 Welcome, everyone. It's the ninth of June. This is the get lab plug in modernization project meeting. So harsh. I've got items on the agenda, including pending pull requests. Any discussion if we need to have some discussion from last week. I put mentor checklist on there is a standing thing, just in case and an open question is this time still working for everyone. Are there other things you'd like to put on the agenda. Like, for the midterm evaluation, like my and semester examinations are classic. Right, right. Okay, that's a very important one. So upcoming upcoming schedules and events. Very good. Okay, so midterm evaluation midterm mid or not. That's the mid you said examinations for university are coming right. Dates and preparation. Good. Okay. And you mentioned midterm evaluation. Are you talking about the project or yeah. Okay. All right, term evaluation for the project. Okay, like the presentation that we have to give. So that was clashing. Oh, that will clash with your examinations. Okay, all right, good to know. All right, so we may need to have you record it separately and play it back or something to do it earlier so that it doesn't collide with your exams that examinations take first priority. Okay, good. So upcoming schedules and events is a good topic. Any other topics you want on the agenda. Other than this get lab also announced get lab 16 which did some breaking here and there in the get lab plugin and that is also going to be fixed I think in this project only other than this I can't think of anything. Okay, so well let's put that one I would love to hear from you on get lab 16 breaking changes because you've, you're probably the first, first explorer of that so let's put it there. Great. Anything else for the agenda. On my side like I would like to discuss more on these topics. Oh, and we will. Great. Okay, so let's go through them then. First check on the action items so we had said a UML diagram but decided that was lower priority than your work on the pull request I assume it's still lower priority are you okay with that harsh. Great. And a snapshot of version six of get lab for Jay. This I think is also lower priority because we, we assume you're using the five dot x version. Okay, good lower priority, but I tried doing that and this was a complete mess for me. Like, I was not able to build a snapshot because I like first of all I was using the MacBook so I had to update the get lab for the library to have the Docker Maven plugin of fabric eight to work. Like, after doing that the integration tests of the library started to fail and I and Chris, I and Chris were trying to fix that I was not able to fix it. And Chris was also tired after a lot of stuff that we tried to do and it was a bad experience overall, like we couldn't. So I just filed an issue in the get lab for the GitHub account. And I don't know, like, how will I do that like I tried it on my m one chip like it did not really work so I also tried it on get bought, and it still didn't work like that was. Like, I was able to do it for just deploying a snapshot it might be easier to skip the tests because you don't need to run the tests. What were you having an issue compiling it or running the tests. Like it was compiling but the tests for not running like I was doing maybe install and it was not working. Yeah, it might be easier to just skip the tests for the purposes of creating the snapshot. Because the point of this is not the point of this is not for us to, you know, do a release of the get lab for J library which which, you know, if we were actually doing a release of that ourselves we would care about the tests not working but we're just getting a snapshot for our own internal use. It would be fine to skip the tests I think. That's, that's actually really good. Otherwise I would have been a dead meat. Okay. All right. Anything else on the action items before we go on to schedules and events. No, I don't think I tried anything else in the week, except the full request of course. Right. Okay, so, so tell us what are your examination dates and your preparation time so we make sure we like. July 6 to July 13 is my examination, and I think the midterm evaluation presentation is also flashing in that time period. So I may need to record the presentation. Okay, so let me let me do a quick look at the where would I look probably the Jenkins calendar, or or let's let's let's just let's not take time in the meeting I'll double check it this is something I can do separately. I can check the the presentation date. Discuss alternatives with John Mark with the org admins, because I think they will agree wholeheartedly your examinations are much more important than you doing a presentation. We want the presentation, but we could do a recording of it earlier. There are all sorts of ways we could consider doing it. Okay, did anything else on upcoming schedules. All right, tell us about get lab 16s breaking release and breaking changes, what have you learned like what did they do what they did it's like for making the stuff projects work better and to improve the performance of get lab server. They introduced delimiters which are actually creating problems when I when the user is trying to click on when the user is trying to go to the changes that they committed using the Jenkins build changes section. They are redirecting to a wrong page so I'll have to correct the URL structure for that so that's like others. No, that's the only thing that they did that I know that the issues were raised in the get lab plugin. So I will have