-
Notifications
You must be signed in to change notification settings - Fork 675
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
ics27 controller API review #753
Comments
My vote goes for |
+1 from me, they would be my preference |
+1 on
|
Food for thought: |
+1 on I will ponder on the I am just hesitant to hide all the magic we are doing with channel creation behind the |
This is interesting. |
GH Action includes: - check for lint errors - check for backward compatibility breaking changes Migrated buf config files from v1beta1 to v1 using command: 'buf config migrate-v1beta1' (https://docs.buf.build/configuration/v1beta1-migration-guide)
Summary
As we move towards a
beta-1
release of interchain accounts I think we should take a moment to review the API exposed to authentication module developers via the controller submodule. When we finalise a release these APIs become harder to change, if not impossible to change without causing disruptions. Raising this issue to open a discussion on this and make a proposal.Proposal
Rename both functions exposed to authentication module developers.
InitInterchainAccount
->RegisterInterchainAccount
Originally I believe this was named
InitInterchainAccount
as it is initialising a channel handshake by creating and routing a newMsgChannelOpenInit
via the sdk's msg service router. I propose we rename this toRegisterInterchainAccount
as essentially its implicit that we are using ibc and channel creation should be opaque from the user's perspective. This function is part of thecontroller
submodule and as such, there is no naming collision withRegisterInterchainAccount
inhost
, in my opinion it aligns nicely.TrySendTx
->SendTx
/ExecuteTx
We should remove
Try
from the function name here. I think this implies that something could often go wrong on packet execution, when in reality we want all valid packets to succeed.Try
is a word seen commonly with error handling syntax and libraries and with that in mind I think it makes a clearer API to just remove it from the function definition.I propose either
SendTx
orExecuteTx
. The latter aligns with the enum naming ofType
used inInterchainAccountPacketData
, but I don't feel particularly strongly in favour of either.Thoughts on this?
For Admin Use
The text was updated successfully, but these errors were encountered: