WEBVTT 00:00:00.001 --> 00:00:06.920 Hello and welcome to Python Bytes, where we deliver Python news and headlines directly to your earbuds. 00:00:06.920 --> 00:00:11.800 This is episode 23, recorded April 25th, 2017. 00:00:11.800 --> 00:00:14.200 And I'm one of your hosts, Michael Kennedy. 00:00:14.200 --> 00:00:15.260 And I'm Brian Okken. 00:00:15.260 --> 00:00:18.380 And we're here to share a bunch of cool Python stuff with you. 00:00:18.380 --> 00:00:21.760 We've got six cool items queued up and ready to go. 00:00:21.760 --> 00:00:24.260 But before we get to that, I want to say thanks to Advanced Digital. 00:00:24.260 --> 00:00:26.560 They have an awesome Python job. 00:00:26.560 --> 00:00:28.980 You can check it out at python.advance.net. 00:00:29.200 --> 00:00:30.620 And we'll talk more about that later. 00:00:30.620 --> 00:00:32.240 Right now, I want to talk about the GIL. 00:00:32.240 --> 00:00:32.920 Brian, what do you think? 00:00:32.920 --> 00:00:34.580 I think it's great to talk about the GIL. 00:00:34.580 --> 00:00:36.880 And I'm really glad that... 00:00:36.880 --> 00:00:42.080 So this is an article called Grok the GIL, How to Write Fast and Thread Safe Python. 00:00:42.080 --> 00:00:51.640 And we talk about the GIL as the reason why we can't do parallel computing and programming just built in in Python. 00:00:51.640 --> 00:00:54.940 But, you know, I haven't really jumped into it a lot. 00:00:54.940 --> 00:00:58.860 And this article is from A. Jesse Giroud Davis. 00:00:59.120 --> 00:01:01.280 Who, by the way, is an excellent writer. 00:01:01.280 --> 00:01:04.740 If you want to have some examples of great writing, read his stuff. 00:01:04.740 --> 00:01:05.780 It's great. 00:01:05.780 --> 00:01:08.040 So he has this little... 00:01:08.040 --> 00:01:11.540 It's a very lightweight introduction to what the GIL is. 00:01:11.540 --> 00:01:12.900 And to talk... 00:01:12.900 --> 00:01:16.360 And I like the approach of not just the details of it. 00:01:16.360 --> 00:01:20.900 Because most of us aren't going to go in and start hacking the CPython core. 00:01:21.100 --> 00:01:26.500 But a little peek into the CPython core to see that it's a mutex inside an interpreter lock. 00:01:26.500 --> 00:01:30.340 The GIL is the global interpreter lock. 00:01:30.340 --> 00:01:33.340 I love how he pulls out little snippets from CPython. 00:01:33.340 --> 00:01:34.660 He's got a section. 00:01:34.660 --> 00:01:36.520 Behold, the global interpreter lock. 00:01:36.520 --> 00:01:38.100 And it just shows you, like, the C code. 00:01:38.100 --> 00:01:39.180 Yeah, it's just one line. 00:01:39.320 --> 00:01:39.760 Yeah, exactly. 00:01:39.760 --> 00:01:42.440 And you don't really need to know a lot of C to appreciate it. 00:01:42.440 --> 00:01:43.480 But there's just, like... 00:01:43.480 --> 00:01:45.780 There's enough to make it super concrete. 00:01:45.780 --> 00:01:50.180 You're like, this is actually the code that runs when you, like, call the socket. 00:01:50.180 --> 00:01:53.040 Like, and that's how the GIL gets released, for example, right? 00:01:53.160 --> 00:01:54.340 Yeah, talking about sockets. 00:01:54.340 --> 00:01:55.900 And it really... 00:01:55.900 --> 00:01:57.280 He really talks about that. 00:01:57.280 --> 00:01:59.520 That it's focused around... 00:01:59.520 --> 00:02:02.020 The lock is around I.O. 00:02:02.020 --> 00:02:03.700 Whenever you're waiting for I.O. 00:02:03.700 --> 00:02:05.560 And I think there's other places, too. 00:02:05.560 --> 00:02:11.520 But that's the main place where your code will pause and let some other thread run. 00:02:11.520 --> 00:02:13.040 I like the... 00:02:13.040 --> 00:02:16.280 He has a thing that says it's so simple that you can... 00:02:16.280 --> 00:02:19.400 The effect on threads is so simple you can write it on the back of your hand. 00:02:20.060 --> 00:02:24.120 One thread runs Python while in others sleep or await I.O. 00:02:24.120 --> 00:02:26.860 And he actually has a picture of his hand. 00:02:26.860 --> 00:02:27.780 I think it's his hand. 00:02:27.780 --> 00:02:29.400 Yeah, I was wondering if that's actually his hand. 00:02:29.400 --> 00:02:33.060 Yeah, and if he wrote it, that means he must be left-handed 00:02:33.060 --> 00:02:34.740 because it's written on the back of his right hand. 00:02:34.740 --> 00:02:35.980 Or he had somebody else write it. 00:02:35.980 --> 00:02:38.560 So I was always curious about this. 00:02:38.560 --> 00:02:39.440 What are the limitations? 