WEBVTT 00:00:00.001 --> 00:00:05.260 Hello and welcome to Python Bytes, where we deliver Python news and headlines directly to your earbuds. 00:00:05.260 --> 00:00:09.800 This is episode 60, recorded January 10th, 2018. 00:00:09.800 --> 00:00:10.880 I'm Michael Kennedy. 00:00:10.880 --> 00:00:11.700 And I'm Brian Okken. 00:00:11.700 --> 00:00:14.060 And we've got a bunch of stuff lined up for you, as always. 00:00:14.060 --> 00:00:16.100 Before we get to it, I want to say thank you to Datadog. 00:00:16.100 --> 00:00:19.540 Datadog is sponsoring the show at pythonbytes.fm/datadog. 00:00:19.540 --> 00:00:21.580 More to say about that later. 00:00:21.580 --> 00:00:23.220 Brian, I want to know who's at nine. 00:00:23.220 --> 00:00:24.480 Who's at nine? 00:00:24.480 --> 00:00:25.980 Yes, who is at nine? 00:00:25.980 --> 00:00:32.300 Wouldn't it be awesome if you could actually just utter this question into the air and have it automatically answered for you? 00:00:32.300 --> 00:00:37.320 Yeah, well, that's the promise of Alexa and Google Home and all these home assistants, right? 00:00:37.320 --> 00:00:37.880 That's right. 00:00:37.880 --> 00:00:39.700 This came up, what was it? 00:00:39.700 --> 00:00:42.420 Episode 33 of Python Bytes, actually. 00:00:42.420 --> 00:00:45.380 We talked about you brought up building an Alexa skill. 00:00:45.380 --> 00:00:47.640 And then somebody saw that. 00:00:47.640 --> 00:00:49.560 And we have a link to a Twitter thread. 00:00:49.560 --> 00:00:53.300 Yeah, it was Greg Quinlan who posted that on Twitter. 00:00:53.300 --> 00:00:53.720 That's cool. 00:00:53.720 --> 00:00:55.540 Okay, well, he ran with it. 00:00:55.540 --> 00:01:00.300 And he looked into programming both skills for Alexa and Google Home. 00:01:00.300 --> 00:01:05.460 And he came up with this who's at nine video that we have a link to. 00:01:05.460 --> 00:01:07.720 And it's just, I think it's hilarious. 00:01:07.720 --> 00:01:15.600 It's basically the old who's on first routine, but done with an Alexa and a Google Home together. 00:01:15.600 --> 00:01:16.840 It's funny. 00:01:17.000 --> 00:01:21.140 Yeah, so he's got like a little dot and a Google Home. 00:01:21.140 --> 00:01:23.940 And he activates them both and they talk to each other, right? 00:01:23.940 --> 00:01:25.160 It cracked me up. 00:01:25.160 --> 00:01:25.680 It's good. 00:01:25.680 --> 00:01:29.020 We should have like had a clip for it or something for the show. 00:01:29.380 --> 00:01:30.700 Yeah, we should just play it. 00:01:30.700 --> 00:01:31.240 Yeah, that'd be fun. 00:01:31.240 --> 00:01:34.500 One of the things that strikes me, first of all, nice work, Greg. 00:01:34.500 --> 00:01:35.020 That's awesome. 00:01:35.020 --> 00:01:38.600 Flask Ask is really pretty interesting to work with. 00:01:38.600 --> 00:01:42.580 And the Flask Assistant is the Google Assistant equivalent of Flask Ask. 00:01:42.580 --> 00:01:44.200 It's like just ported to that API. 00:01:44.200 --> 00:01:49.880 But what's really interesting to me is the API here effectively is Flask. 00:01:49.960 --> 00:01:52.220 And I'm noticing this in lots and lots of places. 00:01:52.220 --> 00:01:56.580 Like someone just sent us something that was almost a web API for the command line. 00:01:56.580 --> 00:02:01.320 But I'm noticing Flask being like the de facto API for so many things these days. 00:02:01.320 --> 00:02:10.580 Yeah, it must just be that people are just having it's easier to set up a Flask, a small Flask application for something. 00:02:10.580 --> 00:02:11.120 I guess. 00:02:11.120 --> 00:02:14.640 I mean, pay attention to number four on our picks as well. 00:02:14.640 --> 00:02:15.360 We'll come back to this. 00:02:15.360 --> 00:02:20.440 There's something coming up related to the Who's at Nine that's on Talk Python, right? 00:02:20.440 --> 00:02:20.800 Yeah. 00:02:20.800 --> 00:02:29.400 So I have three guys, a panel discussion, who are all involved in building Echo Assistant type things. 00:02:29.400 --> 00:02:32.760 Some of the people behind Flask Ask and some of the people behind Flask Assistant. 00:02:32.760 --> 00:02:38.120 Those are sometimes the same people on Talk Python this week. 00:02:38.120 --> 00:02:42.400 So it should come out just about the same time this episode comes out, which is pretty awesome. 00:02:42.400 --> 00:02:47.300 So if you really like this and you want to dig into it, check out talkpython.fm/146. 00:02:47.300 --> 00:02:48.860 There should be a whole lot there. 00:02:48.860 --> 00:02:49.420 That's cool. 00:02:49.420 --> 00:02:49.680 Yeah. 00:02:49.680 --> 00:02:50.400 I'll check that out. 00:02:50.400 --> 00:02:50.840 Sounds good. 00:02:50.840 --> 00:02:51.720 Yeah. 00:02:51.720 --> 00:02:58.740 So I came across an article that I think it's pretty interesting because you and I have been 00:02:58.740 --> 00:03:01.760 singing the praises of Python, how it's really, really popular. 00:03:01.760 --> 00:03:06.640 It's becoming quite the important language in so many ways, right? 00:03:06.640 --> 00:03:09.000 We talked about the incredible growth of Python, things like that. 00:03:09.420 --> 00:03:15.520 But I guess this struck a chord with me because I feel like one of the places where Python really 00:03:15.520 --> 00:03:22.800 falls down is this sort of the intersection of packaging for delivering an application to 00:03:22.800 --> 00:03:29.600 non-developers or non-servers or packaging, and especially in that area around UIs, like 00:03:29.600 --> 00:03:30.580 native UIs. 00:03:30.580 --> 00:03:31.400 Yeah, definitely. 