You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Tooling such as the wallet need access to time conversions, and it makes sense to make this available through the LSQ protocol, because the HFC goes through great lengths to track the necessary information. Since tooling might need to do many conversions, it might make sense to make a LSQ query available that returns an interpreter for Qry (from Ouroboros.Consensus.HardFork.History.Qry). Tooling should be aware however that such an interpreter will have a limited range, and will need to be updated regularly.
The text was updated successfully, but these errors were encountered:
Incidentally, we have been wondering whether the simplification that the consensus layer makes that time conversions should not be subject to rollback is perhaps not strictly required, thereby giving is the possibility of removing the requirement for the double stability window. If we are now making time conversions available to tooling (such as the wallet), that simplifying assumption becomes more important, I think. Perhaps we can check our uses of conversions and verify that if they are subject to rollback it’s fine, but now that we are making this available to tooling, I much prefer to be able to say “yes, we know the answer for sure, here it is” or “we don’t know that yet” rather than “here’s a guess, but be careful, if we rollback, it might change”.
Fixes#2365.
* Introduce `QueryHardFork` for queries specific to the hard fork combinator,
unrelated to any specific era. Provide a query to ask for an `Interpreter`,
which can convert a slot to an epoch or time, etc.
* Generalise `QueryAnytime` so that the first era can also be queried. We
still require there to be more than one era. Add the `QueryAnytimeByron`
pattern synonym.
* Rename the `EraStart` constructor of `QueryAnytime` to `GetEraStart` for
consistency.
* Change the encoding format of hard fork queries. Since they are not part of
the final Shelley release yet (right?!), this is fine.
2370: Introduce QueryHardFork r=mrBliss a=mrBliss
Fixes#2365.
* Introduce `QueryHardFork` for queries specific to the hard fork combinator,
unrelated to any specific era. Provide a query to ask for an `Interpreter`,
which can convert a slot to an epoch or time, etc.
* Generalise `QueryAnytime` so that the first era can also be queried. We
still require there to be more than one era. Add the `QueryAnytimeByron`
pattern synonym.
* Rename the `EraStart` constructor of `QueryAnytime` to `GetEraStart` for
consistency.
* Change the encoding format of hard fork queries. Since they are not part of
the final Shelley release yet (right?!), this is fine.
Co-authored-by: Thomas Winant <[email protected]>
Tooling such as the wallet need access to time conversions, and it makes sense to make this available through the LSQ protocol, because the HFC goes through great lengths to track the necessary information. Since tooling might need to do many conversions, it might make sense to make a LSQ query available that returns an interpreter for
Qry
(fromOuroboros.Consensus.HardFork.History.Qry
). Tooling should be aware however that such an interpreter will have a limited range, and will need to be updated regularly.The text was updated successfully, but these errors were encountered: