WEBVTT 00:00:00.001 --> 00:00:04.000 Hello and welcome to Python Bytes, where we deliver Python news and headlines directly to 00:00:04.000 --> 00:00:09.940 your earbuds. This is episode 238, recorded June 15th, 2021. I'm Michael Kennedy. 00:00:09.940 --> 00:00:11.060 And I'm Brian Okken. 00:00:11.060 --> 00:00:12.160 And I'm Julia Sidnell. 00:00:12.160 --> 00:00:14.460 Hey, Julia. Thanks for coming on the show. 00:00:14.460 --> 00:00:15.920 Yeah, thanks for having me. 00:00:15.920 --> 00:00:19.420 Yeah, it's great. Why don't you tell folks a bit about yourself? 00:00:19.420 --> 00:00:25.680 Yeah, so I'm the head of open source at SaturnCrowd and a maintainer of Dask. So I split my time 00:00:25.680 --> 00:00:31.300 half and half. I spend half my time just doing regular like maintenance stuff on Dask. And then 00:00:31.300 --> 00:00:37.240 half my time doing like engineering and product management on SaturnCrowd. SaturnCrowd is a 00:00:37.240 --> 00:00:42.620 data science platform that really specializes in distributed Dask clusters in Jupyter and making 00:00:42.620 --> 00:00:46.800 it really easy for people to get up and going with those things on AWS. 00:00:46.800 --> 00:00:52.280 Yeah, Dask is really interesting. You know, when I first heard about it, I thought, okay, 00:00:52.280 --> 00:00:57.080 this is a like a grid computing scale out thing, which I probably don't have a lot of use for. 00:00:57.080 --> 00:01:01.520 But then I was speaking with Matthew Rocklin about it. And it has a lot of applicability, 00:01:01.520 --> 00:01:07.700 even if you have not huge data, huge clusters, right? Like you can say even on your local machine 00:01:07.700 --> 00:01:12.520 scale this out across my cores or, you know, allow me to work with more data than will fit in RAM 00:01:12.520 --> 00:01:15.040 on my laptop and stuff like that. Right. It's a cool. 00:01:15.040 --> 00:01:19.540 Yeah. Yeah. It has like a whole different, a whole number of different ways of interacting with it. 00:01:19.540 --> 00:01:25.040 Right. Like there's that there's like, just make this thing go faster by paralyzing it. There's all the data 00:01:25.040 --> 00:01:30.540 framey stuff. There's all the array stuff for more dimensional data. So it's got a, it's got a large API. 00:01:30.540 --> 00:01:36.700 Yeah. Cool. And we're going to touch on a couple of topics that are not all that unrelated to, to those things 00:01:36.700 --> 00:01:41.460 here. And so, yeah. Speaking of data science, Brian, you want to kick us off? 00:01:41.460 --> 00:01:48.320 Sure. Yeah. The first thing I want to cover is an article called the practical SQL for data analysis. 00:01:48.320 --> 00:01:57.580 This is by Aki Benita. So I, one of the things I like, liked about this is it was, it's kind of talking 00:01:57.580 --> 00:02:03.960 about the first bit of the article was talking about basically that with, with data science, 00:02:03.960 --> 00:02:08.960 you've got pandas and, and numpy and stuff. And you also have often you're dealing with a database. 00:02:08.960 --> 00:02:17.520 So the, and SQL on the backend. So there's the first part of the article talks about how some 00:02:17.520 --> 00:02:26.860 things you can do both in pandas and in, in, in SQL, like SQL queries it's faster in SQL. So there's a big 00:02:26.860 --> 00:02:33.520 chunk that's just talking about how that's faster. But then, but then, you know, he also talks about just 00:02:33.520 --> 00:02:38.700 basically there's, there's a lot of benefits to the flexibility and the comfortableness you can have with 00:02:38.700 --> 00:02:44.440 pandas though. So trade offs as to where, you know, you can, where you're going to push the, push it too far 00:02:44.440 --> 00:02:50.340 into SQL or, or have a nice split is good. But then he goes through and talks about a whole bunch of 00:02:50.340 --> 00:02:57.060 great examples of different things like pivot tables and roll-ups and, and choices and different things 00:02:57.060 --> 00:03:02.500 you can do with either pandas or SQL. And really what his recommendations are for whether it should be 