00:02:39.440 --> 00:02:43.140 And how do you utilize it to have faster code? 00:02:43.140 --> 00:02:47.480 And the gist of it is if you've got some code that's waiting I.O., 00:02:47.480 --> 00:02:51.080 like maybe pulling off a whole bunch of different... 00:02:51.080 --> 00:02:53.940 Taking a bunch of connections or pulling off a bunch of URLs... 00:02:53.940 --> 00:02:55.180 Downloading a bunch of URLs. 00:02:55.180 --> 00:02:58.660 That's a great place to use multi-threading 00:02:58.660 --> 00:03:01.380 because the guild doesn't really get in your way. 00:03:01.380 --> 00:03:02.540 There's... 00:03:02.540 --> 00:03:05.080 In places where you really have multiple processing, 00:03:05.080 --> 00:03:08.260 where you really want your Python code to run at the same time, 00:03:08.260 --> 00:03:11.580 then you have to jump into multi-processing. 00:03:11.780 --> 00:03:13.860 And he actually gives an example of that. 00:03:13.860 --> 00:03:15.500 And it's not that bad either. 00:03:15.500 --> 00:03:19.380 So anyway, I liked the quick jump into it. 00:03:19.380 --> 00:03:23.220 And I think I'm going to be a better Python programmer for reading this. 00:03:23.220 --> 00:03:24.840 Yeah, this is really nice work. 00:03:24.840 --> 00:03:25.780 Good job, Jesse. 00:03:25.780 --> 00:03:26.760 He's a great writer. 00:03:26.760 --> 00:03:29.560 I actually had him on Talk Python on episode 69, I think, 00:03:29.560 --> 00:03:33.140 about design patterns for programmer blogs. 00:03:33.140 --> 00:03:35.360 And we did a whole session on blogging. 00:03:35.360 --> 00:03:35.800 It was great. 00:03:36.440 --> 00:03:40.920 And one of the things I like about this is he talks about cooperative multitasking, 00:03:40.920 --> 00:03:44.500 concurrency versus parallelism, preemptive multitasking, 00:03:44.500 --> 00:03:49.140 how sometimes you still need to actually lock your Python code, 00:03:49.140 --> 00:03:52.660 even though you might think of like, well, this is all straight Python. 00:03:52.660 --> 00:03:54.320 It's not going to get interrupted. 00:03:54.320 --> 00:03:59.580 But there's certain mechanisms that slightly vary between Python 2 and 3, 00:03:59.580 --> 00:04:02.640 where if you hang on to the guild too long, 00:04:02.640 --> 00:04:05.980 it will be potentially taken from you and given to another thread. 00:04:05.980 --> 00:04:10.460 And so that might still cause what would appear to be parallel race conditions. 00:04:10.460 --> 00:04:12.360 So that's also worth reading about. 00:04:12.360 --> 00:04:12.660 Yeah. 00:04:12.660 --> 00:04:14.600 And one of the things that surprised me is, 00:04:14.600 --> 00:04:16.840 and I do realize I don't really worry about that. 00:04:16.840 --> 00:04:18.560 I deal with multi-threading in C++. 00:04:18.560 --> 00:04:24.940 And with C++, you have to do it fine-grained locking of data structures, 00:04:24.940 --> 00:04:28.040 any data structure shared by multiple threads. 00:04:28.040 --> 00:04:33.320 But I was surprised how much you can share between threads in Python 00:04:33.320 --> 00:04:36.980 because the GIL won't interrupt a bytecode. 00:04:36.980 --> 00:04:43.120 And it'll only interrupt, yeah, between bytecodes, not in the C code. 00:04:43.120 --> 00:04:47.640 So the, like, things like sorting a list will happen atomically, 00:04:47.640 --> 00:04:50.020 and you won't be interrupted with that, 00:04:50.020 --> 00:04:51.880 which is, that surprised me. 00:04:51.880 --> 00:04:52.660 I didn't know that. 00:04:52.740 --> 00:04:56.480 It is where, ironically, incrementing a variable could be interrupted. 