Seraphis wallet workgroup meeting #26 - Monday, 2023-06-12, 18:00 UTC about meta HOT 1 CLOSED

<rbrunner7[m]> Meeting time. Hello! https://github.com/monero-project/meta/issues/847
<dangerousfreedom> Hello
<rbrunner7[m]> Hmm ... just the two of us? :)
<Rucknium[m]> waves
<valldrac[m]> Hi
<dangerousfreedom> Now 3 :p
<jberman[m]> hey I'm here, sorry
<rbrunner7[m]> Splendid! Well then, anything to report? I have been reviewing dangerousfreedom[m] 's PRs a bit and left some comments
<dangerousfreedom> This week I continued implementing the wallet knowledge proofs (for legacy and sp) and took a look at jberman's scanner. A question raised from that: what is the easiest way to communicate with the node in c++?
<rbrunner7[m]> From where?
<rbrunner7[m]> I don't know anything else than the RPC interface
<dangerousfreedom> In jberman's code and at wallet2 you have to set the m_http variable... I was wondering if I should create a module for that or if there is something easier to do
<dangerousfreedom> I will need to request many information from the node and was wondering the easiest way to do it
<rbrunner7[m]> Isn't that just the object for the RPC connection?
<dangerousfreedom> I didnt look deep what this object is doing and its dependencies yet but I guess it wont be hard but instead I just asked to save me some work :p
<dangerousfreedom> But I will find out this week and the goal is to have the spend_proof by next week
<dangerousfreedom> I need something like this: https://github.com/monero-project/monero/blob/master/src/wallet/wallet2.cpp#L10864
<jberman[m]> I don't think the code I have there is the simplest way, but basically you initialize an http client and use the RPC interface to communicate with the daemon
<dangerousfreedom> Ok. Yeah, I will do something like that
<jberman[m]> this is sort of contained right here: https://github.com/j-berman/monero/blob/1180c8262e405708411d166ef3cb36bd5f98cb92/src/blockchain_utilities/blockchain_scanner.cpp#L203-L213
<dangerousfreedom> Would it be worth to write something modular?
<dangerousfreedom> Or is there something already?
<rbrunner7[m]> I don't think it will get much easier than that, once you know the basics ...
<rbrunner7[m]> Nor is the code very hard to write
<rbrunner7[m]> Tedious maybe, and quite low-level, but why complicate things with yet another layer ... IMHO
<dangerousfreedom> Ok thanks.
<valldrac[m]> In wallet2 there are two functions to call RPC endpoints: epee::net_utils::invoke_http_json_rpc and epee::net_utils::invoke_http_bin. jberman snippet code uses the former 
<valldrac[m]> You need to provide the http client to these functions. I think it's the m_http variable you said 
<shalit[m]> Hello
<dangerousfreedom> Thanks. Sometimes I'm ashamed of my lazyness but I'm improving :p
<rbrunner7[m]> I wondered about another little thing when I reviewed the code in the PRs: What do we do with links to web pages in source code comments
<rbrunner7[m]> I mean, we all know about link rot, but is this already reason enough to not include any links?
<rbrunner7[m]> For example we have a very special Base32 alphabet. There is some page written by Tevador explaining the reasoning.
<rbrunner7[m]> Does it make sense to link to that from the source coude or not?
<rbrunner7[m]> What's the sentiment here?
<dangerousfreedom> I would say it is worth. When the link is not valid anymore we remove.
<jberman[m]> +1^ and include a snippet of reasoning behind what the source link says so it's not doing all the lifting on explanation
<valldrac[m]> Yeah, I often include links to external pages in source code
<Rucknium[m]> Create a snapshot with the Wayback Machine and directly link the snapshotted URL
<rbrunner7[m]> Yeah, the most important, core info copied and a link is of course best :)
<rbrunner7[m]> Ah, you can ask the Wayback Machine for a snapshot?
<Rucknium[m]> Yes
<rbrunner7[m]> Interesting
<valldrac[m]> Unless it's something essential or part of the specification, you can just put it in our wiki or directly in the source code as a long comment :)
<Rucknium[m]> https://archive.org/web/  "Save Page Now" in lower right corner
<rbrunner7[m]> Sure, but I guess our own resources are not necessarily more stable
<rbrunner7[m]> Rucknium[m]: Good to know, thanks for the tipp
<valldrac[m]> Another option is as .md into /docs in the repo
<rbrunner7[m]> There are many options, if the willingness is there, right. But I read the sentiment as "Links can make sense, and they are at least not discouraged"
<valldrac[m]> +1
<rbrunner7[m]> Alright. Anything else for today's meeting?
<dangerousfreedom> Not from me
<valldrac[m]> Is there something that needs to be reviewed?
<rbrunner7[m]> You could have a look anytime of the 3 PRs that dangerousfreedom[m] made
<dangerousfreedom> I will keep updating this draft so my work will be visible there. 
<dangerousfreedom> Now it is in a very raw situation
<dangerousfreedom> Not worthy of detailed review but general advices
<rbrunner7[m]> Understood.
<rbrunner7[m]> Does not look like we have more. Thanks for attending, read you again next week at the latest, the last meeting before Monerokon, hurray.
<rbrunner> Our PRs are here, by the way: https://github.com/seraphis-migration/monero/pulls