00:03:02.500 --> 00:03:08.400 in pandas or in, in SQL query, and then how to do those queries, because I mean, 00:03:08.400 --> 00:03:15.120 really the gist of the articles and, and this problem space is people are comfortable with pandas, 00:03:15.120 --> 00:03:21.560 but they don't really understand SQL queries. So this sort of a good cheat sheet for, for, for how to do 00:03:21.560 --> 00:03:24.880 the queries is a, is a, I think really kind of a cool thing. So. 00:03:24.880 --> 00:03:30.380 Yeah. I think it's really neat. And you, you have these problems, you know how to solve them in one 00:03:30.380 --> 00:03:35.900 or the other. And I think this compare and contrast is really valuable, right? Like I know how to take 00:03:35.900 --> 00:03:40.680 the mean of some column and SQL, but I haven't done it in pandas yet. Let's go see how to do that. Or 00:03:40.680 --> 00:03:45.620 I'm really good at doing pivot tables and pandas, but boy, always kind of avoided joins in SQL. They 00:03:45.620 --> 00:03:49.640 scared me. And then how does that even translate? Right. I think that back and forth is really valuable. 00:03:49.780 --> 00:03:53.740 Yeah. Yep. And I, then it covers, covers things that I don't even know what they are, 00:03:53.740 --> 00:03:58.680 like aggregate expressions. I don't even know what that is, but currently that's a, that's a thing 00:03:58.680 --> 00:04:03.660 that people do. So I can help you out at aggregate stuff. No, just kidding. Julia, what do you think 00:04:03.660 --> 00:04:08.740 of this? Yeah, no, it seems, it's really cool. Like, I agree that like the, having the, the, 00:04:08.740 --> 00:04:13.860 having independence and then in SQL, that like comparison is super helpful. Like SQL is always super 00:04:13.860 --> 00:04:18.640 scary to me. And I always end up like Googling a bunch of stuff whenever I have to mangle my SQL. 00:04:18.640 --> 00:04:23.100 but I know it's so fast, so it's cool to see a way to access that. 00:04:23.100 --> 00:04:26.400 Yeah, absolutely. This is a good one, Brian. I think a lot of people will find it useful. 00:04:26.400 --> 00:04:31.580 I also want to just give a quick shout out to the past a little bit, not too long ago. We caught, 00:04:31.580 --> 00:04:37.320 we talked about efficient SQL on pandas with duck DB, where you actually do the SQL queries against 00:04:37.320 --> 00:04:43.280 pandas data frames. So if you're finding that you're trying to do something and maybe it would be 00:04:43.280 --> 00:04:47.920 better in SQL, but you don't want to say completely switch all your, your data over to a relational 00:04:47.920 --> 00:04:52.240 database, you just kind of want to stay in the panda side, but is that one or two things like, 00:04:52.240 --> 00:05:00.640 this is really cool. This sort of upgrade your data frame to, execute SQL with the duck DB query 00:05:00.640 --> 00:05:06.680 optimizer is also a kind of a nice intermediary there. Yeah. Dask also does some, I'm going to try not 00:05:06.680 --> 00:05:11.440 not to make everything about Dask, but, Dask does some things that are kind of, that kind of 00:05:11.440 --> 00:05:16.760 take some of the ideas in the, from this article of like doing predicate push down of like, of 00:05:16.760 --> 00:05:22.660 pushing down some of the like filters into the read because it, because it evaluates lazily. It doesn't 00:05:22.660 --> 00:05:27.280 have to like grab all the data greedily upfront. It can like do that later. so you can get some 00:05:27.280 --> 00:05:31.420 of the benefits. That's cool. And it can also distribute the filter bit, I guess at that point. 