00:03:31.400 --> 00:03:32.980 We don't have that really. 00:03:33.360 --> 00:03:35.920 It's not as obvious what to do. 00:03:35.920 --> 00:03:36.740 It's not obvious. 00:03:36.740 --> 00:03:38.800 There's a few like half solutions. 00:03:38.800 --> 00:03:40.740 They sometimes work, except for when they don't. 00:03:40.740 --> 00:03:46.980 And, you know, it's the thing that I ran across that made me sort of bring this up again. 00:03:46.980 --> 00:03:50.220 And it's not that I think Python completely sucks because it doesn't have it. 00:03:50.220 --> 00:03:51.440 Obviously, I don't think that. 00:03:51.440 --> 00:03:53.760 But I see it as a major weakness. 00:03:53.760 --> 00:03:59.660 And if we could solve this challenge, I feel like it would just make Python even stronger 00:03:59.660 --> 00:04:05.180 and reach even deeper into like big enterprises that have to have desktop UIs and things like 00:04:05.180 --> 00:04:05.380 that. 00:04:05.380 --> 00:04:09.960 So the article I found is called Retiring Python as a Teaching Language. 00:04:10.460 --> 00:04:12.160 So this is by James Higg. 00:04:12.160 --> 00:04:17.040 And basically he said, for the last 10 years, my advice for someone getting started programming 00:04:17.040 --> 00:04:19.700 is to say, guys, start with Python. 00:04:19.700 --> 00:04:20.220 That's awesome. 00:04:20.220 --> 00:04:24.940 But then he wrote this article to say, actually, that's not my belief anymore. 00:04:24.940 --> 00:04:26.900 That's kind of sad, right? 00:04:26.900 --> 00:04:27.520 It is. 00:04:27.520 --> 00:04:33.020 And it seems, okay, so his big beef really is part of this not being able to do the user 00:04:33.020 --> 00:04:37.200 interfaces like desktop applications easily with it. 00:04:37.200 --> 00:04:37.860 Yeah, absolutely. 00:04:37.860 --> 00:04:38.900 And mobile as well. 00:04:38.900 --> 00:04:41.200 I mean, these kind of are sort of cousins. 00:04:41.200 --> 00:04:42.940 It says, it's all great. 00:04:42.940 --> 00:04:45.300 People work with Python when they're getting started and so on. 00:04:45.300 --> 00:04:50.020 But then one day the student will innocently ask, instead of running a poker simulator on 00:04:50.020 --> 00:04:53.260 the command line, can I just put it on a window with like a button and some cards? 00:04:53.260 --> 00:04:54.280 Yeah. 00:04:54.280 --> 00:04:55.120 Okay. 00:04:55.120 --> 00:04:56.340 So what do you do then, right? 00:04:56.340 --> 00:04:56.720 Yeah. 00:04:56.720 --> 00:04:59.080 Well, there's TK Enter, but I don't know. 00:04:59.080 --> 00:05:03.280 That looks like it's from the early 90s, late 80s. 00:05:03.280 --> 00:05:07.140 It definitely doesn't look fresh and exciting like modern applications do. 00:05:07.780 --> 00:05:11.000 We have Pygame, but that's only if you want a game, right? 00:05:11.000 --> 00:05:14.560 What if I want like an app that I want to... 00:05:14.560 --> 00:05:20.420 So suppose I work at like, say, a stock trading place and I want an app that is like across 00:05:20.420 --> 00:05:26.040 three monitors and it's showing me the status of all my trades running real time, maybe like 00:05:26.040 --> 00:05:31.540 a web socket back to some server as fast as it can be to like completely show the traders 00:05:31.540 --> 00:05:32.220 what's going on. 00:05:32.220 --> 00:05:34.460 Like, what's the answer for Python there? 00:05:34.460 --> 00:05:34.860 Yeah. 00:05:34.860 --> 00:05:38.380 Well, I don't want to diss Pygame too much because I know some people have done some 00:05:38.380 --> 00:05:42.440 non-game things with it, but I don't really have an answer to that either. 00:05:42.440 --> 00:05:43.580 I think that's a big shortcoming. 00:05:43.580 --> 00:05:47.580 And if we had a really good answers for that, like some of the other frameworks, like even 00:05:47.580 --> 00:05:53.060 JavaScript has really good answers for that, which is kind of bizarre, but like Visual Studio 00:05:53.060 --> 00:05:58.120 Code, Atom, those are all Node.js plus JavaScript front-end apps, right? 00:05:58.120 --> 00:05:58.900 That's Electron.js. 00:05:59.200 --> 00:06:03.960 I would like to see a modern sort of thing for Python. 00:06:03.960 --> 00:06:06.800 And I think this is just worth considering. 00:06:06.800 --> 00:06:11.880 You know, actually the Twitter thread that we'll link to the Twitter thread, it's pretty 00:06:11.880 --> 00:06:12.260 interesting. 00:06:12.260 --> 00:06:16.920 If you open up my initial sort of tweet, I don't know if you want to call it a complaint or 00:06:16.920 --> 00:06:20.620 warning to the community, but if you open that, you can scroll down and you can see all the 00:06:20.620 --> 00:06:21.100 conversations. 00:06:21.100 --> 00:06:25.300 And even Guido gets in there a little bit to talk about some of the history, which is pretty 00:06:25.300 --> 00:06:25.760 interesting. 00:06:25.760 --> 00:06:26.660 And thanks for that. 00:06:26.660 --> 00:06:30.560 So I think it's, you know, I just, I don't know that there's a great answer right now, 00:06:30.560 --> 00:06:32.480 but I'm trying to inspire people. 00:06:32.480 --> 00:06:35.000 I want to play devil's advocate here though, a little bit. 00:06:35.000 --> 00:06:40.160 I know that you've said that you have written a lot of desktop applications, but to be honest, 00:06:40.160 --> 00:06:40.860 I have not. 00:06:40.860 --> 00:06:46.380 I've written a couple little tiny things that I've needed for utilities at work. 00:06:46.380 --> 00:06:51.060 There's a whole bunch of programming that is not writing desktop applications. 