WEBVTT 00:00:00.000 --> 00:00:07.999 Welcome to No Compromises. A peek into the mind of two old web devs who have seen some things. This is Joel. 00:00:08.000 --> 00:00:09.999 And this is Aaron. 00:00:16.000 --> 00:00:23.999 Aaron, thanks for joining me here. I would like to discuss a recent PR in public and just have a chance to debate it a little bit more. 00:00:24.000 --> 00:00:32.999 Because I seem to recall in a code review, I think I was reviewing your pull request, and I didn't really care. Like I didn't think, "Oh, this is terrible." 00:00:33.000 --> 00:00:36.999 But I'm like, "Why did you do this?" And I made an offhand comment, and you're like, "Yeah, I disagree." I'm like, "You know what? 00:00:37.000 --> 00:00:39.999 That'll be a good podcast topic." So here we are. 00:00:40.000 --> 00:00:40.999 Okay. 00:00:41.000 --> 00:00:52.999 So the scenario to set up what I had observed in the PR that I was commenting on was that you had added a bunch of values to our phpunit.xml file. 00:00:53.000 --> 00:01:00.999 And these were keys that are also specified in the .env.example file, right? And so, it doesn't matter what they're for, 00:01:01.000 --> 00:01:02.999 but they're for talking to an external service. 00:01:03.000 --> 00:01:06.999 Yeah. Well, and before you go on too far, let's make sure when we say it doesn't matter what they're for. 00:01:07.000 --> 00:01:15.999 The .env locally is filled up of two different things, right? The .env file. It's filled up of workflow control and credentials. 00:01:16.000 --> 00:01:16.999 Okay. 00:01:17.000 --> 00:01:19.999 So workflow control, I mean, things like is this enabled or is it not enabled? 00:01:20.000 --> 00:01:20.999 Oh, sure. 00:01:21.000 --> 00:01:26.999 Are we in debug mode? What database do we even... Like, what connection do we pick? Not the credentials, but what connection? 00:01:27.000 --> 00:01:32.999 So it's all workflow. And then there's a second reason for it, which is credentials. Like API keys, passwords, things like that. 00:01:33.000 --> 00:01:33.999 Yeah. 00:01:34.000 --> 00:01:38.999 So, I just want to make sure that we bring those up because I think that might be the core of some of the things here too. 00:01:39.000 --> 00:01:45.999 Oh, okay. All right, I'm prepared to be wrong about this, but... So these were credentials, right? These were things that you would use 00:01:46.000 --> 00:01:58.999 to talk to another third-party or some other external system. And the ones you had added were actually things I had coded in a previous PR, 00:01:59.000 --> 00:02:07.999 and I did not add them to the phpunit.xml. But then I believe what happened is you were working on this code, your tests were passing, 00:02:08.000 --> 00:02:16.999 then you ran the whole suite, like a good developer, and some tests failed related to these keys that I had done in the previous PR. 00:02:17.000 --> 00:02:22.999 And you're like, "Ah, Joel, blah blah blah," and then you added this to the phpunit.xml, and then your tests passed, and you moved on. Is that? 00:02:23.000 --> 00:02:23.999 Yep. 00:02:24.000 --> 00:02:32.999 Okay. And my contention was like, "Aaron, whenever somebody adds something to their projects, .env example, 00:02:33.000 --> 00:02:38.999 it is your responsibility as a developer to copy that into your .env," right? Like that's part of- 00:02:39.000 --> 00:02:39.999 Absolutely. 00:02:40.000 --> 00:02:40.999 You agree with that? 00:02:41.000 --> 00:02:41.999 Yep, absolutely. 00:02:42.000 --> 00:02:45.999 Okay. Good, then I'm right. 00:02:46.000 --> 00:02:47.999 No, we didn't go all the way through. 00:02:48.000 --> 00:02:54.999 I'll keep going. All right, so defend yourself and explain it to me because I maybe see where you're going, but- 00:02:55.000 --> 00:02:56.999 "Okay. Defend yourself." 00:02:57.000 --> 00:03:04.999 Let me say this. If you had copied that .env example to your .env, your tests would have passed. Correct? 