Runtime ENVironment configuration #179
Conversation
| "local" -> Right Local | ||
| s -> Left $ TextDecodingError $ | ||
| T.unpack s <> " is neither \"mainnet\", \"testnet\", \"staging\" \ | ||
| \nor \"local\"." |
There was a problem hiding this comment.
Moved to Cardano.Environment
| @@ -64,4 +66,4 @@ spec = do | |||
| threadDelay 1000000 | |||
| (handle,) <$> (mkWalletLayer | |||
| <$> MVar.newDBLayer | |||
| <*> HttpBridge.newNetworkLayer "testnet" port) | |||
| <*> HttpBridge.newNetworkLayer Testnet port) | |||
There was a problem hiding this comment.
Seeing this… 🤔 won't this lead to inconsistencies between the network layers and the CLI param in future tests?
There was a problem hiding this comment.
Hmmm.. I am not sure to see what you mean here? There's no more "--network" in the CLI, and the CLI itself reads the environment to get this network value.
There was a problem hiding this comment.
If we call mkStdTx the protocol magic it uses will depend on network and the ENV var. If we ever create a network layer with a custom network, like here (Testnet), we can never safely use the two together.
Would it be possible to have one network for the entire integration tests?
Also:
- Does the choice of network matter for the integration tests? (Could we default to
testneteverywhere and be done with it?)
There was a problem hiding this comment.
@Anviking Hmm... That's not possible no, because we are running some integration tests against testnet, but some against a local cluster (the scenarios) because we need the ability to have faucets wallets there, and submit transactions at will.
So yeah, if the application is configured with a protocol magic / network that is different from the one used by the bridge, we can an error; but that's pretty much what we have already anyway.
| -- | ||
| -- ENV[NETWORK] = ? | ||
| -- | ||
| -- What about trying to provide a valid ENV var `NETWORK=value cardano-wallet-launcher` ? |
There was a problem hiding this comment.
will this error be presentet using stack exec -- cardano-wallet server .... as well? If it will be, maybe this error message will be missleading as it constrains only to cardano-wallet-launcher binary
EDIT: I see command is not hardcoded - so everything is ok
| deriving (Generic, Show, Eq, Enum) | ||
|
|
||
| instance FromText Network where | ||
| fromText = \case |
There was a problem hiding this comment.
I guess both FromText and ToText instances can be made more generic by using Show and Read classes - but Network seems so small and selfcontained that I don't find that necessary
| -- Note that the 'ProtocolMagic' for 'Local' and 'Testnet' are the same. | ||
| protocolMagic :: Network -> ProtocolMagic | ||
| protocolMagic = \case | ||
| Mainnet -> ProtocolMagic 764824073 |
There was a problem hiding this comment.
I hope these hardcoded values are right - because I am not sure where to look up for the proof/source that these are the right numbers?
Maybe it would be usefull for other readers to add a comment where from or why these specific numbers were picked
There was a problem hiding this comment.
Actually, I am not even sure where I got them from :| ... I think I now mostly go for the ~/.hermes configuration files generated by the http bridge to be honest...
….CLI to Cardano.Environment
KtorZ
force-pushed
the
KtorZ/95/runtime-environment-configuration
branch
from
7a9d3b7 to
4931fb5
Compare
Issue Number
#95
Overview
Comments
Deprecates #162