00:06:51.060 --> 00:06:51.880 Oh, absolutely. 00:06:52.100 --> 00:06:57.220 And a lot of it services, a lot of it's on the web, or it's just command line utilities, 00:06:57.220 --> 00:06:57.760 like you said. 00:06:57.760 --> 00:07:03.500 But here's where I think the actual, this is why I think we're underestimating the importance 00:07:03.500 --> 00:07:04.100 of this. 00:07:04.100 --> 00:07:10.860 If you could open up your editor and it was just, you know, here's a few of the widgets. 00:07:10.860 --> 00:07:17.740 I could just drop them in here and I could just put a .app or a .exe somewhere. 00:07:18.060 --> 00:07:20.740 And my entire company could just pick it up and run it locally. 00:07:20.740 --> 00:07:27.480 How often have you not even can entertain that idea because you're just like, well, that doesn't 00:07:27.480 --> 00:07:27.740 exist. 00:07:27.740 --> 00:07:29.920 So why would I even consider this weird idea? 00:07:29.920 --> 00:07:30.160 Right? 00:07:30.160 --> 00:07:32.240 Like we'd probably also see a lot less. 00:07:32.380 --> 00:07:38.600 I mean, there's a lot of internal projects that are, they're set up as web applications, internal 00:07:38.600 --> 00:07:44.860 web applications that could definitely be desktop applications with a shared database backend 00:07:44.860 --> 00:07:48.780 that we don't do it that way because it's not easy to do that. 00:07:48.780 --> 00:07:49.200 Exactly. 00:07:49.200 --> 00:07:49.960 It's super hard. 00:07:49.960 --> 00:07:56.020 But it would be really nice to have some sort of ability to take advantage of the local 00:07:56.020 --> 00:07:57.880 application, a local OS. 00:07:57.880 --> 00:08:00.920 And maybe you talk to a web service for shared data. 00:08:00.920 --> 00:08:02.880 Maybe you just use SQLite locally. 00:08:02.880 --> 00:08:04.020 That might be interesting. 00:08:04.640 --> 00:08:11.640 But I don't think we do this because it's just like, it's so not working that it's just 00:08:11.640 --> 00:08:15.000 like, even if you got the UI working, it's like, okay, well now what do we do for packaging? 00:08:15.000 --> 00:08:15.440 Right? 00:08:15.440 --> 00:08:17.360 I know there are answers, but they're always kind of tricky. 00:08:17.360 --> 00:08:17.700 Right? 00:08:17.700 --> 00:08:23.500 So if that was like entirely obvious, the tooling was just like, yeah, of course you do this. 00:08:23.500 --> 00:08:28.120 Like how many stores in the, how many apps in the Mac app store would be Python apps? 00:08:28.120 --> 00:08:32.920 Right now I bet it's like 0%, you know, like basically, right? 00:08:32.920 --> 00:08:35.700 Once you round it, then round it down to the significant digits, right? 00:08:35.700 --> 00:08:36.900 They don't exist. 00:08:36.900 --> 00:08:39.480 And these things could just open up, right? 00:08:39.480 --> 00:08:44.480 If we could somehow as a community get like a really nice cross-platform UI set up. 00:08:44.480 --> 00:08:45.140 I agree. 00:08:45.140 --> 00:08:50.160 I think some of the stumbling blocks on that are that little thing that you throw in that's 00:08:50.160 --> 00:08:50.980 cross-platform. 00:08:50.980 --> 00:08:55.820 I know that that's cool, but I don't know if it's necessarily a requirement. 00:08:55.820 --> 00:08:56.500 That's true. 00:08:56.600 --> 00:09:03.460 Actually, I would say if you had a killer way to build macOS apps, stop. 00:09:03.460 --> 00:09:05.380 That would be a huge benefit for Python. 00:09:05.380 --> 00:09:10.500 If you had a killer way to build Windows apps, there's still like the majority of, you know, 00:09:10.500 --> 00:09:13.380 business people sit down and log in on a Windows machine. 00:09:13.380 --> 00:09:17.600 Like if you could do that, that would be an amazing advancement for Python. 00:09:17.600 --> 00:09:18.120 So you're right. 00:09:18.120 --> 00:09:19.300 It doesn't necessarily have to be. 00:09:19.300 --> 00:09:22.980 I mean, one model is what the people at Xamarin are doing. 00:09:23.160 --> 00:09:28.660 They basically have cross-platform stuff, but every platform has its own UI bit, but they 00:09:28.660 --> 00:09:30.480 have these kind of designer things that help with that. 00:09:30.480 --> 00:09:30.960 So I don't know. 00:09:30.960 --> 00:09:34.980 I just wanted to put that out there and let people know about this conversation and think 00:09:34.980 --> 00:09:35.760 about it. 00:09:35.760 --> 00:09:38.820 I think we're having an open session at PyCon on it. 00:09:38.820 --> 00:09:39.700 Oh, that would be good. 00:09:39.700 --> 00:09:40.000 Yeah. 00:09:40.000 --> 00:09:45.320 I think I'm glad we're bringing it up because we do need to hammer on this until it's fixed. 00:09:45.440 --> 00:09:49.440 If I look across all of Python, this is like really the place where there's still a weakness 00:09:49.440 --> 00:09:53.720 and it would be nice to just like fortify that wall and make it like a perfect place to be. 00:09:53.720 --> 00:09:54.080 Definitely. 00:09:54.080 --> 00:09:57.920 And like I said, SQLite as a little database, you don't even need a server. 00:09:57.920 --> 00:09:59.020 That might be a good choice, right? 00:09:59.020 --> 00:10:01.080 Yeah, definitely. 