00:04:56.480 --> 00:04:56.800 Right. 00:04:56.800 --> 00:05:01.320 Because it ends up being, like, a two-step or read-modify-write operation. 00:05:01.320 --> 00:05:01.900 Yeah, exactly. 00:05:01.900 --> 00:05:02.460 Exactly. 00:05:02.460 --> 00:05:06.080 And Jesse uses the disk module to look inside, which is all very good. 00:05:06.080 --> 00:05:06.860 So that's a great article. 00:05:06.860 --> 00:05:09.820 I think that's probably the most substantive thing we're covering. 00:05:09.820 --> 00:05:13.540 Do you want to think about not so substantive but pretty cool? 00:05:13.540 --> 00:05:14.080 Yes. 00:05:14.180 --> 00:05:14.900 I've got one for you. 00:05:14.900 --> 00:05:18.940 Let's talk about the NBA as in National Basketball Association. 00:05:18.940 --> 00:05:20.740 The American basketball. 00:05:20.740 --> 00:05:25.180 So there was a pretty big deal on Twitter the other day. 00:05:25.180 --> 00:05:27.980 So Mark Cuban, he owns the Dallas Mavericks. 00:05:27.980 --> 00:05:31.840 And he's, I don't know if he comes from tech or not. 00:05:31.840 --> 00:05:33.060 I don't really think so. 00:05:33.060 --> 00:05:34.580 But he definitely was an entrepreneur. 00:05:34.580 --> 00:05:36.960 He's, you know, worth, like, he's a billionaire, basically. 00:05:37.500 --> 00:05:46.140 But as a billionaire, owner of a NBA team, he posted out a pretty interesting thing on Twitter saying, here's the new NBA. 00:05:46.140 --> 00:05:54.540 And it was a picture of him learning Python machine learning with, I think, iPython, an iPython notebook open. 00:05:54.540 --> 00:05:57.600 And he's like, I need to understand the Mavericks in the NBA. 00:05:57.600 --> 00:05:58.260 I'm on it. 00:05:58.260 --> 00:05:58.700 Wow. 00:05:58.700 --> 00:05:59.480 It's pretty cool, right? 00:05:59.480 --> 00:06:00.240 It is pretty cool. 00:06:00.460 --> 00:06:09.600 I don't know much about basketball or Mark Cuban or any of that, but it's neat that somebody that high up is wanting to learn Python and notebooks. 00:06:09.600 --> 00:06:11.420 That was basically the main takeaway. 00:06:11.420 --> 00:06:15.860 A bunch of people, like our friends over at Partially Derivative, invited him to be on the show. 00:06:15.860 --> 00:06:17.100 They're like, oh, we have to hear your story. 00:06:17.100 --> 00:06:18.880 He's like, no, no, I'm just getting started. 00:06:18.880 --> 00:06:20.380 They have a team at the Mavericks. 00:06:20.380 --> 00:06:27.280 I just want to understand what they're doing when they use machine learning to help make predictions and planning. 00:06:27.620 --> 00:06:32.820 And that's kind of cool to think of how machine learning is actually driving these professional sports teams as well. 00:06:32.820 --> 00:06:33.240 Yeah. 00:06:33.240 --> 00:06:34.600 Very interesting. 00:06:34.600 --> 00:06:35.100 Indeed. 00:06:35.100 --> 00:06:35.460 Indeed. 00:06:35.460 --> 00:06:36.660 All right. 00:06:36.660 --> 00:06:39.920 So next up, we have somebody winning an award. 00:06:39.920 --> 00:06:40.540 How cool is that? 00:06:40.540 --> 00:06:40.900 Yeah. 00:06:40.900 --> 00:06:42.060 Ian Kurdasko. 00:06:42.060 --> 00:06:49.280 He got a, or it was announced that he will get the 2017 Community Service Award from the Python Software Foundation. 00:06:49.280 --> 00:06:52.120 And I think that's pretty cool. 00:06:52.120 --> 00:06:56.320 It's apparently, I didn't know that he was doing, that a lot of the stuff that he did. 00:06:56.320 --> 00:06:58.140 I mean, I was familiar with Ian. 00:06:58.140 --> 00:07:01.100 He was on Test and Code episode 13. 00:07:01.100 --> 00:07:09.080 And we talked about Betamax library that he has for recording and playing back requests interactions. 00:07:09.620 --> 00:07:17.680 He's apparently been the election administrator for the PSF since 2015, volunteering all that time, of course. 00:07:17.680 --> 00:07:29.140 And he is active in mentoring new coders and supporting other Python developers with apparently really a focus on trying to be active in mentoring women in Python. 