 Hello. Hey, Peter, can you hear me? Hey, yeah. Can you hear me? Yes. Okay. That's Jake Sandeep. Okay. He's joining. So, uh, you had to, you were trying to do some experiments on the package publishing, right? Yes, or Sandeep showed me something you had in, uh, in a pinch. Yeah. Unfortunately, I have not had the time to do more on it since then, but I'll definitely get something done this week. Okay. So, from my side, not too many updates just yet, but I've been working on it. Sure. Yeah. I'm also been waiting for the packages to get published so we can start experimenting with them. Uh, so the API, uh, I think Sandeep had, uh, so we have one PR and Sandeep has another now. Can you talk about them? Okay. Yeah. So the PR that got merged, that I updated all the names and the targets that it should be published to CACTA. And now the second PR, I am adding that will add the workflows that will publish the packages in the CACTA. Sounds good to me. Yeah. Maybe Peter, can you review this? So I added the logic where on the tag release, it will take the version from that and make a release of the package. I tried to use something similar to what CACTA already has. CACTA's already has some published workflows. Oh, sorry if I didn't look at it yet. Yeah. I mean, it's not hurry. I just, I opened it yesterday only. So. Yeah. Honestly, I, I want to, my aim is to review pull requests either the same day or the next day, the latest. Uh, it's not easy, but we should, I think, strive for that because especially if the CI was faster, we would, we should be able to merge much more pull requests. Much faster than what we're currently at. So I definitely want to improve that significantly. I mean, I want to speed it up 10 times easy. Yeah, I'm guilty of being slow at reviewing myself. It's something better to remind me, but I will, I'll also try to be faster at it. Thank you. And we have, I mean, the final thing we've been discussing the, in the context of project perspective is to write. Yeah. Okay. I proved it. I was just looking at it right now. It looks good to me. You know, it's, I didn't look at it in too much detail, but I did take glance and if there's something in there that I very much disagree with for whatever reason that I didn't realize we can always talk more about it later and then just change it maybe if it ends up being like that. Okay. How many does it appear? It's two votes. It's two companies. But since I just approved it from Accenture, you Rama could approve it from IBM and then that's it. Yeah, I'll do that. The only thing is that if at least two organizations agree that something should happen then it should happen because that's currently the majority. That's a 66% majority. Right. And I'm, I have this other idea to write a GitHub action based bot that would make it possible for us to actually enforce this logic. Automatically. So that it would have, it would parse the maintainers file. And then it would have a configuration that says not just that it's any two reviewers, but two reviewers from two separate organizations. And then it would block the merge until at least two different organizations approved it. And then with that, we could enable auto merge so that the moment the necessary approvals are there, it just gets merged immediately. And then we are on to the next poll request. That's cool. So right now, it's just the block is only for two words, any two words, right? Yeah, any two organizations out of the three. No, I mean, currently, so say, if you and Jacpreet voted, would the PR go through? Oh yeah, yeah. So GitHub's logic is not aware of the organizations. So yeah, that's the problem. That's why we cannot enable auto merge right now. Yeah, because then me and Jacpreet could accidentally get things merged just because both of us approved it and you didn't even have a chance. For example. I understand. Yeah, I think among us. All the lead maintenance, we know to it for other companies. Yeah, they're right now we're all just complying with the rules that we know of. Yeah. But it would be more efficient if more automation could be in it. Yep. So Sunday back with PR. This for this is the last when this will publish all the packages we need right for the previous. So you can come again. The PR that you currently have that will publish the packages we need for the previous version. Yes, whenever we'll have to release a tag for that and then it will publish. Okay. And the go packages for go packages it's not there right now so we I still couldn't figure out how to do that in one single release. Yeah, that's on me. I said I would try it as well but I have not yet time to finish my investigation. I'll put some more effort in this week. No problem. But in the meantime, how do we probably to go packages. Ah, it was per package isn't it. Yeah. Sorry that's just super inefficient. I was trying to create a script that would generate all the check go checksms in one commit. But I'm not sure whether it will be 100% for proof. Yeah. Yeah, not easy. If you can share that script. I'm also happy to take a look at that maybe, you know, the two of us together, figure something out faster. Sure. If you want you can also commit it to that. I can invite you to be a collaborator on on my sandbox repo you know the one that I showed last time. Yeah, yeah, sure. Yeah. Let me invite you give me one second operators. All right, give me give me two more seconds because it just hit me with the two factor authentication the phone is in the other room. Give me a sec. Okay, it says that the invite has been sent. Okay, let me check. Can you share the link to the local. Yes, one second. Read on the zoom chat. Okay. So, I'll add the script here to test. Okay, then I can also take a look and also feel free to use the repo as well if if there's something that you want to test that is also difficult to test in in the actual official repo. This is a good lab environment and you know, nothing bad is going to happen if you put commits in there. I don't care about it. It's just a test repo. Okay. Yeah, I don't do anything else right now. I had asked something on the discord so it's about the node modules. There are some conflicts in the node modules, because there's there's a global node modules folder in the root of the cacti right when we clone it. And that's causing some conflicts with the viewer packages. I don't know, like how cactus have avoided that but just wanted to know how to fix that. So that's the mono repo. Both yarn and npm supports it and they are called a sort of a mono repo they're called workspaces. So if you want to read up on it, then there's here's a link I put on the chat. So what I'm thinking is that probably your package JSON is just not yet set up to be part of the workspaces. Okay. And yarn is getting confused about whether it's a workspace base repository or a non workspace base repository. That's the best thing I can imagine. And probably the solution is to tell it that the viewer package is the node JS viewer packages are part of the list of packages that are in the project. And you can do that with globs in the learner that Jason file. The packages and tree that has globs. So I would try to add it there. I'm trying to add here's the logs image. Yes, I put that also in the group chat on discord. Okay. I'm happy to fix it, but if not, I'm happy to take a look myself. Okay, I'll try this. If nothing else, then I'm sorry for joining. I'll talk to you next time. Thank you. Thanks. Bye.