P-chain UTs consolidation 2 - Context and Config #2121
Conversation
abi87
force-pushed
the
pchain_UTs_setup_consolidation
branch
from
0a3cb46 to
079eac8
Compare
abi87
changed the title
abi87
changed the title
|
|
||
| var ( | ||
| fork = ts.DurangoFork | ||
| forkTime = ts.ValidateStartTime.Add(-2 * time.Second) |
There was a problem hiding this comment.
different packages still have different fork times set. We'll fix this in next PR #2169
| var ( | ||
| fork ts.ActiveFork | ||
| forkTime time.Time | ||
| ) | ||
| switch { | ||
| case postDurango: | ||
| fork = ts.DurangoFork | ||
| forkTime = ts.ValidateStartTime.Add(-2 * time.Second).Add(-2 * time.Second) | ||
| case postCortina: | ||
| fork = ts.CortinaFork | ||
| forkTime = ts.ValidateStartTime.Add(-2 * time.Second).Add(-2 * time.Second) | ||
| case postBanff: | ||
| fork = ts.BanffFork | ||
| forkTime = ts.ValidateEndTime.Add(-2 * time.Second) | ||
| default: | ||
| fork = ts.ApricotPhase5Fork | ||
| forkTime = ts.GenesisTime | ||
| } |
There was a problem hiding this comment.
kind of arbitrary, but this is what we have been using fo far. Will be cleaned up in next PR #2169
| ApricotPhase3Time: ts.ValidateEndTime, | ||
| ApricotPhase5Time: ts.ValidateEndTime, | ||
| BanffTime: time.Time{}, // neglecting fork ordering this for package tests |
There was a problem hiding this comment.
this is partially fixed in this PR, meaning that at least we sort forks in the right order in testsetup package. In next PR #2169 we'll complete the fix by using the same fork times across different packages
| fork = ts.ApricotPhase5Fork | ||
| forkTime = ts.ValidateEndTime |
There was a problem hiding this comment.
these values come from defaultConfig (now ditched in favour of ts.Config())
| func Context(tb testing.TB, baseDB database.Database) (*snow.Context, *MutableSharedMemory) { | ||
| ctx := snowtest.Context(tb, snowtest.PChainID) | ||
| m := atomic.NewMemory(baseDB) | ||
| msm := &MutableSharedMemory{ | ||
| SharedMemory: m.NewSharedMemory(ctx.ChainID), | ||
| } | ||
| ctx.SharedMemory = msm | ||
|
|
||
| return ctx, msm | ||
| } |
There was a problem hiding this comment.
moving this to its own package seems over fragmenting
| }, | ||
| ApricotPhase3Time: genesistest.ValidateEndTime, | ||
| ApricotPhase5Time: genesistest.ValidateEndTime, | ||
| BanffTime: time.Time{}, // neglecting fork ordering this for package tests |
There was a problem hiding this comment.
I start changing this partially in this PR and finally fix fork times ordering in next PR
| var ( | ||
| fork configtest.ActiveFork | ||
| forkTime time.Time | ||
| ) | ||
| switch { | ||
| case postDurango: | ||
| fork = configtest.DurangoFork | ||
| forkTime = genesistest.ValidateStartTime.Add(-2 * time.Second).Add(-2 * time.Second) | ||
| case postCortina: | ||
| fork = configtest.CortinaFork | ||
| forkTime = genesistest.ValidateStartTime.Add(-2 * time.Second).Add(-2 * time.Second) | ||
| case postBanff: | ||
| fork = configtest.BanffFork | ||
| forkTime = genesistest.ValidateEndTime.Add(-2 * time.Second) | ||
| default: | ||
| fork = configtest.ApricotPhase5Fork | ||
| forkTime = genesistest.GenesisTime | ||
| } |
There was a problem hiding this comment.
this is just a refactoring of the way we set fork times right now. We'll fix this in next PR
|
This PR has become stale because it has been open for 30 days with no activity. Adding the |
Why this should be merged
In #2107 we moved to a
genesistestthe machinery to build UTs genesis. Here we keep on going withvm.Configandvm.Ctx, moving them to aconfigtestpackageThanks @joshua-kim and @marun for the discussions that brought me to these changes.
How this works
Just refactor UTs setup
How this was tested
CI