Meeting Minutes 7
===

###### tags: `BDK Core` `Meeting`

:::info
- **Location:** Online
- **Date:** Nov 11, 2022 09:00 AM (UTC)
- **Agenda**
- Weekly standup
- **Participants:**
- Evan (@evanlinjin)
- Daniela (@danielabrozzoni)
- Lloyd (@LLFourn)
- Raj (@rajarshimaitra)
- Vlad (@vlad_kwasi)
- **Reference:**
- [Project board](https://github.com/users/LLFourn/projects/2/views/1)
- [Last meeting](https://hackmd.io/@evanlinjin/bdk_core_5)
:::

:timer_clock: Minutes
--
- Standup
- Evan
- Metadata for sparsechain
- LL
- New idea for persistence (database instead of flat file)
- Alekos & Daniela
- Not too much progress on planning module :(
- Alekos and Daniela worked on reviewing the PR and bdk_core
- Questions to be asked later :sparkles:
- Daniela worked on hws
- Vlad
- Persistence: use a simple database (linklist?)
- Kind of working! :grin:
- Raj
- Not much time on core related work
- Q & A
- Alekos: SpkTracker keeps track of unused spks, wouldn't it be easier/more efficient to keep track of the tracked ones?
- Evan: Yay!
- Raj: `iter_unused` becomes too slow
- Alekos: In the Sparsechain code there are checks on the tx when applying the update, do you make sure that the update doesn't give you txs that were confirmed before the checkpoint that you're updating from? i.e., if I see an update that builds on top of my checkpoint (no reorgs), do you make sure that in the update there isn't a tx that says that it's confirmed before the checkpoint? (Can't insert txs that were confirmed in the past)
- Evan: you can update txs in the past and it's on purpose, maybe for some implementations you want to update keychains separately. We do ensure that we have a checkpoint that it connects to the update. For example, sparsechain has blocks height 10 & 11, if I do an update with height 10 & height 12, we wouldn't know which history to use; if we introduce blocks before and connect to height 11 it works.
- Alekos: Let's say I'm doing an update from block 10 to 12, is it garanteed that the update contains all the txs between blocks 10 and 12?
- Evan: no guarantee, just some
- Raj: Is it possible that block 12 contains txs confirmed in block 9, but didn't get in block 9 for some reason?
- Evan: basically yes
- Alekos: is the txgraph monotone?
- Evan: sparsechain is non-monotone, txgraph is monotone
- Raj: How we're going to test our assumptions?
- Evan: tests should already cover most, but we might have to write more