[3/n permissions] feat: add execute user op #69
Conversation
adamegyed
left a comment
adamegyed
left a comment
There was a problem hiding this comment.
Looks good overall, had a few questions for changes earlier in the stack. Would be nice to have a test verifying that a hook gets an entire user op, too.
|
|
||
| PostExecToRun[] memory postExecHooks = | ||
| _doPreExecHooks(_storage.selectorData[selector].executionHooks, data); | ||
| _doPreHooks(_storage.selectorData[selector].executionHooks, data, false); |
There was a problem hiding this comment.
not rebased on top of that yet!
|
|
||
| // Run any pre exec hooks for this selector | ||
| PostExecToRun[] memory postExecHooks = _doPreExecHooks( | ||
| PostExecToRun[] memory postExecHooks = _doPreHooks( |
There was a problem hiding this comment.
Same here, the executeFromPluginExternal path should be gone in an earlier PR.
| (FunctionReference hookFunction, bool isPreHook, bool isPostHook, bool requireUOContext) = | ||
| toExecutionHook(key); | ||
|
|
||
| if (!isPackedUO && requireUOContext) { | ||
| revert RequireUserOperationContext(); | ||
| } | ||
|
|
||
| if (isPreHook) { | ||
| bytes memory preExecHookReturnData = _runPreExecHook(hookFunction, data); | ||
| bytes memory preExecHookReturnData; | ||
|
|
||
| if (isPackedUO) { | ||
| if (requireUOContext) { | ||
| preExecHookReturnData = _runPreExecHook(hookFunction, data); | ||
| } else { | ||
| PackedUserOperation memory uo = abi.decode(data, (PackedUserOperation)); | ||
| preExecHookReturnData = | ||
| _runPreExecHook(hookFunction, abi.encodePacked(msg.sender, msg.value, uo.callData)); | ||
| } | ||
| } else { | ||
| preExecHookReturnData = | ||
| _runPreExecHook(hookFunction, abi.encodePacked(msg.sender, msg.value, data)); | ||
| } |
There was a problem hiding this comment.
Slight preference for not using nested if/else branches, and instead early-terminating with a continue; statement to reduce indent depth, but this is just stylistic. The workflow itself looks good for handling the different possible cases.
There was a problem hiding this comment.
The difference is mostly bytecode size by reducing JUMPs? I think this is more readable, will leave it because its RI, but production accounts should consider making this change
There was a problem hiding this comment.
Ah no I think it should be equivalent, I meant it as a readability nit.
howydev
force-pushed
the
howy/add-permission-hooks
branch
from
b183d5a to
f3ad732
Compare
howydev
force-pushed
the
howy/add-permission-hooks
branch
3 times, most recently
from
7d9f003 to
3c65bc6
Compare
howydev
force-pushed
the
howy/add-execUO
branch
2 times, most recently
from
de53ae8 to
8e244e6
Compare
howydev
changed the title
howydev
force-pushed
the
howy/add-permission-hooks
branch
from
4f8c2dc to
5d2c4ff
Compare
|
addresses: erc6900/resources#50 |
|
addresses erc6900/resources#35 |
changes:
executeUserOpin UO validationabi.encode(PackedUserOperation)if UO context is required, else sendabi.encodePacked(msg.sender, msg.value, msg.data)installValidationanduninstallValidation