00:10:01.080 --> 00:10:07.920 So I came across, well, this is an older interview, but somebody recommended to us that we go back 00:10:07.920 --> 00:10:11.440 and listen to episode 201 of the Change Log. 00:10:11.940 --> 00:10:17.300 It's another podcast, but there's an interview with the, now I should have looked all these 00:10:17.300 --> 00:10:18.700 names up, but the dude that wrote? 00:10:18.700 --> 00:10:19.260 Richard Hipp. 00:10:19.260 --> 00:10:19.720 Yeah. 00:10:19.720 --> 00:10:20.080 Okay. 00:10:20.080 --> 00:10:21.880 You're awesome for looking that up. 00:10:21.880 --> 00:10:27.220 He started at, he calls it SQLite, which I've never heard it pronounced that way. 00:10:27.220 --> 00:10:31.620 SQLite, as in like a meteorite or something. 00:10:31.620 --> 00:10:32.020 Yeah. 00:10:32.020 --> 00:10:32.940 Or is that mineral? 00:10:32.940 --> 00:10:33.440 Yes. 00:10:33.440 --> 00:10:34.900 A very interesting way of pronunciation. 00:10:34.900 --> 00:10:35.960 None of which I would have guessed. 00:10:35.960 --> 00:10:36.320 Right. 00:10:36.320 --> 00:10:38.880 So the rest of us know this is SQLite, I think. 00:10:38.880 --> 00:10:39.620 That's how I know it. 00:10:39.620 --> 00:10:40.440 That's what I'm calling it. 00:10:40.440 --> 00:10:41.960 I didn't quite realize that. 00:10:41.960 --> 00:10:43.420 So this, this is the database. 00:10:43.420 --> 00:10:49.900 If you just say what SQLite three import in Python, you, you get a little single file database. 00:10:49.900 --> 00:10:52.160 This is beautiful because it's built into Python. 00:10:52.160 --> 00:10:57.780 Every instance, every buddy who has Python has this, so you can just count on it being there 00:10:57.780 --> 00:10:58.740 and it runs in process. 00:10:58.740 --> 00:11:00.160 So there's no configuration, right? 00:11:00.160 --> 00:11:01.000 Yeah, definitely. 00:11:01.240 --> 00:11:08.720 And I didn't realize how dang cool the thing is until I listened to this episode that this change log episode. 00:11:08.720 --> 00:11:17.900 One of the comments from the interviewers were, was they just assumed it was a starter database that you always eventually have to move over to something else. 00:11:18.180 --> 00:11:19.520 And that's not true. 00:11:19.520 --> 00:11:39.360 There's as long as your application is not like a web application with extreme client server concurrency, you probably can get away with like a lot of desktop applications are small or the like, if you have a client side part of your application, that could just be a SQLite database and it'll be fine. 00:11:39.500 --> 00:11:40.120 Yeah, that'd be awesome. 00:11:40.120 --> 00:11:40.320 Yeah. 00:11:40.320 --> 00:11:52.320 So if you had like some sort of client side thing that you periodically wanted to sync with a real database, you could store it all in, in SQLite locally and just synchronize that data across some kind of service with your real data. 00:11:52.320 --> 00:11:57.840 And that would be a real nice way to take the load off the server, to have a local offline version, all sorts of stuff. 00:11:57.840 --> 00:12:05.560 Some things I didn't know about SQLite before I listened to this was that it started its life as a TCL application. 00:12:05.560 --> 00:12:06.360 Yeah. 00:12:06.360 --> 00:12:07.840 Which is bizarre. 00:12:07.840 --> 00:12:09.440 It's very bizarre, but that's what happened. 00:12:09.440 --> 00:12:11.980 Well, I guess that's where like TK started, right? 00:12:11.980 --> 00:12:13.020 In the TCL world. 00:12:13.020 --> 00:12:20.340 But the, also that one of the things that you got to watch out for is that it's not as type safe, it's type flexible. 00:12:20.340 --> 00:12:22.040 Yeah, that's what he called it. 00:12:22.040 --> 00:12:27.540 When you are changing over to another database and your other database might be a little pickier than SQLite. 00:12:27.540 --> 00:12:28.620 So be careful of that. 00:12:28.620 --> 00:12:34.620 So an example, you might have a column, which is in the DDL designated as an integer. 00:12:34.900 --> 00:12:40.080 And if you passed, if you tried to assign quote seven to it, it would just convert that to seven. 00:12:40.080 --> 00:12:41.920 Things like that. 00:12:41.920 --> 00:12:43.600 So it's very funky in that regard. 00:12:43.600 --> 00:12:44.900 It's kind of scripty like that. 00:12:44.900 --> 00:12:47.740 He comes from a background of interpreted languages. 00:12:47.740 --> 00:12:50.940 So it makes, and, and weekly typed. 00:12:51.240 --> 00:12:52.400 So that makes sense. 00:12:52.400 --> 00:12:53.000 It's kind of cool. 00:12:53.000 --> 00:12:53.780 Yeah. 00:12:53.780 --> 00:12:55.980 I thought that, that interview was really interesting. 00:12:55.980 --> 00:12:58.960 Definitely a lot of respect for what he's doing. 00:12:58.960 --> 00:13:07.340 He's got some very different opinions and perspectives on open source, on say version control and editors. 00:13:07.340 --> 00:13:11.420 I mean, I'm not sure I agree with all of them, but they're certainly interesting to think about. 00:13:11.420 --> 00:13:12.340 A couple more things. 00:13:12.340 --> 00:13:16.200 The SQL, how there's a page of how SQLite is tested. 00:13:16.200 --> 00:13:21.940 And I think it's really cool that they put that out there of like what their checklist is for testing it. 00:13:21.940 --> 00:13:23.580 The amount of tests are huge, right? 00:13:23.580 --> 00:13:24.440 Yeah. 00:13:24.440 --> 00:13:28.380 And then also that the, I didn't realize it was in everything. 