Seraphis wallet workgroup meeting #63 - Monday, 2024-03-25, 18:00 UTC about meta HOT 1 CLOSED

<r​brunner7> Meeting time. Hello! https://github.com/monero-project/meta/issues/982
<r​brunner7> Can you please add "reading" to your hello
<r​brunner7> so I know who can read me and who probably not
<d​angerousfreedom> Hello. I can read you now :)
<j​berman> *waves*
<r​brunner7> jberman: Can you read me?
<SNeedlewoods> hey, I joined IRC so hopefully see the best of both worlds
<s​needlewoods_xmr> hey
<r​brunner7> Perfect, SNeedlewoods . That probably works.
<SNeedlewoods> rbrunner your message didn't come through to matrix side
<jeffro256> Howdy
<s​needlewoods_xmr> jberman can you join irc? message from rbrunner does not show here
<jeffro256> I'm joining from IRC since monero.social -> matrix.org Matrix messages don't work 
* jberman (~jberman@2001:41d0:a:2aab::1) has joined
<r​brunner7> jeffro256: Do you see what I write as "rbrunner7" on Matrix?
<jberman> *waves from irc*
<jberman> I don't, I only see sneedlewood's message
<rbrunner> jeffro256: Do you see what I write as "rbrunner7" on Matrix?
<d​angerousfreedom> I only changed the server to monero.social and now I can see the messages
<j​effro256> rbrunner7: yes, but i'm on the monero.social server
<rbrunner> So maybe better I also use IRC for today?
<rbrunner> Alright, seems more or less ok then to start.
<rbrunner> What is there to report from last week?
<d​angerousfreedom> This week I could advance in putting everything together (EnoteStore, TxHistory, KeyContainer) and created a demonstrator of a wallet. The goal was to make a few transactions between two wallets, retrieve the proper balance, show the enotes and make knowledge proofs on them. Here it is:
<jeffro256> Holy moly that's awesome !
<d​angerousfreedom> I would like to kindly ask your feedback about the way things are going. I believe this could really look like a version0_pre_alpha_proto seraphis wallet :p
<jeffro256> How were transactions transmitted?
<system> file Screencast from 25.03.2024 18:21:17.webm too big to download (17663360 > allowed size: 1000000)
<d​angerousfreedom> Screencast from 25.03.2024 18:21:17.webm
<SNeedlewoods> I got feedback from koe and jberman and worked on https://github.com/UkoeHB/monero/pull/40
<SNeedlewoods> but currently my local branch is quite messy and I introduced bugs I need to fix, can still push if someone wants to see?
<d​angerousfreedom> Of course there are really a lot of things to do like:
<d​angerousfreedom> - Use the proper KeyContainer. Right now I'm using the previous version of seraphis and waiting for ghostway  KeyContainer.
<d​angerousfreedom> - Update everything (address scheme, knowledge proofs, etc) with the jamtis updates proposed by jeffro256 
<d​angerousfreedom> - Use a real ledger instead of the `MockLedgerContext`. Some comments here:
<d​angerousfreedom>     - Right now I'm de/serializing the MockLedgerContext to have a local permanent storage. This is not optimal, it is buggy now but it allows to test many wallet functions. (Thanks to Koe and his modular library design).
<d​angerousfreedom>     - Only the storage of seraphis enotes is working now. I will try to fix the serialization for the legacy ones next and try to incorporate the legacy scanner from jberman
<jberman> Update: the async wallet scanner is now backwards compatible with today's node versions, scanning perf is on par with wallet2 today when pointing to a current node (the async scanner's significant speed-up requires the daemon to update). I'm currently wrapping up the PR with final touches, cleaning up class structure/naming/commit history etc.
<jberman> Shooting for ready-for-review this week
<SNeedlewoods> before that I started this https://github.com/UkoeHB/monero/compare/seraphis_lib...SNeedlewoods:seraphis_wallet:x_identify_edge_cases
<d​angerousfreedom> I'm serializing the MockLedgerContext and loading/saving it everytime I open/close a wallet
<rbrunner> ghostway announced shortly before the meeting started that they will make a PR tomorrow.