00:07:29.140 --> 00:07:31.720 And I think that's just pretty awesome. 00:07:31.720 --> 00:07:32.660 Yeah, that's really awesome. 00:07:32.820 --> 00:07:34.020 So congratulations, Ian. 00:07:34.020 --> 00:07:39.020 And his project that you talked about, like replaying requests, that's called Betamax? 00:07:39.020 --> 00:07:39.480 Yes. 00:07:39.480 --> 00:07:40.460 That's an awesome name. 00:07:40.460 --> 00:07:40.900 Yeah. 00:07:40.900 --> 00:07:41.220 Yeah. 00:07:41.300 --> 00:07:48.000 And it's, I guess, the idea, of course, of there's a VCR type library in some other languages. 00:07:48.000 --> 00:07:52.320 But he chose Betamax because, well, everybody knows Betamax was better. 00:07:52.320 --> 00:07:52.800 That's right. 00:07:52.800 --> 00:07:55.780 But yeah, you should listen to it. 00:07:55.780 --> 00:07:57.340 It's a pretty interesting tool. 00:07:57.340 --> 00:08:00.620 So that was one that the community asked me to do. 00:08:00.620 --> 00:08:07.440 There were community members of listeners of Test and Code that said, hey, could you go find Ian and talk to him about Betamax? 00:08:07.440 --> 00:08:08.300 That's awesome. 00:08:08.300 --> 00:08:13.000 We love to get those recommendations for all the shows, including some stuff that we're covering here today, right? 00:08:13.000 --> 00:08:14.020 Yeah, definitely. 00:08:14.020 --> 00:08:14.460 Definitely. 00:08:14.460 --> 00:08:25.740 So if you want to work with these kind of fun things, maybe you work at a company where you're doing Java and you dabble in Python or you don't really get to do all the cool things you'd like, 00:08:25.740 --> 00:08:30.580 Advanced Digital has a cool job offer for everyone out there who might want to make a change. 00:08:30.580 --> 00:08:33.760 I wish I was near Jersey City because this sounds fun. 00:08:33.760 --> 00:08:34.780 It does sound fun. 00:08:34.780 --> 00:08:39.860 So, right, they're in Jersey City just across the Hudson from Manhattan there. 00:08:39.860 --> 00:08:41.800 Small, agile environment. 00:08:41.800 --> 00:08:45.100 They're mostly a Python shop, but they play with other cool technologies. 00:08:45.100 --> 00:08:48.760 They fund you guys to go to conferences, professional development. 00:08:48.760 --> 00:08:55.200 And most importantly and coolest, I think, is they run one of the 10 largest news sites by traffic in the U.S. 00:08:55.200 --> 00:08:56.440 And they do it with Python. 00:08:56.440 --> 00:08:59.900 So if you want to be part of that team and you want to play with cool stuff like that, 00:09:00.020 --> 00:09:03.520 just visit python.advance.net and check it out. 00:09:03.520 --> 00:09:10.160 So there's a couple of things coming up, Brian, that have to do with Python versus legacy Python. 00:09:10.160 --> 00:09:19.020 Remember, Matthias from the IPython project, Matthias Boussournier, I'm sorry if I mess up your name, but I think that's pretty close. 00:09:19.020 --> 00:09:25.520 He was the original guy who got us talking Python versus legacy Python instead of Python 3 versus Python 2. 00:09:25.680 --> 00:09:26.520 Oh, yeah, right. 00:09:26.520 --> 00:09:26.680 Yeah. 00:09:26.680 --> 00:09:29.740 So he works on IPython and Jupyter and all that stuff. 00:09:29.740 --> 00:09:32.760 And he's back with a new blog post, which is my next item. 00:09:32.760 --> 00:09:35.080 And it's a pretty big deal. 00:09:35.080 --> 00:09:39.260 We just talked about Mark Cuban, the new NBA machine learning, IPython. 00:09:39.760 --> 00:09:43.200 And so they just released IPython 6. 00:09:43.200 --> 00:09:44.340 Okay. 00:09:44.340 --> 00:09:45.840 So that's pretty big news. 00:09:45.840 --> 00:09:46.740 That is big news. 00:09:46.740 --> 00:09:47.120 Yeah. 00:09:47.120 --> 00:09:50.620 And so people who use IPython, you know, there's a brand new version. 00:09:50.620 --> 00:09:51.260 That's awesome. 