Move secrets to partkey registry #3052
Conversation
Codecov Report
@@ Coverage Diff @@
## feature/partkey #3052 +/- ##
===================================================
+ Coverage 43.96% 43.97% +0.01%
===================================================
Files 394 394
Lines 87467 87511 +44
===================================================
+ Hits 38451 38486 +35
- Misses 42928 42934 +6
- Partials 6088 6091 +3
Continue to review full report at Codecov.
|
winder
force-pushed
the
will/install-secrets
branch
from
17c719b to
f4cfae1
Compare
|
|
||
| // update cache. | ||
| // TODO: simplify to re-initializing with initializeCache()? | ||
| db.cache[id] = ParticipationRecord{ |
There was a problem hiding this comment.
this looks an awful lot like .Duplicate() ?
(and also things should only be assigned if the source was not nil)
There was a problem hiding this comment.
Yes, I didn't see a good way to reuse it.
| record.LastValid, | ||
| record.KeyDilution) | ||
| record.KeyDilution, | ||
| rawVRF) |
There was a problem hiding this comment.
I see that you're storing nil value directly into the database here. This isn't strictly wrong - but it might require you to work abit harder on the query side ( i.e. sql.NullString ). Please make sure you have at least a single unit test to verify the correctness of this code.
There was a problem hiding this comment.
It turns out you don't need special handling here because the type is []byte, not string. So nil checking is done in the normal way by looking at len(vrf) and len(voting).
There is a unit test called TestParticipation_EmptyBlobs that is supposed to cover that case. I just used the debugger to double check and can confirm that the query returns nil when scanning results and it is handled as expected.
4 participants
Summary
Start moving partkeys away from the old file format.
Test Plan