3509/files#diff-718c35fab848e1ee6f5e2a387c71b0a0ea8d84039fedce530cb9f0584c164495R381

So your request @matejcik is both fulfilled and unfulfilled at the same time.

Do you think we should raise an error if the data is too long rather than truncating? Or simply not truncate and keep scrolling?

[matejcik]

not truncate, keep scrolling.
we can employ the mechanism for skipping long messages that we came up with for Ethereum; not sure if that was ever implemented or not

[obrusvit]

utACK