Magento GraphQL Community, 18 Oct 2018

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Magento GraphQL Community Meeting, 18 October 2018

Description

Agenda:
-) General updates about Pull Requests (@vnayda)
-) Add GraphQL tests support to Travis (@Yaroslav Rogoza)

GraphQL Community Project Wiki: https://github.com/magento/graphql-ce/wiki
Mind Map: https://www.mindmeister.com/1094880388?t=a5LVAQ71Vq
Backlog (with Zenhub): https://github.com/magento/graphql-ce#boards?repos=128075669&showPRs=false
Backlog (without Zenhub): https://github.com/magento/graphql-ce/issues

Zenhub Chrome Add-on: https://chrome.google.com/webstore/detail/zenhub-for-github/ogcgkffhplmphkaahpmffcafajaocjbd

A

Soho guys, we start our weekly group all demo. Today we have pretty short agenda because the last week I have was at Romania and I. Just come back, so we don't have a big changes in processing requests and general update from my side, a sheer stream.

A

Like during contribution date, Romania, we finish it took requests. I will show.

A

This poor request was about my product reviews at my orders also looks like all change all fixes after pod review was fix it and we have fair word static bills, but I think it will be no big problem and I think this poor requests will be noted as quick as possible. Yes, we have only problem is static test and next update from my side. I am going to contribution day at :. It will be next week. It will be very interesting event.

A

There are two days of contribution and like and looks like: there are no free tickets, so I expected I, don't know 10, maybe 50, maybe more people's, because I asked Alexander zoom in he told that there are no free tickets and so I will be missed at four days next week. I will try to provide some answer as weak as possible, but maybe maybe I will be unreachable sometime and generally say week after the next field. Bill will be very hard because we need to process interview and not shove.

A

A lot of requests, the proquest from contribution to Romania and what a question contribution day from poem general is. It's also observe all update from my site and next I miss it Patrick I in agenda, but he he wrote me is that he would like to share some status from contribution day at Barcelona. So please are.

B

You connected to the to this or recognizes: can I do all right? Can you open issue 145 I'm not connected to the maybe goal area you could share your screen on exclude.

A

To share your screen.

C

B

I noticed you're, not I'm, not.

B

Okay issue 145, it's on the third column, so what's number 145.

B

Right so as a line I'm from Poland, she never contributed before to the graph QL, and this was her first item. I would expect that she would have done a poor request by now, but I guess she didn't. However, he has all the code and she did about all the co plus plus the test, and she has a functional functionality for all these things. I, don't know why I cannot say the pork was because I was almost sure she ate. She did a poor request for this I communicate.

A

With son today and I provide some advices and he promised it create two requests today, maybe tomorrow a sweet was possible. Yeah.

B

And she also said that she's gonna take another item which was fairly similar to this, so she seems that there's gonna be a great contributor to this. Also, there was another issue of being worked on, and that was issue number four and.

B

This was this was actually done by a by a guy. There was a very experienced developer and he finally abandoned it because he thought he was too complicated. Didn't know how to work with the media gallery and all that stuff, and it was not very performant when he looked at. How can he he put up like a coating a performing way? So there was a no poor request in this and he abandoned that in me term.

B

The contribution.

B

So there was just this to attempt the race was for PWI and stuff, because we had a shared table between PWA and grasscloth.

B

That's that's it for my site.

A

Thank you, Chris, and also we have a birthdate about running active functional tests on trails. You know that, generally we have on the upper functional test for Imperial. For us, it's absolutely important since, and this task was market, a special achievements and hundreds and Yaroslav will share results.

D

Hey guys sharing my screen, so I haven't been working on this issue for some time and finally, a few days ago, I had some success with running other functional tests with Travis. So if you check this will request.

D

The bills, so everything is green here, except the other functional tests, but the situation it's not as bad, because actually we have only a few tests with errors I'm going to check what happens, but I'm, not sure this issue is related to Travis, somehow, because sometimes on my local environment, I have the same errors with some tests, but I need to double check it, but, as you can see, almost all tests except three one run without issues, and actually this operation is being executed.

D

Pretty fast I believe it's about just a few minutes to run all other functional tests using Travis and after we figure out what happens with this three errors. I believe we can use this approach in our usual routine.

A

So I run only craft real tests on the trails yeah.