00:05:31.420 --> 00:05:37.840 Yeah. Nice. All right. I want to talk about the usual suspects. So, okay. That was, that was a 00:05:37.840 --> 00:05:41.860 pretty good show. Was that Quentin Tarantino or something like that? It was not actually about 00:05:41.860 --> 00:05:47.300 this. This comes to us from wrestling port port. Noy. And thank you for sending this in mentioned 00:05:47.300 --> 00:05:55.060 an article that has this really interesting idea. How do you apply get blame when you encounter a 00:05:55.060 --> 00:06:01.080 Python traceback? So here's the scenario, your code crashes and you either print out the traceback or 00:06:01.080 --> 00:06:06.720 Python does it for you because it's just crashed. And normally it says, here's the value. Here's the 00:06:06.720 --> 00:06:12.100 line of code. Here's the file. It's in here's the next line in the call stack. Here's a line of 00:06:12.100 --> 00:06:18.980 code. It's in the idea is you can take get blame, which is a command that says, show me who changed 00:06:18.980 --> 00:06:24.180 this line of code or who wrote this line of code, at least touched it last on every single line of 00:06:24.180 --> 00:06:28.360 code. And I love this whole idea of like, all right, who did this? And sometimes I'll come across code. 00:06:28.360 --> 00:06:33.820 I'm like, this is so crappy. Like who did this? Oh wait, that's me. Okay. Well, at least I know 00:06:33.820 --> 00:06:38.820 what I would feel about it. But the idea is what if your traceback on each line where it had an 00:06:38.820 --> 00:06:45.260 exception could also show who wrote that line of code. Cool. Huh? Yeah. So let's check it out. 00:06:45.260 --> 00:06:51.800 It's pretty straightforward. This is an article by offer core or in, and it basically uses two libraries 00:06:51.800 --> 00:06:57.380 that are themselves both pretty straightforward. So like here's a straightforward example of a traceback, 00:06:57.380 --> 00:07:02.960 like trying to pop something off of an empty list says on this line in the function pop some, 00:07:02.960 --> 00:07:06.940 you know, there's this line here in the call stack. And then the next line, this line, 00:07:06.980 --> 00:07:12.240 the call stack and eventually raise a value error, you know, empty range, can't pop nothing off, 00:07:12.240 --> 00:07:16.400 you know, something off of nothing basically. But this doesn't show you any information about like 00:07:16.400 --> 00:07:22.040 maybe who wrote that line and who wrote this other line up here. Right. So, what they did is they 00:07:22.040 --> 00:07:28.660 took a couple of modules traceback and then line cache. And it turns out when traceback shows you this 00:07:28.660 --> 00:07:34.540 traceback that line, it uses line cache to figure out, okay, from this actual, I'm guessing, 00:07:34.940 --> 00:07:41.480 bytecode that it's going to run this, CPython interpreter code. Where did it like, 00:07:41.480 --> 00:07:48.300 what line of file did this actually come from? Right. So here's the insight or the thing 00:07:48.300 --> 00:07:54.580 you can actually change what's in the cache. And because it's a cache, once it's figured out what 00:07:54.580 --> 00:08:00.260 the lines are, it's not going to read it again. So it's like, like a list for each line that you 00:08:00.260 --> 00:08:05.320 get back and you can just change the value. So it said, okay, well, here's like return random. 00:08:05.320 --> 00:08:08.740 That's what the line of text was. They're like, no, no, no, there's nothing to see here. Move along. 00:08:08.740 --> 00:08:13.480 If you make that and then you cause it to crash again, what comes out is, if you go a little 00:08:13.480 --> 00:08:17.800 bit further down, normal code, normal code, or normal traceback, normal traceback, then it just, 00:08:17.800 --> 00:08:20.960 instead of the line of code, it says nothing to see here. Please move along. 00:08:21.880 --> 00:08:25.740 All right. So what are you going to do with that? Now that you realize like you can actually 00:08:25.740 --> 00:08:33.200 change what appears in the traceback. So you write a little regular expression to go and execute, 00:08:33.200 --> 00:08:39.580 get blame on the various files, and then to re-inject that back into line cache. And so what they do is 00:08:39.580 --> 00:08:45.060 they just put, if they know the blame, they just put, you know, like 80 lines, 80 characters up to 80 00:08:45.060 --> 00:08:50.180 characters of the line and then edited on such and date, such and such date by such and such person. 