00:09:51.700 --> 00:09:57.440 The bigger thing is that this is the first release where IPython goes Python 3 only. 00:09:57.440 --> 00:09:59.620 They've dropped support for Python 2. 00:09:59.620 --> 00:10:00.220 That's great. 00:10:00.220 --> 00:10:05.620 So, or as Matthias would say, they now support Python and not mixing in legacy Python with it. 00:10:05.620 --> 00:10:05.940 Yeah. 00:10:05.940 --> 00:10:08.600 And what I thought was nice is, you know, it's a pretty major project. 00:10:08.600 --> 00:10:15.420 They did a little write-up of what was their experience of converting a mixed source code to Python 3 only. 00:10:15.420 --> 00:10:17.060 What were the benefits and what were the drawbacks? 00:10:17.740 --> 00:10:21.320 So let's see, a couple of things, a couple of stats that Matthias put out. 00:10:21.320 --> 00:10:26.480 The size of the IPython code base has decreased by 1,500 lines. 00:10:26.480 --> 00:10:28.620 That's pretty solid, right? 00:10:28.620 --> 00:10:29.340 That's significant. 00:10:29.340 --> 00:10:30.640 Less code means less maintenance. 00:10:30.640 --> 00:10:31.020 Right. 00:10:31.020 --> 00:10:35.840 They said it's not just because they're dropping Python 2, but it's a significant amount is. 00:10:35.840 --> 00:10:41.760 And even more impressive is they added some entirely new features that required hundreds of new lines of code. 00:10:42.340 --> 00:10:51.960 So they're really the decrease in amount of code they had to support for Python 3 or really for 2 they were able to get rid of when they went to Python 3 is actually probably more. 00:10:51.960 --> 00:10:54.220 So that's pretty cool. 00:10:54.220 --> 00:11:01.140 And they said one of the benefits, they think, is that contributors can spend less time worrying about, well, how does this work if we do it in Python 2? 00:11:01.800 --> 00:11:03.720 Or, you know, this has happened to me. 00:11:03.720 --> 00:11:04.700 You make a pull request. 00:11:04.700 --> 00:11:05.440 You submit it. 00:11:05.440 --> 00:11:07.640 It runs on the continuous integration. 00:11:07.640 --> 00:11:09.360 And it works fine in Python 3. 00:11:09.360 --> 00:11:14.200 But then it fails in Python 2 because, you know, you forgot the B in the string or whatever, right? 00:11:14.200 --> 00:11:16.100 So they don't have to worry about that. 00:11:16.100 --> 00:11:17.820 CI runs faster. 00:11:18.820 --> 00:11:21.600 They said basically, in summary, we're totally happy. 00:11:21.600 --> 00:11:26.420 We're entirely pleased with having switched to basically have the ability to write Python 3 only code. 00:11:26.420 --> 00:11:33.880 And they're looking forward to using a lot of the improvements in Python 3, specifically async and await, which will be cool. 00:11:33.880 --> 00:11:37.240 So an async and await REPL inside of IPython. 00:11:37.240 --> 00:11:37.900 How cool is that? 00:11:37.900 --> 00:11:38.620 That's neat. 00:11:38.620 --> 00:11:42.680 Is async and await in all of the three versions or did that get introduced? 00:11:42.680 --> 00:11:43.940 It came in 3.5. 00:11:43.940 --> 00:11:44.540 Okay. 00:11:44.540 --> 00:11:47.940 The async I.O. stuff was introduced in Python 3.4, I think. 00:11:47.940 --> 00:11:52.080 And then 3.5, they're like, let's put some proper syntax on this and make it really easy. 00:11:52.080 --> 00:11:52.380 Yeah. 00:11:52.380 --> 00:12:02.760 I'm trying to, I'm writing a little thing that I want to have available on Python 2 also and at least 2.7. 00:12:02.760 --> 00:12:10.360 And even if I were to just do Python 3, all of the three versions, I still can't use f-strings, which I wish I could use f-strings. 00:12:10.360 --> 00:12:11.080 I know. 00:12:11.080 --> 00:12:11.660 They're so new. 00:12:11.660 --> 00:12:12.480 It's 3.6 only. 00:12:12.880 --> 00:12:14.740 Even on my production server, it's 3.5. 00:12:14.740 --> 00:12:16.280 So it is what it is. 00:12:16.280 --> 00:12:17.500 That's a move in the right direction. 00:12:17.500 --> 00:12:22.820 And I think it's great that Matias and others talked about their experience with that change. 