D

Yeah the PHP unit that XML file is configured only they're after functional tests that are related to graph QL I.

A

Am not true. Can we move this in develop branch Alex? What would we know to this request develop if there are will be run it on the graph co-opted? Because if you notice this to our man line, it will be affected over versions all of chicken.

A

Can we just keep those tests that don't work right now, I mean? Maybe we need to run all of our functional tests? What only graphical, but oh, we need yes for.

C

The tests all executed know just what I thought yeah well, the rest I believe others tests all these kids and the difference is that you use different pitch per unit for graph Keo and for other tests, at least that's what my understand might be here. No.

A

No, no, we don't help any of you functional test on time. Yes,.

D

That's that's the very first approach to implement a functional test in Travis and, as Bolero said, we have a separate PHP unit file that points PHP. You need to execute only krotkiy all related at the functional tests, but I'm not sure we have the same HP unit that XML file in my line. So.

A

If we would like to notice for this pull request to malign, maybe we need to run all of tests, because if you run on the graph QL test, I'm, not sure that we'll have some approve.

C

So we should have different jobs anyway, like we have on our Jenkins. We have separate job for crafting separate forests and separate for soaps. That's why it is possible to merge this separately, but then we can just use the same approach and most likely to be trivial to enable rest and soap support when we have graph key of support there and and also your stuff regarding these fellows, it is possible test. Module is not installed, as it should be before the tests execute, because I said.

B

It seems yeah yeah.

C

Because all these feathers are related to that module. If you know, if you're aware about the functionality, but we can discuss it after offline, if you wish yeah.

D

Yeah sure I saw this issue as well, but I'm not sure. Maybe we should just script a little bit to have this modules installed in Travis environment or we can skip this test somehow. So we can discuss a separate thing.

B

Is I, don't know how can I skip them just with Travis if.

C

You should not skip them. We should figure out like why they are not working, not specific contracts, because we have same scenario is currently working on Jenkins and they were working on. One boy, I believe as well, and everything was fine. So maybe there is some small misconfigurations. So after the meeting, okay, yeah sure.

D

And yeah, and that's actually all from the Travis perspective and also I, created another pull request regarding setting shipping methods for shopping, cart and I covered in this palookas two issues. Actually, it's managing shipping methods and I have introduced the concept of schema for working with shopping carts. It's for setting shipping addresses billing addresses and shipping matters to the shopping cart, but I am stuck here because the approach is working great but as we discussed before, we need to figure out how we implement the output of our shopping cart operations.

D

So we need to return the entire cart object on all our operations and it makes sense completely but I'm not sure about the best way, to create some Universal method for all our operations. To turn the entire cart, so my thought was to create a separate resolver for cart object. So, let's check the schema.

D

We have the cart object here, and maybe we should create a separate resolver for this cart and in this resolver we will be able to get all this information for all our operations. But it's like matter for discussion, because I haven't had a chance to discuss it with Valera or someone else, and that's the only roadblock for me to finalize this pool finalize this pool request. So maybe, if we have some time for a grooming, we can discuss this point.

A

We can't bow guru, Manhattan, Monday or Tuesday great, because I will left.

A

D

That would be great because I can finalize this pull request and we will have like a basis for our father mutations for a shopping cart, so the small refactoring and the idea how to introduce the cart object. It's the only thing that needs to be discussed.

A

Okay, so I will be in human teaches sometime.

A

Also, we have obtained from Alex. Isn't he finish at work at Port, West show on the active categories, but now he will in some trip, so he cannot show and declare it on.

A

Okay looks like maybe because we don't have any updates, maybe some more some questions. Yeah.

C

Sure I have one more update, I'm, currently working on unification of our tests and specifically looks like we can unify scenarios for functional tests and Web API tests. But all this is working progress right now and we'll have more details in coming weeks, I suppose, but at least just for you to know that we are working on improving our state with test duplication and also stability of the tests and coverage for equation looks like we should be able to have pretty good coverage.

C

Dennis Lee in the web guys and our UI yeah he'll, give more details later.

A

See you next week, I will try to join them off from contributions day. Maybe generally I would like that guys showed some functionality in soon, but I know trust. It will be possible if you have some separate stage. I will provide some information more clear to next demo. So thank you guys and see you next week.