00:08:50.180 --> 00:08:56.600 And here's the, commit message. Right. And so just basically shelling out to get blame when it 00:08:56.600 --> 00:09:02.000 crashes. Now you get some really cool stuff. Like on this line, it says this is edited by, you know, 00:09:02.000 --> 00:09:07.940 many, many days ago by so-and-so in this, get commit and so on. And what's interesting, 00:09:07.940 --> 00:09:13.220 like this is already in itself useful, I think, but what's more interesting is other tools use this as 00:09:13.220 --> 00:09:19.180 well. So for example, if you use PUDB, which is a sort of visual debugger, kind of, it's like a 00:09:19.180 --> 00:09:24.320 command line one that I know visual in the sense of like Emacs is visual, not like PyCharm is visual, 00:09:24.320 --> 00:09:29.960 but it will actually pull up that data. So you can see they, they jumped into the PUDB bugger and it's 00:09:29.960 --> 00:09:35.220 actually showing all of this get blame attribution as well that they've added. So yeah, pretty interesting. 00:09:35.220 --> 00:09:40.580 What do you all think? Yeah, I think that looks really cool. I mean, I always do get blame whenever I run 00:09:40.580 --> 00:09:43.980 into something that's weird with the hope that someone else will be able to explain it to me. 00:09:43.980 --> 00:09:47.700 Exactly. Who knows about this or who do I talk to about breaking this? 00:09:47.700 --> 00:09:51.820 Right. Yeah. You could even put like PR numbers and stuff in here. Right. And that'd be pretty cool. 00:09:51.820 --> 00:09:54.620 Yeah. Yeah. That'd be super cool. 00:09:54.620 --> 00:09:59.660 Yeah. One of the things I like, I don't really like that the name get blame, but it's there. 00:10:00.280 --> 00:10:05.980 But I agree with Julia that the main thing I use it for isn't to try to figure out who broke it, 00:10:05.980 --> 00:10:09.040 but who to ask about this, this chunk of the code. 00:10:09.040 --> 00:10:13.960 Yeah. I agree. Cause usually when you see something that's really confusing and weird, you're like, 00:10:13.960 --> 00:10:19.020 I know they didn't just pick the hard way of doing this because they didn't want to do the easy way. 00:10:19.020 --> 00:10:24.520 There's something that I don't fully understand. Some edge case that's crazy here. I'm going to go talk to 00:10:24.520 --> 00:10:29.360 that person. So yeah. Also the, how long ago it was edited. So if there was something edited 00:10:29.360 --> 00:10:33.800 yesterday, that's probably the problem. Yeah, exactly. Like in this little screenshot here, 00:10:33.800 --> 00:10:41.120 some of these are edited like 1,427 days ago. That's probably not the problem. Maybe, but probably not. 00:10:41.120 --> 00:10:44.760 I feel like I have the opposite assumption. Like if something is from six years ago and it's weird, 00:10:44.760 --> 00:10:48.760 I'm like, well, probably things were different back then. And like, you know. 00:10:48.760 --> 00:10:53.480 Yeah. Yeah. Yeah. It's no, no longer applicable to the new data, new situation. Yeah. 00:10:53.940 --> 00:10:58.240 Oh, that'd be an interesting thing also is to have like a tool that would tell you if something's like 00:10:58.240 --> 00:11:02.820 over a thousand days old or something like that, you probably should go refactor it to make sure 00:11:02.820 --> 00:11:08.560 somebody understands that code. Yeah. Yeah. For sure. All right. Jumping back to the first item 00:11:08.560 --> 00:11:13.160 really quick and the live stream, Alexander out there. Hey, Alexander says, I wonder if graph databases 00:11:13.160 --> 00:11:18.500 with Gremlin queries could be more suitable for data science. You know, SQL joins are way harder. 00:11:18.500 --> 00:11:22.420 Yeah. Graph databases are pretty interesting. If you're trying to understand the relationships, 00:11:22.420 --> 00:11:25.460 that may well be better. I don't know. Hila, do you got any thoughts on this? 00:11:25.460 --> 00:11:30.000 I don't know anything about graph databases. So out of my league. 00:11:30.000 --> 00:11:36.400 I didn't have a desire to understand graph databases until I found out that there were Gremlin queries. 