fix: Use median-time-past for input sequence checks rather than block…

keep-starknet-strange / raito

Bitcoin ZK client written in Cairo.

https://raito.wtf

MIT License

40 stars 34 forks source link

fix: Use median-time-past for input sequence checks rather than block… #239

Closed fishonamos closed 3 weeks ago

fishonamos commented 4 weeks ago

… time

[x] resolves #231
[x] follows contribution guide
[x] code change includes tests

vercel[bot] commented 4 weeks ago

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
raito	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Oct 11, 2024 0:38am

fishonamos commented 3 weeks ago

@m-kus. Kindly review so i can proceed with the scripts.

m-kus commented 3 weeks ago

Please rebase

fishonamos commented 3 weeks ago

Please rebase

DOne

maciejka commented 3 weeks ago

MTP in test data is larger than the block_time before changes — this means something is wrong (previous timestamps can be larger than current block time, but those are exceptions)

Concrete example would be helpful.

m-kus commented 3 weeks ago

Ok @fishonamos I misguided you regarding the MTP for the outpoints, we need MTP for the blocks where those outpoints were created not the blocks where they are spent. It's a more broad issue actually, because we also need to re-build our Google BigQuery dataset.

I'll make few changes to the PR and merge after.

fishonamos commented 3 weeks ago

Ok @fishonamos I misguided you regarding the MTP for the outpoints, we need MTP for the blocks where those outpoints were created not the blocks where they are spent. It's a more broad issue actually, because we also need to re-build our Google BigQuery dataset.

I'll make few changes to the PR and merge after.

Ok ser

m-kus commented 3 weeks ago

I'll follow up in a separate PR