Conversation
grod220
force-pushed
the
create-ix-part-1
branch
2 times, most recently
from
4c63763 to
9e39e98
Compare
|
With anza-xyz/pinocchio#349 in progress, was able to remove the shim! 🎊 |
grod220
force-pushed
the
create-ix-part-1
branch
from
1fc9658 to
add440f
Compare
|
Update: going to investigate p-token style fixtures approach to testing both binaries |
| cargo $(nightly) test --features mollusk-svm/fuzz --manifest-path program/Cargo.toml \ | ||
| --test create_always \ | ||
| --test create_idempotent \ | ||
| --test create_shared && \ |
There was a problem hiding this comment.
Found this to be the way we can do conformance testing only on the code paths that have been migrated so far. It also serves a dual purpose as think it's helpful to break down large test files into smaller scoped ones.
grod220
force-pushed
the
create-ix-part-1
branch
from
c0474ab to
1190109
Compare
|
|
||
| harness | ||
| .ctx | ||
| .process_and_validate_instruction(&instruction, &[Check::err(ProgramError::IllegalOwner)]); |
There was a problem hiding this comment.
nit: This error is a bit misleading, since the owner is correct. Maybe is_valid_existing_ata_for_idempotent needs to return specific errors depending of which assert failed. But I do see that the processor is not complete, so maybe is something that you will address later.
There was a problem hiding this comment.
You're right. It's a strange case. From the caller's perspective they submitted a valid wallet/mint/token-program and the ATA address has a token program owned account sitting there (but with invalid data).
This error is inherited from the current program. Wouldn't changing it break conformance?
There was a problem hiding this comment.
Ah, I see, if unpack fails in the original program (which will in this case), it will then reach the ProgramError::IllegalOwner error. You are right, we can't change that. So this is correct, it is replicating the same behaviour.
grod220
force-pushed
the
create-ix-part-1
branch
from
a6ba167 to
2cc4432
Compare
febo
left a comment
febo
left a comment
There was a problem hiding this comment.
Looks good! I think it only needs a rebase.
|
|
||
| harness | ||
| .ctx | ||
| .process_and_validate_instruction(&instruction, &[Check::err(ProgramError::IllegalOwner)]); |
There was a problem hiding this comment.
Ah, I see, if unpack fails in the original program (which will in this case), it will then reach the ProgramError::IllegalOwner error. You are right, we can't change that. So this is correct, it is replicating the same behaviour.
grod220
force-pushed
the
create-ix-part-1
branch
from
2cc4432 to
6453e8d
Compare
2 participants
This PR adds the first portion of the Pinocchio
create_associated_token_accountflow covering the idempotent w/ existing ATA case.Also integrates mollusk fixtures to serve as conformance testing between old and new programs.