eth_multicall Meeting, Sep 23, 2024

Transcript

so from our side I am investigating the rhl problem that we have created while we were fixing an issue also I've uh cleaned up the stuff that Kari told me that should not be there in the block and in the transaction related calization part of what we return uh so uh soon will be the pair with that uh also I found another way I could check uh the timings uh because debugging on hive is sadly hard uh so I found another way I could check the

timings of a block that is getting created so yeah I think uh will'll post update for you guys to update the do image soon and uh yeah let's continue see how is your progress yeah um not a lot from my side um been busy with other stuff uh but yeah so I I was h i was U digging up some old discussion we had about the uh idea from Dano um pre-computed calls and trying to find the rational

uh we had for choosing the current approach versus the other one so that we can add something to the uh to the documents we have in this spec so that's um that's pretty much it you mind what precomputed call here means it was basically that the caller sends a list of inputs and outputs for an address so you could say that for this pre-compile if the user sends this input then you return the given output directly without processing any codes oh yeah I remember we discussed it some time ago but it was here really

far Yeah so basically the it was that we decided because it's it is touching evm code uh and among other reasons we decided against it so it's a potential for vtu uh yeah sure and karia I understand now we have a V2 dock yeah uh what dog uh a dock that is collecting V2 features potential V2 yeah I agree with that but I haven't checked if anyone has added anything and then I created the documentation for get for the each imlet call and I made a PO

request for it but no one has reviewed that yet and I wonder if NE mind also has some some kind of similar documentation and if so maybe I can add stuff there as well but I wonder that we could just create one documentation and then I think it's probably copied in many places or I don't know how the not providers for example how they are doing documentation so but I thought that we could make some one documentation somewhere and then that could be maybe copied in other places and maybe modified a bit depending on the different client uses yep appreciate oh that's I'm going to take a look at that like I was browsing internet and

trying to find documentation for each simulate and I found that nind has that kind of generated dogs and then there were some problems there and then I mentioned those to ol but I I guess that's AOG generated and and there's no yeah it's s generated and we really like generate those fields when we to meet responses to you so I'll think about how we can eliminate them but does NE mind have some kind of documentation for the apis as well or is it included in the or you should be writing it in the code and then it autog generates or I don't know how it works but I didn't find that kind of like not aerated documentation for at least it was a mix when I last

time edited them so there you can write an article and like just a mugdown article and we can publish that or we can like have Parts document of documents generated from code yeah I like how Lucas just let that claim in the in the chat that they left yep it happens sometimes that's all right so if you uh prepare any markdown I'll be happy to help to push it out into the docks uh and uh I think we have a repo with like manually created

articles on on our current state but it is in GI book so so yeah just piing me up and I'll I'll try to connect you with the repo can you share me with the location for that like for get it's also quite bizarre that it's in the get repository but it's in the different branch and it's the in the different branch there's completely different kind of project kind of similar situation here oh really okay but yeah I don't know why it's Implement well Cena said why it's Implement like like that but it's for me it's still weird do we have anything

else I don't know if Mah has been able to contact any of the providers surprisingly I haven't been able to get a hold of quick node um usually they're pretty responsive to us because our team does their support um but they've just been kind of radio silent I might try again like tomorrow since it's the weekend right now still for a lot of people I know you've posted about reth so how is re do we have anybody from rest today maybe IO don't think so yeah so apparently re released I'm

guessing they just copy it from gu is that how that works like well R right in Rust yeah so it they must have implemented a little bit different yeah I'm I'm surprised they just kind of implemented it silently but I'm happy well I haven't tested it so I don't know how good sorry sorry could you include it into the test seit yeah I could maybe see how it works I don't know if Cena would be better doing that because I'm quite terrible in the kind of figuring how that that test Su works so I know if Cena can easily add it

there yeah I can have with that yay and there'll be three of us yeah but interesting to see how well it matches that the get and N Minds outputs yeah definitely yeah agreed okay so this week hopefully we can get re integrated into the hive test suite and hopefully this week I can get a hold of quick node and after them Alchemy and infura about getting clients updated and so it's available in the wild um and then we we're still working on getting the test to actually sync up

and pass right yeah and we don't know what would be when we'll add rest how far it would be or how near worst case scenario we have three different results best case scenario we have uh two results and we can use bre as a tiebreaker I'm quite confident that we will get three different results yeah I agree that does seem most likely how far are the E call on those three what do you mean we proper tested for it oh that's that's why all right and now we had hashes and stuff

like that that definitely makes it harder there's noes there all it's easier yeah we're doing this all on hard mode we're actually writing tests we're doing hashes yep all right guys do we have anything else for today not from [Music] me all right then thank you all for coming it was productive and short and we'll do what Lucas did yeah the best kind of me yeah yeah perfect all right see you guys next week and I'll bring you up on

telegram soon bye