00:12:22.820 --> 00:12:23.220 Yeah. 00:12:23.220 --> 00:12:23.860 That's awesome. 00:12:23.860 --> 00:12:24.200 Yeah. 00:12:24.200 --> 00:12:24.680 Thanks, Matias. 00:12:24.680 --> 00:12:25.140 Excellent. 00:12:25.140 --> 00:12:28.340 I think I'm, to use an American expression, beating a dead horse. 00:12:28.340 --> 00:12:29.980 But we have another. 00:12:29.980 --> 00:12:31.240 Is that dead horse called Source? 00:12:31.240 --> 00:12:31.780 SRC? 00:12:31.780 --> 00:12:32.460 Yes. 00:12:32.460 --> 00:12:33.180 Yes. 00:12:33.580 --> 00:12:38.100 The other package I was talking about me building up, it's for the book. 00:12:38.100 --> 00:12:47.020 But it's, I wanted to make sure I was representing the community correctly and how to, how to put together a distributable package and do it correctly. 00:12:47.020 --> 00:12:48.960 At least with best practice. 00:12:48.960 --> 00:12:50.240 I know there's not really a correct. 00:12:50.560 --> 00:12:57.180 But somebody pointed me to the direction of an article by, I'm going to probably get this wrong. 00:12:57.180 --> 00:12:58.040 I think it's Enoch. 00:12:58.040 --> 00:13:00.640 And it's called Testing and Packaging. 00:13:00.900 --> 00:13:05.860 And it's basically, he's the guy that did the Adders project or ATTRS. 00:13:05.860 --> 00:13:07.080 Great project. 00:13:07.080 --> 00:13:08.500 That we've talked about a couple times. 00:13:08.500 --> 00:13:26.920 And how there were issues, at least with one package that wasn't using the source, SRC, that the testing that was done was, there was a bug that showed up in, with installed applications that doesn't show up in non-installed. 00:13:27.460 --> 00:13:37.040 So one of the benefits of using SRC is you can more easily make sure that you're only testing the installed package and not the non-installed. 00:13:37.040 --> 00:13:41.260 And he also just shows that it's really just two lines of code change. 00:13:41.260 --> 00:13:44.680 So to do the right thing is not that much work. 00:13:44.680 --> 00:13:44.960 Right. 00:13:44.960 --> 00:13:51.580 So basically in your setup, PY, the call to setup, you set the packages to be looking in the source directory. 00:13:51.580 --> 00:13:54.220 And you set the package to be in the source directory, right? 00:13:54.300 --> 00:13:54.440 Yeah. 00:13:54.440 --> 00:14:02.500 So when you would normally say find packages, he recommends specifically saying find packages and then give it a where equals SRC. 00:14:02.500 --> 00:14:07.100 But you can also just put SRC in the, as the first argument. 00:14:07.100 --> 00:14:07.860 And that works also. 00:14:07.860 --> 00:14:10.500 And then listing it in the packages dir. 00:14:10.900 --> 00:14:18.080 And then one of the things I noticed, which I don't think people have really talked about, is the entire repository looks better. 00:14:18.080 --> 00:14:24.960 You've got all of the package junk, like your setup and your manifest and all that stuff at the top level. 00:14:24.960 --> 00:14:30.080 And the stuff you really care about on a day-to-day basis is separated into subdirectories. 00:14:30.080 --> 00:14:34.180 You've got the docs in one and the tests in another, and then your source in another. 00:14:34.660 --> 00:14:37.680 And that separation just, it pleases my organization. 00:14:37.680 --> 00:14:39.060 It just is nice. 00:14:39.060 --> 00:14:39.280 Yeah. 00:14:39.280 --> 00:14:40.620 I'm coming around to this as well. 00:14:40.620 --> 00:14:41.480 It sounds pretty solid. 00:14:41.480 --> 00:14:41.860 Anyway. 00:14:41.860 --> 00:14:46.500 But that's, I'll probably try to drop talking about that every episode. 00:14:46.500 --> 00:14:47.560 But there you go. 00:14:47.560 --> 00:14:48.240 One more article. 00:14:48.240 --> 00:14:52.080 Well, I'm not quite done beating the Python versus legacy Python horse yet. 00:14:52.080 --> 00:14:53.780 So I'm going to keep going on that one. 00:14:53.780 --> 00:14:55.120 Because there's some more big news. 00:14:55.200 --> 00:14:57.520 We've heard that IPython went to Python 3 only. 00:14:57.520 --> 00:15:05.860 And now, same week, last week, AWS Lambda goes to Python, or ads not only, but ads. 00:15:05.860 --> 00:15:08.000 Python 3.6 support was just 2.7. 