00:11:36.400 --> 00:11:38.160 Now I think I want them. 00:11:38.160 --> 00:11:44.280 Well, Brian, they don't start out as a Gremlin queries. They're Mogwai inserts. And then if you 00:11:44.280 --> 00:11:48.800 insert them after midnight, then they become a Gremlin query. I mean, come on, we all know how it goes. 00:11:49.080 --> 00:11:53.920 You definitely don't want to get them wet. Oh, that's an old show. I'm not sure if everyone's 00:11:53.920 --> 00:11:57.860 going to get that reference, but yeah, that was, I love that show. Okay. Anyway, let's, 00:11:57.860 --> 00:12:01.080 let's move on to the next one. The next one is you, Julia. 00:12:01.080 --> 00:12:08.640 Yeah. So I wanted to highlight FS spec. So file system spec for people who can't hear letters very well. 00:12:08.640 --> 00:12:19.100 So this is the basis for S3 FS. FS, I'm not getting the letters right, but there's, there's one for GCP. 00:12:19.240 --> 00:12:27.560 There's one for S3 and basically it's a file system storage interface or like the basis for a file system. 00:12:28.020 --> 00:12:41.020 And so you can do things like you can open just files as you can just take a path and open it as a, as a, as a file object in Python and read it with all the normal, like read, write operations. 00:12:41.020 --> 00:12:42.680 Oh, interesting. 00:12:43.020 --> 00:13:02.580 But from anywhere. So like there's all these different ones for S3 for GCFs and for, for like, even for like HTTP and just basically anything you've, you can imagine anywhere you can imagine a file being either there's already been one of these written. 00:13:04.020 --> 00:13:11.340 It's kind of like a, it's an interface and then you write different packages on top of it that are like drivers or some, they have some name for it. 00:13:11.340 --> 00:13:18.120 And it allows you to treat the file system as like this interchangeable building block. 00:13:18.120 --> 00:13:26.380 So you don't get, you don't end up writing like photo three code or something that's like very specific to a specific cloud storage. 00:13:26.580 --> 00:13:38.240 You write like this more general code and then it's really useful for like a lot of free datasets that are hosted on different clouds, but like they'll sometimes be on one cloud and sometimes be on another, but like basically it's the same data. 00:13:38.240 --> 00:13:45.000 Or if you're at a company and you want to like switch clouds, it just makes that whole thing so much easier. 00:13:45.000 --> 00:13:50.300 It looks really, really useful, especially for avoiding cloud lock-in. 00:13:50.300 --> 00:13:51.620 Yeah. Yeah. 00:13:51.760 --> 00:13:54.680 And you can always write, like you can always write your own one. 00:13:54.680 --> 00:13:57.900 If something else pops up, you can write your own implementation of that. 00:13:57.900 --> 00:13:58.360 Right. 00:13:58.360 --> 00:14:08.080 So there's an example here talking about using a file system in the docs that says something to the effect of, well, you want to open up a CSV and feed it off to pandas, read CSV. 00:14:08.080 --> 00:14:14.500 So normally you would say open CSV file, and then you just say pandas, read CSV and give it the file stream. 00:14:14.500 --> 00:14:16.000 But what if that's on the internet? 00:14:16.000 --> 00:14:18.260 What if that's on S3 with authentication? 00:14:18.260 --> 00:14:18.900 What's that? 00:14:18.900 --> 00:14:20.700 What if that's somewhere else? 00:14:20.700 --> 00:14:20.960 Right. 00:14:21.020 --> 00:14:25.080 And so with this one, you can just say FS, file system spec, open. 00:14:25.080 --> 00:14:26.260 Here's a URL. 00:14:26.260 --> 00:14:28.120 And now that's a stream, right? 00:14:28.120 --> 00:14:32.080 Or that could be, here's an S3 location, S3 bucket. 00:14:32.080 --> 00:14:32.840 Go get that, right? 00:14:32.840 --> 00:14:33.760 Yeah. 00:14:33.760 --> 00:14:34.120 Yeah. 00:14:34.120 --> 00:14:39.880 So instead of passing the path directly into the read function, you pass in the file object. 00:14:39.880 --> 00:14:43.220 And it's really powerful. 