00:03:05.000 --> 00:03:13.999 Yeah, your tests in my suite would have passed if I copied the .env.example files or line over that I honestly forgot. Did not notice. 00:03:14.000 --> 00:03:14.999 Okay. 00:03:15.000 --> 00:03:15.999 Correct. 00:03:16.000 --> 00:03:16.999 All right, so- 00:03:17.000 --> 00:03:20.999 But now I'm going to take it over now here because I'm glad I didn't. 00:03:21.000 --> 00:03:26.999 I'm glad I didn't do that because that is the reason I still agree with what I did. And the reason is because- 00:03:27.000 --> 00:03:27.999 Of course you do. 00:03:28.000 --> 00:03:29.999 Well, yeah, because we talk about... remember it was credentials. 00:03:30.000 --> 00:03:30.999 Yes. 00:03:31.000 --> 00:03:36.999 So, when we work with Stripe, we put in something in our phpunit.xml. We're like, "Stripe key do not use." 00:03:37.000 --> 00:03:37.999 Right. 00:03:38.000 --> 00:03:44.999 So we know that there's a key in there and it hits. I don't remember what was in your .env.example, but maybe it was like, "Do not use," 00:03:45.000 --> 00:03:51.999 or maybe it was like some credentials that would have worked for yours, or whatever. Well, I don't know what they were, but even if I had copied them over, 00:03:52.000 --> 00:03:58.999 it would have probably hidden the fact that I wanted to at least zero out the API key credential or something like that. 00:03:59.000 --> 00:04:06.999 And if there wasn't an API key or credential, then at least the hostname. Because I want to make sure my tests never accidentally use my .env locally 00:04:07.000 --> 00:04:11.999 to load up my test development .environment and hit things third-party. 00:04:12.000 --> 00:04:12.999 Sure. 00:04:13.000 --> 00:04:19.999 So, while I should have copied over to .env.example, the fact that I didn't showed me that there was tests that would hit those. Whether or not they 00:04:20.000 --> 00:04:28.999 would actually call out I don't know, but they were dying because they were missing those keys. So I put them there to basically say that I wanted to stop this. 00:04:29.000 --> 00:04:35.999 Now, when I think about it, I think I did it wrong in another way. And I'll tell you how so that you can feel happy here. 00:04:36.000 --> 00:04:36.999 Thank you. 00:04:37.000 --> 00:04:38.999 There was like four or five keys. What I should have done- 00:04:39.000 --> 00:04:41.999 I'm looking at it, there are seven. 00:04:42.000 --> 00:04:48.999 Whatever. So, what I should have done is copy over from the .env.example from my local copy when I first noticed it. 00:04:49.000 --> 00:04:56.999 That's what I should have done. Then I should have added in one key to make sure that they could never hit the external world in the phpunit.xml 00:04:57.000 --> 00:05:06.999 All right. So, I'm going to push back a little bit. Because I agree in principle with everything you said, but I think it might not completely have 00:05:07.000 --> 00:05:15.999 been the way you're describing it. Because there was, let's call the service... Let's say Stripe because Stripe is something people know about. 00:05:16.000 --> 00:05:26.999 So, I had a key in there called Stripe Host that was some invalid URL blah blah blah. Stripe isn't really a great example because this- 00:05:27.000 --> 00:05:34.999 But it doesn't matter. Third party, like insurance company. We're hitting insurance company, and you had a URL for insurance company API, right? 00:05:35.000 --> 00:05:45.999 Yeah. Okay, so all of the keys in here, the ones that you added, that I did not choose to put in the phpunit.xml, were not credentials. 00:05:46.000 --> 00:05:52.999 They were like port numbers, and they were like configuration. Like, this is the location ID It's going to look up. 00:05:53.000 --> 00:05:53.999 Yeah, absolutely. 