00:15:08.000 --> 00:15:09.380 So that's a big jump, right? 00:15:09.380 --> 00:15:10.280 Wow, that's a big jump. 00:15:10.280 --> 00:15:10.500 Yeah. 00:15:10.500 --> 00:15:10.760 Yeah. 00:15:10.760 --> 00:15:12.400 So that's pretty awesome. 00:15:12.400 --> 00:15:14.660 And do you have much experience with Lambda? 00:15:14.660 --> 00:15:15.280 Have you played with it? 00:15:15.280 --> 00:15:15.680 No. 00:15:15.680 --> 00:15:18.200 I've heard a lot about it, but I haven't played with it yet. 00:15:18.200 --> 00:15:27.360 So Lambda is one of these things from AWS, from Amazon, that fits into this serverless architecture. 00:15:27.360 --> 00:15:29.940 So basically, you say, here's a function. 00:15:29.940 --> 00:15:33.600 And when something happens, run this, please. 00:15:33.600 --> 00:15:34.800 So run it on a schedule. 00:15:34.800 --> 00:15:36.580 Somebody changes a database. 00:15:36.580 --> 00:15:39.180 Somebody uploads a file to S3, whatever. 00:15:39.180 --> 00:15:40.760 And it just runs. 00:15:40.760 --> 00:15:43.000 There's no servers that you deal with. 00:15:43.000 --> 00:15:43.940 Obviously, there are servers. 00:15:43.940 --> 00:15:48.040 But it just distributes your code to run when it needs to. 00:15:48.040 --> 00:15:49.540 So I'll cross a whole bunch of servers. 00:15:49.540 --> 00:15:51.780 So it scales basically infinitely. 00:15:51.780 --> 00:15:55.500 As long as you have infinite money, you can infinitely scale this. 00:15:55.500 --> 00:15:56.180 It's fine, right? 00:15:56.180 --> 00:15:58.400 And that's pretty cool. 00:15:58.400 --> 00:15:58.760 Yeah. 00:15:58.760 --> 00:16:00.120 So have you tried it? 00:16:00.120 --> 00:16:03.560 No, I have not had a real reason to do it. 00:16:03.560 --> 00:16:06.200 I mean, I guess there's a couple of things that I could do. 00:16:06.200 --> 00:16:14.140 Like on the websites, there's a job that runs every couple hours that will completely re-index the database 00:16:14.140 --> 00:16:17.220 and reorganize it for super fast queries. 00:16:17.220 --> 00:16:20.660 Like the queries on the various websites run. 00:16:20.660 --> 00:16:21.960 And I'm going to be adding to Python bytes. 00:16:21.960 --> 00:16:22.500 No worries. 00:16:22.500 --> 00:16:25.620 They run like sub-millisecond, right? 00:16:25.620 --> 00:16:28.080 In order to get that stuff, you've got to kind of pre-compute some things. 00:16:28.080 --> 00:16:31.200 Maybe that's a perfect Lambda operation, right? 00:16:31.200 --> 00:16:37.680 Well, I mean, especially now that they have 3.6 support, I'm intrigued enough that I might give it a shot anyway. 00:16:37.680 --> 00:16:41.540 Just to make up some excuse to play with it. 00:16:41.860 --> 00:16:42.220 Exactly. 00:16:42.220 --> 00:16:43.040 We need to run this. 00:16:43.040 --> 00:16:49.320 But I mean, if you're using other AWS stuff, like their database services, Dynamo or RDS or S3, 00:16:49.540 --> 00:16:54.820 like here's a way to run code like really near your resources on triggers with no effort. 00:16:54.820 --> 00:16:58.440 And one of the things I thought was pretty cool, like this announcement just came out. 00:16:59.120 --> 00:17:07.160 And Zappa, so Zappa, if you look at their page, which I linked, it's called serverless Python web services. 00:17:07.160 --> 00:17:09.100 That's interesting, right? 00:17:09.660 --> 00:17:17.600 So basically you can set it up so that using the AWS architecture, you can route web requests to these Lambda functions. 00:17:17.600 --> 00:17:20.520 But you don't really have servers or anything like that. 00:17:20.520 --> 00:17:25.300 And people have been asking for Python 3 support. 00:17:25.300 --> 00:17:26.740 They've been saying, no, no, no, no. 00:17:26.740 --> 00:17:29.800 As soon as this dropped, they're like, yes, it has Python 3 support. 