<d​angerousfreedom> There are really a lot of clean-ups and more comments to make but I think it would be better to have your feedback on it first. Even in a high level. You can play with the wallet (open many wallets and transfer enotes between them). -> https://github.com/DangerousFreedom1984/seraphis_lib/commit/9340e037e763cc69619d3a27883327e2bf060437
<d​angerousfreedom> Just compile and run ./monero-wallet3
<rbrunner> Great, dangerousfreedom!
<rbrunner> A lot of things finally start to drop into place
<rbrunner> My little contribution was yet another rebase of our "seraphis_wallet" branch to koe's latest "seraphis_lib"
<rbrunner> I know and understand now the necessary trick, as explained patiently by jeffro256
<SNeedlewoods> dangerousfreedom I'll try to check  your monero-wallet3 out soon
<rbrunner> jberman: I think everybody is also looking forward to your PR. Another big piece.
<rbrunner> Will this a lib PR, a wallet PR, or a two-part PR?
<rbrunner> *Will this be
<d​angerousfreedom> Thanks! I will look at your PR soon also since my serialization is not working for legacy yet and it might be better to even go to a real ledger and then I would need to use your PR with jberman's scanner. I will think about it this week.
<jeffro256> Me: https://github.com/UkoeHB/monero/pull/31, https://github.com/UkoeHB/monero/pull/34, and https://github.com/UkoeHB/monero/pull/38 were merged into seraphis_lib. I'm working on a PR which will switch the way we index legacy enotes to match how we currently index enotes (amount, global index), so that we can have CLSAG rings in seraphis transactions w/o re-indexing the database 
<rbrunner> Sounds a bit like magic ...
<rbrunner> Ah, no, just a switch back to the "old" system?
<jeffro256> yes
<jeffro256> So heads up @dangerousfreeedom, the internal structure of MockLedgerContext will change with this PR I'm working on, and thus the serialization code will have to update 
<rbrunner> jberman, do you copy, this was a question meant for you: Will this a lib PR, a wallet PR, or a two-part PR?
<jberman> rbrunner I'm going to divide up the commits as cleanly as possible including small commits to the daemon RPC today, and then from there I think it'll be easier to see what should be PR'd where and at what time etc. Probably will be a number of small PR's
<rbrunner> Just making sure that we all are on the same boat regarding PRs
<d​angerousfreedom> Yeah, everything is changing everytime haha. But hopefully we will start having an idea about the core of the wallet
<rbrunner> Thanks, jberman, sounds good
<jberman> jeffro256 have you seen this PR from SNeedlewoods that introduces an `enote_same_amount_index` ? https://github.com/UkoeHB/monero/pull/40
<jberman> *enote_same_amount_ledger_index
<rbrunner> Does this solve the same problem?
<rbrunner> Or is this an addition?
<jberman> It would enable indexing by amount + global index, same as today
<jeffro256> No IIUC, they should work together. That PR is for adding contextual information to legacy enote records, while the PR I'm working on changes the `LegacyRingSignatureV4` to include an enote amount for indexing the input members 
<rbrunner> I guess jeffro256 will sort it out :)
<jberman> gotcha
<jberman> noice
<rbrunner> Alright. Do we have any separate subject to discuss today? Anybody want to bring up some general things?
<rbrunner> Nice that my incremental pool update code now becomes useful, "thanks" to the spam wave. Now we "only" have to get everybody to update
<d​angerousfreedom> Haha thanks :p
<jberman> +1
<jeffro256> yeah that's great that incremental pool updating was implemented and tested before it was really really needed 
<jeffro256> Usually not how the dev cycle goes here ;)
<rbrunner> Well, I did not have spam waves on my mind ...
<rbrunner> Ok, if nothing else I think we can close for today. Thanks everybody for the fantastic news regarding progress, and your attendance. Read you next week.
<jeffro256> Unrelated, and not to be selfish, but https://github.com/UkoeHB/monero/pull/26 still needs review, and it will affect most wallet-related things downstream. If anyone is brave enough to help review, Koe and I would greatly appreciate it 
<jeffro256> But yes thank you everyone! Things are speeding up which is nice to see 
<jberman> on my list :)
<rbrunner> Good luck, I hope you find a reviewer. That won't be me, however, that's definitely beyond my reach