to solve them also. That issue would affect users of rest easy to would it not. Yeah, yeah, it's, it will be affecting the rest easy as well as get lab. Yeah, so this seems like it's kind of another one of those orthogonal changes to this effort in the sense that you know we could improve this. But I would feel, I would feel strongly that such an improvement would be done in a separate PR from the one that migrates to get lab for Jay. So I didn't really do it in the this PR which I made regarding the first milestone of migrating the resource because I'll be, I'll be making it another PR for this. So, and that other PR we could, we could conceivably merge and release so that we fix it for get lab 16 users already right that doesn't, I think I have I understood you correctly. Okay, yeah, yeah. So limiters are you talking about pagination of the rest API. I mean like did they, they had a page number to it or I'm not sure what you're referring to. Like they added a dash in between just so that they could separate the projects and stuff projects. That's what that's the only thing that they did that I found out just to improve the performance of the get lab server like I don't know how they are doing it but yeah that's what they did. Well that's yeah that seems like a pretty easy thing for us to adjust. Right. Well, and it seems like we will probably need to somehow do it conditionally based on the version of get lab that they're using because the old one doesn't respond that way 15 and earlier, still honor the old URL is that right. There's a lot of research on that like if I'm trying to like if I'm trying to click on the old URL it's automatically directing to the newer URL. So I'll have to research a bit on that I didn't really work on that I was more on the PR. Yeah, no problem and you did the right thing that that was the PR is certainly the more important thing. If we needed to I suspect others could do this change if you're, if you're feeling like hey you you don't have the time to work on this one. Yeah, it's a pretty easy thing like it's not good. Okay. Anything else on the get lab 16 breaking changes then. Nope. Okay next topic then pending pull requests that need discussion so I see three but I may have them in the wrong order for what you would prefer for discussion. Is there which which of these would you like to start first. Ignore the draft PR that I made, like the main PR the normal PR which I did. I think we need to review this. I also added a Docker based functional test with like I adapted them also, but it was a stress goal but I just forgot about it and just added it anyways. So, like, I am waiting for a review for this. Okay, and so this 1501. This is the one that's not draft and you're ready for you feel like you're ready for review. Yeah. Good. All right. Okay. So the tests are now compiling with this version right. Yeah, I adapted the test also so they are working at least on my side. And Chris also confirmed that they are working. Yeah, they are compiling. I don't think we're running them yet in the CI build. Are you running them look I think we are running them. As far as I can see, I would be surprised if CI was running. Oh, it's running. Oh yeah, these are just the injected tests aren't they know right. The CI build is not running the get lab tests yet. So if they are working then, if they are working that we could re-enable them in this pull request. But if they aren't, then it's markets in the palm file. Oh, so I can't, I can't just do a replay. Okay, got it. No cheating. Okay. If they are, if they are not just compiling but running, then we could re-enable them in this PR by basically reverting the changes that I had originally made to disable them. And if they aren't work, if they're not working then it would be fine to re-enable them in a separate change. But I think we would consider that an important thing to re-enable them before this project branch is put back to the main branch. Yeah, so, so harsh, would you like to be the one who re-enables the tests in that in your pull request? Are you comfortable how to do that? I don't really know. Well, let's see. Let's, all you need to do is the way Basil, you can describe it. Yeah, all you need to do is run a diff between the project branch and the main branch. And the only, the only difference right now between the two is the code that I committed to disable compiling and running the tests. So that would just need to be reverted. Yeah, it's basically this one and a predecessor to it, right? And this one. Which, as Basil said, do it, do it with a diff between the master branch and your working branch. You'll see what the changes. Okay, I'll revert it and try and see if it's passing or not. Now don't be dismayed. If you could just at mention me or Basil when you do that, because I suspect you don't have permission to run, to execute a jank. Oh no, this is not a Jenkins file change. Is it Basil? No change. Okay. I didn't say anything harsh, please ignore the last sentence I said it was nonsense. Okay, good. Yeah, and, and, you know, you would, you would want to run them locally before you try to do it in CI. And there's no, there's no pressure, you know, if they, if they aren't working locally yet. There's no pressure to do it in this PR, even just getting them to compile is a first step that's positive. So we could always, you know, do another PR after this one to make sure that they're actually executing, not just compiling. So, you know, if we're already there, that's great. And if we're not there that we'll get there, you