00:14:43.220 --> 00:14:45.860 It seems like a thing that we shouldn't need. 00:14:45.860 --> 00:14:50.680 But files get like the file locations can get so crazy so quickly. 00:14:50.680 --> 00:14:57.980 And this just really helps simplify and like, make it so you don't have to think about this stuff, which I think is what most people want. 00:14:57.980 --> 00:14:58.720 It's what I want. 00:14:58.720 --> 00:15:00.080 Yeah, for sure. 00:15:00.200 --> 00:15:02.660 So like there's a local file system option. 00:15:02.660 --> 00:15:05.400 But then you could also have an FTP file system. 00:15:05.400 --> 00:15:07.420 Or you could have something else, right? 00:15:07.420 --> 00:15:08.460 All sorts of different options. 00:15:08.460 --> 00:15:09.020 Yeah. 00:15:09.020 --> 00:15:09.360 Yeah. 00:15:09.360 --> 00:15:10.400 All sorts of stuff. 00:15:10.400 --> 00:15:10.940 Yeah. 00:15:11.360 --> 00:15:11.660 Okay. 00:15:11.660 --> 00:15:12.140 That's cool. 00:15:12.140 --> 00:15:13.360 Brian, what do you think? 00:15:13.360 --> 00:15:15.220 Does it have any applicability for you? 00:15:15.220 --> 00:15:15.980 Oh, yeah. 00:15:15.980 --> 00:15:16.440 Definitely. 00:15:16.440 --> 00:15:24.760 And that's a great abstraction layer to put in place to just have reading as if it was a file and have it moved. 00:15:24.760 --> 00:15:30.580 It also helps you develop tools locally and then be able to deploy them into a larger space. 00:15:30.580 --> 00:15:31.540 So it's cool. 00:15:31.540 --> 00:15:32.380 Yeah, for sure. 00:15:32.380 --> 00:15:38.080 One of the things that always makes me a little hesitant when I hear people say things like, we're cloud native. 00:15:38.080 --> 00:15:40.000 Like my app is cloud native. 00:15:40.340 --> 00:15:41.800 That's always code word for me. 00:15:41.800 --> 00:15:44.860 Like I will never be able to run my app unless I'm connected to the internet. 00:15:44.860 --> 00:15:47.540 You know, it's like it depends on all these services together. 00:15:47.540 --> 00:15:49.880 And there's no way I can recreate that locally. 00:15:49.880 --> 00:15:54.480 But something like this could allow you to say, well, we're going to have a local file system version. 00:15:54.480 --> 00:15:57.880 But then when we go to production, we'll switch to S3 or, you know, pick it. 00:15:57.880 --> 00:15:58.520 Pick something. 00:15:58.520 --> 00:16:03.900 I've always wanted to make it either a t-shirt or a sticker or both that says not a cloud native, just visiting. 00:16:03.900 --> 00:16:06.660 Nice. 00:16:06.660 --> 00:16:09.600 I also think, Brian, there might be testing opportunities here. 00:16:09.860 --> 00:16:10.380 Yeah, definitely. 00:16:10.380 --> 00:16:12.400 Give it a test file system. 00:16:12.400 --> 00:16:13.080 That'd be cool. 00:16:13.080 --> 00:16:13.560 Yeah. 00:16:13.560 --> 00:16:19.960 And like Julia said, swapping things out to just have your logic not have to care where it's coming from. 00:16:20.960 --> 00:16:29.260 But I guess you'd have to make sure all of the interfaces, the different storage systems really are equal. 00:16:29.260 --> 00:16:32.040 But I guess you should try that out yourself. 00:16:32.420 --> 00:16:34.440 Yeah, there's like kind of a bucket, right? 00:16:35.220 --> 00:16:38.600 There's kind of like a dict that you can pass, which is like storage options. 00:16:38.600 --> 00:16:44.540 So I think that can, that might get a little wonky depending on what the different backends need. 00:16:44.540 --> 00:16:47.320 But the general principles are the same. 00:16:47.320 --> 00:16:58.160 And it also, I should have said this originally, but it also allows, like FS spec itself can contain logic to do things that are general to all the different libraries, like caching and things like that. 00:16:58.400 --> 00:16:59.320 It's all the different. 00:16:59.320 --> 00:16:59.780 Oh, interesting. 00:16:59.780 --> 00:17:03.600 Like you could put a caching layer on top of arbitrary things like S3. 00:17:03.600 --> 00:17:04.340 Yeah. 00:17:04.520 --> 00:17:07.520 Google storage and Azure buckets or blob storage. 00:17:07.520 --> 00:17:07.900 Yeah. 00:17:07.900 --> 00:17:08.400 Yeah. 00:17:08.400 --> 00:17:12.320 Maybe even save money on bandwidth there if you can do some caching. 