00:17:29.800 --> 00:17:31.720 So that is pretty cool as well. 00:17:31.720 --> 00:17:38.040 So you've seen things that basically are layered on top of Lambda also started to support Python 3, which is great. 00:17:38.040 --> 00:17:38.760 Yeah, definitely. 00:17:38.760 --> 00:17:39.460 Cool. 00:17:39.680 --> 00:17:40.900 All right, yeah, maybe we should play with Lambda. 00:17:40.900 --> 00:17:41.440 I don't know. 00:17:41.440 --> 00:17:41.960 Yeah. 00:17:41.960 --> 00:17:43.060 Very nice. 00:17:43.060 --> 00:17:44.340 All right, well, that's it for the news, Brian. 00:17:44.340 --> 00:17:46.680 You got anything personally you want to share with everyone? 00:17:46.680 --> 00:17:57.860 No, I'm going to be, I guess I'll be in the Munich area the second week in May if there's anybody around that wants to have a beer or something with me. 00:17:57.860 --> 00:17:58.440 Hit me up. 00:17:58.440 --> 00:17:59.500 Yeah, that sounds awesome. 00:17:59.500 --> 00:18:00.420 I'm jealous. 00:18:00.420 --> 00:18:01.500 I'd love to go visit Germany. 00:18:01.500 --> 00:18:03.660 Well, I'll do that at the end of the summer maybe. 00:18:03.660 --> 00:18:04.260 We'll see. 00:18:04.260 --> 00:18:05.920 But no news for me. 00:18:05.920 --> 00:18:08.400 I just want to say thank you, everyone, for listening. 00:18:08.400 --> 00:18:08.940 Oh, you know what? 00:18:09.020 --> 00:18:09.780 Actually, one more thing. 00:18:09.780 --> 00:18:12.060 This is not personal news, but it falls right in here. 00:18:12.060 --> 00:18:15.540 I also saw CheckIO at CheckIO.org. 00:18:15.540 --> 00:18:19.360 These guys have a pretty cool gamification of learning Python. 00:18:19.360 --> 00:18:21.660 They also just went Python 3. 00:18:21.660 --> 00:18:22.300 Oh, cool. 00:18:22.500 --> 00:18:24.360 So just to keep on this. 00:18:24.360 --> 00:18:26.140 Hey, Python 3 is starting to really roll. 00:18:26.140 --> 00:18:27.980 I'd say it's really starting to roll this week. 00:18:27.980 --> 00:18:35.620 I use CheckIO for, hopefully I won't get in trouble for this, but I've gone through a bunch of this stuff and I use them for interview questions. 00:18:35.840 --> 00:18:37.180 Yeah, I think it's actually pretty good. 00:18:37.180 --> 00:18:42.720 What I really like is you can solve a puzzle and then you can look at other people's solutions. 00:18:42.720 --> 00:18:53.640 And I found after solving a bunch and looking at the solutions that I unknowingly have an implicit bias towards performance over ease of reading or simplicity or whatever. 00:18:53.640 --> 00:18:56.240 It's interesting that it uncovered that for me. 00:18:56.320 --> 00:18:57.060 Oh, that's interesting. 00:18:57.060 --> 00:18:58.180 And I totally have the opposite. 00:18:58.180 --> 00:19:01.280 I like them to be readable more than anything else. 00:19:01.280 --> 00:19:01.580 Yeah. 00:19:01.580 --> 00:19:01.960 Yeah. 00:19:01.960 --> 00:19:02.620 Funny, huh? 00:19:02.620 --> 00:19:03.600 All right. 00:19:03.600 --> 00:19:04.700 Well, thanks, Brian. 00:19:04.700 --> 00:19:05.920 Thank you, everyone, for listening. 00:19:05.920 --> 00:19:07.060 And we'll catch you next week. 00:19:07.060 --> 00:19:07.660 And thank you. 00:19:07.660 --> 00:19:08.060 Yep. 00:19:08.060 --> 00:19:08.500 You're welcome. 00:19:08.500 --> 00:19:08.960 Bye. 00:19:09.320 --> 00:19:11.840 Thank you for listening to Python Bytes. 00:19:11.840 --> 00:19:14.400 Follow the show on Twitter via at Python Bytes. 00:19:14.400 --> 00:19:17.280 That's Python Bytes as in B-Y-T-E-S. 00:19:17.280 --> 00:19:20.700 And get the full show notes at pythonbytes.fm. 00:19:20.700 --> 00:19:25.060 If you have a news item you want featured, just visit pythonbytes.fm and send it our way. 00:19:25.060 --> 00:19:27.760 We're always on the lookout for sharing something cool. 00:19:27.760 --> 00:19:31.140 On behalf of myself and Brian Okken, this is Michael Kennedy. 00:19:31.140 --> 00:19:34.780 Thank you for listening and sharing this podcast with your friends and colleagues.