00:13:28.380 --> 00:13:32.840 Like every Android phone has a, is using it and things like that. 00:13:32.840 --> 00:13:34.040 I think iOS as well. 00:13:34.040 --> 00:13:34.240 Yeah. 00:13:34.240 --> 00:13:34.860 Yeah. 00:13:34.860 --> 00:13:36.440 So it's quite ubiquitous. 00:13:36.440 --> 00:13:38.340 It's definitely ubiquitous and it's very cool. 00:13:38.340 --> 00:13:41.380 And I think it's definitely a good place to start. 00:13:41.380 --> 00:13:52.180 And a lot of, a lot of times you might not need more than, than that, which is one fewer server to configure one fewer things to patch one fewer thing to make sure the firewalls are all right. 00:13:52.180 --> 00:13:54.060 It's really nice to just have that simple thing. 00:13:54.060 --> 00:13:54.700 Nice. 00:13:54.700 --> 00:13:57.940 So speaking of servers, let's talk about data dog real quick. 00:13:57.940 --> 00:14:06.420 So data dog is a monitoring solution that provides deep visibility inside of not just your app, but the infrastructure. 00:14:06.420 --> 00:14:07.500 Your app works with. 00:14:07.500 --> 00:14:14.560 So within a couple of minutes, you could investigate like say a bottleneck in your code by checking out some flame graphs and dashboards. 00:14:14.560 --> 00:14:16.360 You can visualize your Python performance. 00:14:17.360 --> 00:14:23.280 And if you go and do their free trial, you'll get a free data dog t-shirt, which has got a cool little dog on it. 00:14:23.280 --> 00:14:28.140 So check it out for yourself at pythonbytes.fm/data dog and let them know. 00:14:28.140 --> 00:14:29.480 Thanks for sponsoring the show. 00:14:29.480 --> 00:14:29.800 Yeah. 00:14:29.800 --> 00:14:30.240 Thank you. 00:14:30.780 --> 00:14:37.500 So I told you about Flask being this sort of like API that is standing in for all these other, other places. 00:14:37.500 --> 00:14:40.780 So the next thing I wanted to cover is serverless programming. 00:14:40.780 --> 00:14:44.160 So there was this joke, which I actually don't agree with. 00:14:44.240 --> 00:14:51.420 I kind of slightly disagree with it philosophically, but the sort of making fun of cloud computing saying, well, that's just another person's computer. 00:14:51.420 --> 00:14:53.280 There is no cloud. 00:14:53.280 --> 00:14:54.820 It's just another person's computer, right? 00:14:54.820 --> 00:14:55.440 That you're borrowing. 00:14:55.440 --> 00:14:56.400 Isn't that true? 00:14:56.400 --> 00:15:10.140 It's technically true, but my disagreement is like the cloud computing stuff really has to do with a hardware that you can basically create and control through programming. 00:15:10.140 --> 00:15:13.760 So if there's an API that you can say instantly, I want a machine. 00:15:13.760 --> 00:15:15.300 I want to scale my machine. 00:15:15.300 --> 00:15:16.700 I want to replicate my machine. 00:15:16.700 --> 00:15:20.940 That's not the same as just like co-location or like a rented server. 00:15:21.060 --> 00:15:24.980 Because you can't just go and like rent a server and then hit an API and have like another server. 00:15:24.980 --> 00:15:27.860 So anyway, that's my disagreement a little bit. 00:15:27.860 --> 00:15:29.380 But to some degree, that's true, right? 00:15:29.380 --> 00:15:39.860 Well, we've moved beyond that joke to a farther world of potential ridiculousness where we have server-based programs that have no servers, right? 00:15:39.860 --> 00:15:43.740 This whole AWS Lambda, Azure Functions, serverless programming. 00:15:43.740 --> 00:15:44.600 It's kind of funny, right? 00:15:44.600 --> 00:15:45.220 Yes. 00:15:45.220 --> 00:15:45.660 Yes. 00:15:46.140 --> 00:16:00.020 So basically, the idea with this serverless programming, whether it's AWS Lambda or Azure Functions, is you write a single function, preferably in Python, but often they support JavaScript, C#, things like that. 00:16:00.820 --> 00:16:06.200 And that single function is the entry point into some piece of functionality. 00:16:06.200 --> 00:16:11.020 And you're going to host that function on the cloud. 00:16:11.020 --> 00:16:18.120 And it's up to the infrastructure to figure out like what Docker container to create, run that on, make that happen, scale it, etc. 00:16:18.120 --> 00:16:22.320 You just say, here's literally a Python function and its dependencies run that. 00:16:22.320 --> 00:16:24.520 So that's pretty interesting. 00:16:24.800 --> 00:16:29.900 And the main reason here that people go for this, one, is there's no server at all to maintain, not even a virtual one. 00:16:29.900 --> 00:16:32.320 And the scalability is basically infinite. 00:16:32.320 --> 00:16:34.860 And the price is like insanely cheap. 00:16:34.860 --> 00:16:38.380 Because you only pay while your function's literally executing. 00:16:38.380 --> 00:16:39.040 Okay. 00:16:39.040 --> 00:16:40.780 But I mean, there's still servers. 00:16:40.780 --> 00:16:42.120 You just don't have to set it up. 00:16:42.120 --> 00:16:42.340 Yeah. 00:16:42.340 --> 00:16:44.260 And your code is not stuck to one of them, right? 00:16:44.260 --> 00:16:47.460 Like every function call basically, I think, is a new Docker container. 00:16:47.460 --> 00:16:49.220 I'm not entirely sure about the internals. 00:16:49.220 --> 00:16:53.560 But basically, every call is like an isolated execution. 00:16:53.560 --> 00:16:54.440 And then it goes away. 00:16:54.660 --> 00:16:57.640 That's part of the appeal also is that you don't really have to care about that. 00:16:57.640 --> 00:16:58.380 Yeah, it's totally the appeal. 