00:05:54.000 --> 00:05:55.999 But you still feel like that is something that- 00:05:56.000 --> 00:06:00.999 No, those are workflows. Those are configuration or workflow. That's what I'm saying. I would have copied over 00:06:01.000 --> 00:06:04.999 all seven from that example to my .env. 00:06:05.000 --> 00:06:05.999 Okay. 00:06:06.000 --> 00:06:14.999 Then I should put one .env in PHPunit.XML, finding that host and making sure was something crazy. Because remember, we might use this locally, 00:06:15.000 --> 00:06:23.999 and so locally the .env we might actually make a real host. So, if we don't put something in phpunit.XML to make it a fake host and one of our tests runs, 00:06:24.000 --> 00:06:25.999 it'll hit that real host. 00:06:26.000 --> 00:06:29.999 Okay. I did have the hostname in there already as fake, so I was right. 00:06:30.000 --> 00:06:36.999 Right. But what if I wanted to use it in real life locally in my- 00:06:37.000 --> 00:06:37.999 Okay. 00:06:38.000 --> 00:06:41.999 That's what I keep saying. It's like when I configure those properly, it would be working for my development environment, 00:06:42.000 --> 00:06:44.999 and I run my tests. And if a test was configured to hit that, it would hit that. 00:06:45.000 --> 00:06:53.999 Well, but just to clarify. I'm saying I did put the one entry in phpunit.xml that talks, that's the hostname. 00:06:54.000 --> 00:06:56.999 All the other things were configuration, the host name was already there. 00:06:57.000 --> 00:07:01.999 Oh, did you? Oh, I missed that then. You should have pushed back harder then. 00:07:02.000 --> 00:07:08.999 Okay. No, that's fair. But I think this... I'm going to take away from this. I don't care if anybody else is benefiting from this right now, 00:07:09.000 --> 00:07:16.999 but I never really thought about the distinction that you drew between configuration/workflow and credentials. 00:07:17.000 --> 00:07:23.999 And credentials would include things like the hostname. Like, the way you talk to the outside world is- 00:07:24.000 --> 00:07:24.999 Yeah. 00:07:25.000 --> 00:07:30.999 Yeah, okay. All right, I feel like we both had good points, but I'm going to say I won. 00:07:31.000 --> 00:07:32.999 I'm definitely right. Cool. 00:07:38.000 --> 00:07:43.999 You know, it's funny that like things go in sort of cycles, you know. You notice that with trends, 00:07:44.000 --> 00:07:47.999 clothes, design, interior design, houses, all that kind of stuff, right? 00:07:48.000 --> 00:07:48.999 Mm-hmm (affirmative). 00:07:49.000 --> 00:07:56.999 And I never really thought about the cycles of entertainment because it always just seemed like our entertainment sort of moved forward. 00:07:57.000 --> 00:08:03.999 But then I started thinking like, "No, it does a cycle too." And something hit me the other day that is just actually hilarious, and I don't know if anyone else 00:08:04.000 --> 00:08:13.999 finds it hilarious. But I'm just going to say the statement, and then you can agree or disagree. YouTube music is just basically MTV again. 00:08:14.000 --> 00:08:15.999 Okay. Yeah. 00:08:16.000 --> 00:08:24.999 Because all it is is... I've seen this now, people in a generation different than mine, younger, will play YouTube music either in their other 00:08:25.000 --> 00:08:33.999 screen or even like minimize. And it's a music video with the music. I'm like, you're basically watching MTV again. And by the way, don't get me started on 00:08:34.000 --> 00:08:39.999 just being a guy who is stuck in the past thinking about data bandwidth and all that kind of stuff. You just want the audio, so why are you sending the video? 00:08:40.000 --> 00:08:44.999 Ah. But anyway, yeah, just MTV again. It's interesting how those things circle. 00:08:49.000 --> 00:08:51.999 So, as we kind of learned in this podcast, you can do something right and 00:08:52.000 --> 00:08:59.999 wrong all at the same time. It's sort of opinion, but is it opinion? I don't really know, but a lot of that stuff will come out during a code review. 00:09:00.000 --> 00:09:09.999 Code review, that's something we could help your team with. If you'd like Aaron and I to review your code, head over to masteringlaravel.io/codereview.