know, and we'll get there soon. And I'll confirm with this, like we were working on the test cases I shared with him in the draft PR and he said that it was working. So I think he will review it first before I actually enable the test. Great. So compare with the master branch and undo the differences in the in the palm file. Okay. So, so have you with 1501 have you, I remember last week we were running into some problems connecting to the GitLab server, you were getting the TCP time out. So have you been able to work through that since we last talked. Yeah, it was working for me. At least it was I tried, I also tested it interactively, like not that heavily but at least the basic features I tested it and it was working for me fine. Okay, great. Maybe I can share the HPI file with you guys so that you can also test it like I didn't really test it that heavily because I wanted to test things very heavily after the webhook thing was implemented but still you can do it that can share the HPI Yeah, that sounds exactly right because we for this initial pull request. Our goal is to get things, you know, off the ground, not to make it perfect. So, you know, that that sounds exactly like exactly what we were looking for. Now, given that you're already already building. This is the pull request right harsh. This is 1501 and here's the HPI file that generates so we've actually already got the HPI file ourselves so you don't need to share it with us we've got it available and can put it into our own our own use I can put it into my, my development environment very easily. What are you using for the GitLab server are you are using the official Docker image or some other way to install GitLab. I use the Docker image also and also use the GitLab.com for some minor testing, like mostly the Docker image that's slow little thing that Sorry, what was the second thing you mentioned. That's not the Docker image. Like what I what I used to do GitLab.com like the server that that they are using. I also use that for version for testing and GitLab, the server for version three testing like my Docker instance. Okay, great. I'll try to review this today or Monday but I think this looks, this looks very promising so. Yeah, and I'll, I'll, I'll try to do some interactive testing with it as well and look at it. Buzzle's reviews are much more powerful but I'm happy to be a user and happy to do good, good checks to see one more person exploring it is a good thing. Yeah. All right. Now, did you want any discussion on 1491. Or is this one. No, this one's draft you said. Yeah. Great. Yeah, so draft pull requests. They've really been doing what we wanted, which is give you a place to quickly share with us without worrying too much about what it is. Very good. Okay. Good. Any other items you want to discuss on those pull requests harsh. 1497 is the is the fix for the Docker integration tests. Yeah, I was on the test. So I think I remember these because I think I tried to run them when I inherited this plug in two years ago and I don't think they ever worked or at least as far as I can tell the last time that they did work was more than three years ago. So that would be that would be really nice to get to get that working again. Like it's really give us a lot more confidence for us. I'm sorry. Like, it was a stretch goal for the project but I just forgot about it and I started working in working on it. And then in the between I just realized, like, after five or five or six tests were less that this is not what I want to do because I was constantly facing the client client configuration so I had to add the get lab rule for getting the actual get lab client from the Docker instance that I was using. So, I was into some mess and I just, it just pried me that what I was doing. Okay, so this is so this is not ready for review or would you like some comments on this one as well. No, I think it's ready like you okay. We look at the changes, can we look at the changes to the file that are in that draft. You bet. So this is. Yeah, so it's updating the Docker Maven plugin, which is just good, or actually, it looks like it's actually down. Yeah, it looks like it's actually downgrading it. And then, what about the compose file. It's downgrading in the draft PR it's not in the natural PR yet and that they actually are on the normal version. Right, so yeah it looks yeah and and it looks like we've got some changes to the compose file. Yeah I mean this is like this was for the like spot bus changes that Chris did. Yeah so. And did you get any of the, did you get any of the integration tests to pass when they were previously failing with this new configuration. We're not that many. Yeah Chris tried them and he told me that they were passing like I was getting some like job migrated issues that's why it was not passing on my machine but just said that it is passing on his machine. So I am counting on him, like but for me, I think some of them were failing like not more than four or five I guess. I think there's only four or five to begin with. And I think the last time I looked at this they were all failing. I will not feel that for sure. I'm talking about the, the docker based tests not the, the regular, the regular ones are fine in the main branch. It's only the docker ones that are not working. And I don't mean to doubt that you six that you ran them successfully but you should be doubting me. I was having pain in that. I would be very shocked if they passed without changes to the tests themselves, which it doesn't seem like you've done here but I could