00:16:58.380 --> 00:17:00.220 The infrastructure just deals with it. 00:17:00.220 --> 00:17:04.600 Yeah, you do pay for a small latency on every request because it's got to like spin up this bit. 00:17:04.600 --> 00:17:12.000 But that's still for a lot of things that are not super performance critical, at least in the latency, it totally would be fine. 00:17:12.000 --> 00:17:18.160 So what I wanted to talk about this week is this thing called Chalice, a Python serverless micro framework for AWS. 00:17:18.160 --> 00:17:22.160 So basically, this is a Flask API. 00:17:22.160 --> 00:17:24.160 Another Flask API. 00:17:24.520 --> 00:17:26.060 Another Flask API. 00:17:26.060 --> 00:17:29.760 So you implement these functions in Chalice. 00:17:30.080 --> 00:17:32.720 And you decorate them with this Flask API. 00:17:33.720 --> 00:17:36.520 But instead of running a Flask app, you create a Chalice app. 00:17:36.520 --> 00:17:40.820 And then you can even say like Chalice deploy and run it. 00:17:40.820 --> 00:17:45.120 And then you just go and basically request your thing. 00:17:45.120 --> 00:17:51.620 So what you effectively create is something that looks like a Flask app with all the different pieces that are related and so on. 00:17:51.760 --> 00:17:57.900 But then they're hosted as individual function calls on the serverless infrastructure. 00:17:58.460 --> 00:17:59.460 So it's pretty cool, right? 00:17:59.460 --> 00:17:59.680 Wow. 00:17:59.680 --> 00:18:00.400 Okay, that's cool. 00:18:00.400 --> 00:18:00.520 Yeah. 00:18:00.520 --> 00:18:02.260 So I think this is really interesting. 00:18:02.260 --> 00:18:08.000 So this is just like we're going to take Flask and put it onto AWS Lambda, which is pretty cool. 00:18:08.000 --> 00:18:11.540 Maybe the other counterpart that I know pretty well is Zappa. 00:18:12.000 --> 00:18:15.380 So Zappa is, I believe, its own framework. 00:18:15.380 --> 00:18:17.380 I don't think it has a Flask variation. 00:18:17.380 --> 00:18:18.300 I could be wrong. 00:18:18.300 --> 00:18:19.300 I haven't done that much with it. 00:18:19.400 --> 00:18:28.480 But Zappa also lets you basically create this API against this website equivalent scaled out across these different functions. 00:18:28.480 --> 00:18:29.040 Yeah. 00:18:29.040 --> 00:18:30.760 So pretty cool things to be playing with. 00:18:30.760 --> 00:18:32.200 Like you don't even need cloud anymore. 00:18:32.200 --> 00:18:33.280 Now you don't even need servers. 00:18:33.280 --> 00:18:34.080 It's all serverless. 00:18:34.080 --> 00:18:35.440 Yeah. 00:18:35.440 --> 00:18:43.580 So you still, I mean, of course, you still have to set up like an AWS account and get your token or whatever to hook into this, I assume. 00:18:43.580 --> 00:18:44.440 Yeah, exactly. 00:18:44.440 --> 00:18:46.760 You've got to get your own token and all those things. 00:18:46.760 --> 00:18:55.340 And yeah, you probably are talking to like some sort of hosted database or S3 or some combination thereof to actually do your work, right? 00:18:55.340 --> 00:18:55.680 Right. 00:18:55.680 --> 00:19:02.460 And actually, so I've never worked with AWS, but these sorts of things make me kind of want to try because it doesn't look scary at all. 00:19:02.460 --> 00:19:02.800 Absolutely. 00:19:02.800 --> 00:19:04.760 So let's wrap it back to item one. 00:19:04.940 --> 00:19:18.700 When you create your Flask skill, your Ask Flask stuff, and when you work with Alexa, one of the main ways to work with Alexa is to actually tie, see? 00:19:18.700 --> 00:19:20.100 See, she's speaking to me now. 00:19:20.100 --> 00:19:25.960 To have her, you could tie those two Lambda functions, these serverless functions. 00:19:25.960 --> 00:19:27.840 So actually, you could put it all together like that. 00:19:27.840 --> 00:19:28.160 Neat. 00:19:28.260 --> 00:19:28.680 Yeah, cool. 00:19:28.680 --> 00:19:29.140 Definitely. 00:19:29.140 --> 00:19:32.880 So I think you can probably help some people with job interviews coming up here. 00:19:32.880 --> 00:19:35.280 First, how do I reverse a string? 00:19:35.280 --> 00:19:39.940 Second, what's the fastest way to uniqueify a list in Python? 00:19:39.940 --> 00:19:41.180 Reverse a string? 00:19:41.180 --> 00:19:41.880 Yes. 00:19:41.880 --> 00:19:51.620 This is a short little thing, but there's a blog article called The Fastest Way to Uniquify a List in Python greater than or equal 3.6. 00:19:51.980 --> 00:19:56.340 So basically, 3.6, unless you're running 3.7 already. 00:19:56.340 --> 00:20:01.500 I just appreciated the, instead of having just an article that says, okay, here's the answer. 00:20:01.500 --> 00:20:06.580 There is a discussion of the different ways you might do it and the code for how he timed it. 00:20:06.580 --> 00:20:13.340 So you can reproduce all of, if you don't believe it, you can reproduce all the code and get your results on your machine. 00:20:13.340 --> 00:20:14.780 It's probably similar. 00:20:14.960 --> 00:20:23.580 The short answer is, if you want to keep the order in place, you simply do a list of a dict from keys of your sequence. 00:20:23.580 --> 00:20:27.360 And we'll have this in the show notes, of course, and a link to the article. 00:20:27.360 --> 00:20:33.260 And if you don't care about the order, then just convert it to a set and then a list. 00:20:33.260 --> 00:20:35.240 So that's it. 00:20:35.560 --> 00:20:38.580 And I've used the list set before. 00:20:38.580 --> 00:20:42.220 And I was actually, I mean, it makes sense that from keys works. 00:20:42.220 --> 00:20:44.420 I've never really used that before. 00:20:44.420 --> 00:20:45.040 So that. 00:20:45.040 --> 00:20:45.280 Yeah. 00:20:45.280 --> 00:20:49.720 And it's going to get interesting with the promises of dictionaries being ordered. 00:20:49.720 --> 00:20:53.300 You could actually get some of these to be a little quicker, right? 00:20:53.300 --> 00:20:54.440 But still preserving the order. 00:20:54.440 --> 00:20:56.860 Well, isn't that the way it is now? 00:20:56.860 --> 00:20:57.440 Not all. 00:20:57.440 --> 00:21:02.860 Like some of them, there's like two graphs that says not order preserving functions and then order preserving functions. 00:21:03.460 --> 00:21:07.760 And right now they have list of set of list. 00:21:07.760 --> 00:21:09.600 Not order preserving. 00:21:09.600 --> 00:21:10.120 Right. 00:21:10.120 --> 00:21:12.540 Because set isn't an order preserving thing. 00:21:12.540 --> 00:21:13.080 So, yeah. 00:21:13.080 --> 00:21:13.860 But it's pretty cool, though. 00:21:13.860 --> 00:21:14.400 Pretty cool. 00:21:14.400 --> 00:21:18.940 Definitely worth thinking about all the different ways and seeing the tradeoffs and so on. 00:21:18.940 --> 00:21:19.200 Okay. 00:21:19.200 --> 00:21:19.600 All right. 00:21:19.600 --> 00:21:22.580 So I want to wrap up this episode with leaving some people with homework. 00:21:22.580 --> 00:21:28.140 They've got some research and some chilling on the couch watching some Python videos. 00:21:28.140 --> 00:21:28.520 Okay. 00:21:28.680 --> 00:21:33.320 So there's two sets of conference videos that came out. 00:21:33.320 --> 00:21:34.680 And I'm sure there's some that I've missed. 00:21:34.680 --> 00:21:36.560 But these are the two that stood out to me recently. 00:21:36.560 --> 00:21:42.140 Pi Texas and PiCon Australia both have their videos up. 00:21:42.140 --> 00:21:46.560 And I linked to the playlist on YouTube for both of them, which is pretty cool. 00:21:46.560 --> 00:21:49.920 There's, you know, like always like 40 or more videos. 00:21:49.920 --> 00:21:51.680 There's tons of really interesting stuff. 00:21:51.680 --> 00:21:57.140 But I kind of wanted to touch on some that I, these are not necessarily endorsements. 00:21:57.140 --> 00:21:59.040 These are just like, these look really interesting to me. 00:21:59.040 --> 00:22:00.060 And I'd like to go watch them. 00:22:00.060 --> 00:22:03.640 I haven't spent all this time watching all these videos just right now. 00:22:03.640 --> 00:22:07.620 But from Pi Texas, there's a talk on MicroPython. 00:22:07.620 --> 00:22:10.280 Which MicroPython is really, really cool. 00:22:10.280 --> 00:22:10.540 Right? 00:22:10.580 --> 00:22:18.100 You can run your Python on like a $5 microchip and hook your Lambda functions, your Python ones, up to like hardware interrupts. 00:22:18.100 --> 00:22:19.160 Like that's just the coolest. 00:22:19.160 --> 00:22:24.160 And then there's so much talk about machine learning and AI and all that stuff. 00:22:24.160 --> 00:22:28.360 So there was a talk given there called, What is Machine Learning Anyway? 00:22:28.360 --> 00:22:29.960 Or something like that. 00:22:29.960 --> 00:22:33.580 So like a foundational background on machine learning, which is cool. 00:22:33.680 --> 00:22:37.360 And another one is C for yourself with the letter C, not S-E-E. 00:22:37.360 --> 00:22:40.420 So like trying to explore the internals of CPython. 00:22:40.420 --> 00:22:46.420 And then there's one on Python and .NET for this thing called Python.NET, which is not the Iron Python that's been around for a long time. 00:22:46.420 --> 00:22:49.400 But like this newer way to try to integrate those things. 00:22:49.400 --> 00:22:54.920 So all the people that work on the big enterprise apps in C#, they can now plug in some sweet Python or vice versa. 00:22:54.920 --> 00:22:55.440 Yeah. 00:22:55.440 --> 00:22:55.880 Okay. 00:22:55.880 --> 00:22:59.180 Those all look good to me too, especially the MicroPython. 00:22:59.180 --> 00:22:59.460 Yeah. 00:22:59.460 --> 00:23:00.560 The MicroPython is great. 00:23:00.560 --> 00:23:01.360 So that's in Texas. 00:23:01.360 --> 00:23:02.320 And then Australia. 00:23:02.320 --> 00:23:03.400 I don't know. 00:23:03.480 --> 00:23:04.980 Those guys were killing it with names down there. 00:23:04.980 --> 00:23:06.300 They just really did a good job. 00:23:06.300 --> 00:23:06.920 And there's so many. 00:23:06.920 --> 00:23:08.500 I have more I pulled out. 00:23:08.500 --> 00:23:09.180 So I'll go a little quicker. 00:23:09.180 --> 00:23:10.500 Gradual typing. 00:23:10.500 --> 00:23:11.220 Right? 00:23:11.220 --> 00:23:13.880 We've talked about that, the typings and adding them and stuff. 00:23:13.880 --> 00:23:16.660 And I think this is probably a really interesting survey that. 00:23:16.660 --> 00:23:19.760 Hot reloading Python web servers at scale. 00:23:19.760 --> 00:23:21.220 Ooh, that sounds nice. 00:23:21.220 --> 00:23:21.740 Yeah. 00:23:21.740 --> 00:23:23.140 That sounds like a lot of cool internals. 00:23:23.140 --> 00:23:25.980 Prototyping Python microservices in production. 00:23:25.980 --> 00:23:31.160 And secrets of a WSGI master, like understanding the base HTTP protocol. 00:23:31.160 --> 00:23:33.060 And then another one. 00:23:33.280 --> 00:23:35.640 Python 3 for people who have not been paying attention. 