be wrong. Like, in the other changes that you made to the docker based Java tests themselves. The integration test you're talking about so mark you can open the test utility like he's talking about those things and the publisher classes. Well it's not the, it's not, it's not the test utilities it's the, it's the tests themselves. Yeah, I forget where they are but I mean, I'll find them integration test to be down. Yeah, I forget what they're called but it's like, oh, get lab it or something. Yeah, I get lab it that that's cool. Of course it's hard to search for that because get lab itself contains the string it. Yeah, it's there integration get lab it. I should see it on screen but I don't is that what there it is get lab it. Okay, the do we have to so do we have to change this class, other than to adapt it to the new API. No I just adapted it. Yeah, I would be pretty surprised if this test was passing without more deep changes than this, at least less than my, less than my try to run it it looks like it needed a lot of work. So have you run this particular one locally in a past, or you're not sure. Yeah, yeah. Yeah, so this this is, I think this is the one that need to work for a member, and I started, I started climbing this mountain one time, and I still have my, I still have my branch. And the first thing that I, the first thing that I tried to do was, if you, the integration test was using get lab version eight. So I was trying to, what, before I even started before I even started debugging it. I was trying to, to use a more recent version of get lab, and I had, I had made some changes to the palm file to update the Docker image to use to use a more recent version I was getting I remember getting a little farther with a more recent version but still having things to debug in that integration test. So, you know if I were, if I were to, if you weren't going to spend more time on this then I would start by trying to use a more recent version of get lab before stinking a lot of time to debugging the really old version. It's debugging the older versions, harder and more time consuming and ultimately unnecessary, because no one uses it so right when you know when I was looking at this I thought, I thought it would be more productive to start by upgrading the Docker image and then figure out how to get this test to pass. I succeeded in doing neither of those two things in the four hours that I spent doing this. So, I think it would probably take a lot more than four hours to do that. Like, I take my words back the tests are not working. Yeah, there should be some problem. Yeah, so anyway, that's, that's the way that I would suggest doing it. At least they're adapted. Yeah, getting them to compile is a positive thing. Absolutely. So, but yeah, they've been disabled at the running them has been disabled, as long as I can remember, just many years now. So, that's pretty much the definition of technical debt right there. Yes. Okay, so, so this one is likely not not ready the 1491 not ready unlikely to be passing needs more work, not now, right. Don't do that work now. It's no harm that you started at harsh, but we're perfectly fine not doing that work. Do I understand that correctly. Yeah, I think I should be partition side than the testing site like it. The project goal is to make the work of the big the plugin work interactively, not to add the Docker base functional test, it's a stretch goal right. Very good. Okay, good. All right. Anything else on pending or of the pull requests or other discussion topics around pending pull requests from my side I think that's what I did in this week like I was pretty slow but yeah, the good love and strength that I use is very, very true. Any other discussion we need to have on this one on review and results from their view or are you feeling comfortable harsh are there things that we need to do better for you what any any insights or things you want to share there. I think so I'm having any problem like I'm waiting for the review so that I can improve upon them. Like, it should be having some problem right. So, yeah, no problem for me. Okay and puzzle any counselor advice you want to give on on that topic. No, this is good. I got everything that I needed out of this conversation I think I'm ready to look more deeply at the code. Excellent. Thank you. So the next topic then for me was just a safety check of our checklist, and as oh good and this gives the states right so midterm evaluations begin submitting July 10. How does that collide with your calendar now harsh it is. Oh it's right in the middle of it okay so you'll need to be able you'll need to have that midterm evaluation ready well before that. So that you can submit it during your exam period but not spend a lot of time on it. Okay, good. All right now where did I put that page okay then July 14 that's when the evaluations are due, and we don't have a date yet or at least I haven't seen a date for the presentations okay. So those we've talked about at least those dates and the risk they are to be sure we're ready for them. Good. All right. That's all that I had on mentor checklist review. Next question was is this meeting time still okay for everybody harsh this isn't too late for you. Are you still okay with this time of day. Yeah. And Basel does this time still work okay for you. Great. All right. That's all the topics I had any other topics we need to discuss in our session today. Sorry say that again harsh. Oh, good. Okay, great. Then we are set. I'll stop the recording recording will be available in a