00:23:35.640 --> 00:23:40.060 So that's not anybody that's listening to this. 00:23:40.060 --> 00:23:40.600 Exactly. 00:23:40.600 --> 00:23:42.160 They know about legacy Python. 00:23:42.160 --> 00:23:42.660 Come on. 00:23:42.660 --> 00:23:44.320 Then identity 2.0. 00:23:44.320 --> 00:23:48.000 What, what, why, and how of social and federated logins. 00:23:48.000 --> 00:23:51.860 So if you've got to do anything with like identity across providers, that's so painful. 00:23:51.860 --> 00:23:52.540 This would be cool. 00:23:53.020 --> 00:23:56.000 And then here's a sort of a Tesla reference. 00:23:56.000 --> 00:23:59.200 I'm guessing Python ludicrous mode with Django. 00:23:59.200 --> 00:24:04.120 So ludicrous mode in the Teslas is like, you know, like the insane amount of power. 00:24:04.120 --> 00:24:07.880 Like, you know, give you a neck ache if you hit the gas yard sort of thing. 00:24:07.960 --> 00:24:09.380 So something to that effect. 00:24:09.380 --> 00:24:12.880 And then finally, scaling down. 00:24:12.880 --> 00:24:16.420 We often talk about scale up and all this complexity you got to deal with. 00:24:16.420 --> 00:24:18.240 You even talked about how you're not Google. 00:24:18.240 --> 00:24:21.200 You're not Facebook or any of these things a while back, which was cool. 00:24:21.380 --> 00:24:26.320 But this is, if you do have one of these large sites that's got a complicated architecture, 00:24:26.320 --> 00:24:32.620 let's say a Redis back-end message queue type thing and a database and a whole bunch of other 00:24:32.620 --> 00:24:34.600 stuff, maybe some of those are hosted in the cloud. 00:24:34.600 --> 00:24:37.540 Like, how do you actually develop that locally also? 00:24:37.540 --> 00:24:38.480 That's cool. 00:24:38.480 --> 00:24:41.480 That's definitely a need for everyone, I think. 00:24:41.480 --> 00:24:42.920 I'm sure there's a lot of good lessons in there. 00:24:42.920 --> 00:24:47.260 So the links to both of those playlists are up on the show notes. 00:24:47.260 --> 00:24:47.900 So that should be fun. 00:24:47.900 --> 00:24:48.800 All right, cool. 00:24:48.800 --> 00:24:49.160 All right. 00:24:49.160 --> 00:24:49.840 Well, how about you? 00:24:49.840 --> 00:24:51.580 Any other news? 00:24:51.580 --> 00:24:52.520 We're out. 00:24:52.520 --> 00:24:54.620 So that was six. 00:24:54.620 --> 00:24:54.860 Cool. 00:24:54.860 --> 00:24:55.160 Cool. 00:24:55.160 --> 00:24:57.260 You've actually been crushing it lately. 00:24:57.260 --> 00:24:59.280 I've been so busy. 00:24:59.280 --> 00:25:00.220 It's been so fun, though. 00:25:00.220 --> 00:25:01.400 Yeah. 00:25:01.400 --> 00:25:05.000 So last week, I released my Everything Bundle where people can get all the courses, right? 00:25:05.000 --> 00:25:10.040 And so I shipped my first course for that Everything Bundle called Mastering PyCharm, 00:25:10.040 --> 00:25:16.980 which, yes, it covers PyCharm, but it's also basically like a little mini course on like 16 00:25:16.980 --> 00:25:19.000 different parts of software development. 00:25:19.000 --> 00:25:23.940 So there's like a little tiny 20 minute thing on refactoring and a 15 minute thing on unit 00:25:23.940 --> 00:25:24.860 testing and pytest. 00:25:24.860 --> 00:25:25.260 Okay. 00:25:25.260 --> 00:25:27.220 30 minute thing on Git and stuff. 00:25:27.220 --> 00:25:31.200 So it's like sort of all those little pieces put together, but through the lens of working 00:25:31.200 --> 00:25:31.860 with PyCharm. 00:25:31.860 --> 00:25:33.300 So I'll link to that. 00:25:33.300 --> 00:25:36.480 And yeah, people can check that out if they want. 00:25:37.340 --> 00:25:40.360 And I'm also doing a webcast with the JetBrains people, but on MongoDB. 00:25:40.360 --> 00:25:45.440 So if they want to, people want to drop in on that on January 30th, I'm doing a live event 00:25:45.440 --> 00:25:48.360 with JetBrains and the link for that registration will be there. 00:25:48.360 --> 00:25:48.740 It's free. 00:25:48.740 --> 00:25:49.780 You can come check it out. 00:25:49.780 --> 00:25:50.100 Cool. 00:25:50.100 --> 00:25:51.080 I'll come watch. 00:25:51.080 --> 00:25:51.400 All right. 00:25:51.400 --> 00:25:51.700 Awesome. 00:25:51.700 --> 00:25:54.400 Well, yeah, that's all I've been doing this week. 00:25:54.400 --> 00:25:54.900 Nice. 00:25:56.480 --> 00:25:56.760 Great. 00:25:56.760 --> 00:25:57.000 All right. 00:25:57.000 --> 00:25:57.780 Well, thank you, Brian. 00:25:57.780 --> 00:25:58.280 Thank you. 00:25:58.280 --> 00:25:59.500 A lot of good stuff this week. 00:25:59.500 --> 00:25:59.740 Yeah. 00:25:59.740 --> 00:26:01.960 And looking forward to more next. 00:26:01.960 --> 00:26:02.860 Talk to you next week. 00:26:02.860 --> 00:26:03.260 All right. 00:26:03.260 --> 00:26:03.420 Bye. 00:26:04.160 --> 00:26:05.900 Thank you for listening to Python Bytes. 00:26:05.900 --> 00:26:08.480 Follow the show on Twitter via at Python Bytes. 00:26:08.480 --> 00:26:11.360 That's Python Bytes as in B-Y-T-E-S. 00:26:11.360 --> 00:26:14.800 And get the full show notes at pythonbytes.fm. 00:26:14.800 --> 00:26:19.120 If you have a news item you want featured, just visit pythonbytes.fm and send it our way. 00:26:19.120 --> 00:26:21.840 We're always on the lookout for sharing something cool. 00:26:21.840 --> 00:26:25.220 On behalf of myself and Brian Okken, this is Michael Kennedy. 00:26:25.220 --> 00:26:28.840 Thank you for listening and sharing this podcast with your friends and colleagues.