00:17:12.320 --> 00:17:12.860 Yeah. 00:17:12.860 --> 00:17:13.540 If you can do it right. 00:17:13.540 --> 00:17:14.420 Yeah. 00:17:14.420 --> 00:17:15.560 Super, super neat. 00:17:15.560 --> 00:17:18.380 Brian, you're going to tell us about how to slim down our Docker containers. 00:17:18.380 --> 00:17:23.320 But before you do, I want to tell people about our sponsor for this episode brought to you by Sentry. 00:17:23.320 --> 00:17:28.460 So how would you like to remove a little stress from your life in addition to just abstracting your file system? 00:17:28.460 --> 00:17:30.140 Maybe tracking down some errors. 00:17:30.220 --> 00:17:35.760 So do you worry that your users may be having difficulties or encountering errors with your app right now? 00:17:35.760 --> 00:17:38.340 And would you even know it until they send that support email? 00:17:38.340 --> 00:17:44.740 How much better would it be if you got the error or performance details sent right away with all the call stack? 00:17:44.740 --> 00:17:46.160 Maybe you would get blame in there. 00:17:46.160 --> 00:17:51.700 The local variables, the active user who was logged in while this happened, all that kind of stuff. 00:17:51.700 --> 00:17:55.200 So with Sentry, it's not only possible, it's actually really simple. 00:17:55.380 --> 00:17:57.040 I've used this on Sentry. 00:17:57.040 --> 00:17:59.360 I've used Sentry on our websites before. 00:17:59.360 --> 00:18:02.360 So it's on Python Bytes, Talk Python Training, all those different sites. 00:18:02.360 --> 00:18:07.200 And I've actually had someone encounter an error trying to buy a course over on Talk Python Training. 00:18:07.200 --> 00:18:08.960 I got the Sentry notification. 00:18:08.960 --> 00:18:12.140 I said, oh, geez, I can't believe this problem crept in here. 00:18:12.140 --> 00:18:16.180 And I fixed it really quick and started to roll out the fix and actually got an email. 00:18:16.180 --> 00:18:18.340 They said, hey, we're having this problem buying a course. 00:18:18.340 --> 00:18:18.900 I know. 00:18:18.900 --> 00:18:19.980 I've almost got it fixed. 00:18:19.980 --> 00:18:21.840 Just give me a moment and try again. 00:18:21.840 --> 00:18:23.360 And they were just like, what? 00:18:23.360 --> 00:18:24.420 That doesn't make sense. 00:18:24.500 --> 00:18:25.840 So they were very surprised. 00:18:25.840 --> 00:18:30.220 And so it's surprising to let your users create your Sentry account at pythonbytes.fm slash 00:18:30.220 --> 00:18:30.620 Sentry. 00:18:30.620 --> 00:18:32.900 And when you sign up, there's a little got a promo code. 00:18:32.900 --> 00:18:37.780 Make sure that you put Python Bytes, all one word, all caps with a Y in there. 00:18:37.780 --> 00:18:41.200 And you'll get two free months plus a bunch of extra features and so on. 00:18:41.200 --> 00:18:45.780 So also, it really lets them know that you came from us rather than just somewhere else. 00:18:45.780 --> 00:18:46.960 And that helps support the show a lot. 00:18:46.960 --> 00:18:50.340 So pythonbytes.fm/sentry and promo code Python Bytes. 00:18:50.340 --> 00:18:50.900 Awesome. 00:18:50.900 --> 00:18:53.020 Thanks for supporting the show, Sentry. 00:18:53.020 --> 00:18:55.320 And Brian, let's talk Docker. 00:18:55.320 --> 00:18:57.240 Yeah, let's talk Docker. 00:18:57.240 --> 00:19:00.080 I mean, I'm starting to use Docker more and more. 00:19:00.080 --> 00:19:02.120 And I like the experience. 00:19:02.120 --> 00:19:05.780 But I was interested when this article came up. 00:19:05.780 --> 00:19:07.040 So it was in June. 00:19:07.500 --> 00:19:11.400 I saw this article called The Need for Slimmer Containers. 00:19:11.900 --> 00:19:15.120 And this is from somebody, Ivan. 00:19:15.120 --> 00:19:17.220 Ivan, I'm not going to try his last name. 00:19:17.220 --> 00:19:17.860 Ivan something. 00:19:17.860 --> 00:19:20.400 But anyway, it's an interesting discussion. 00:19:20.400 --> 00:19:29.300 And the idea around the original post was that there's now a Docker scan that you can use. 00:19:29.300 --> 00:19:34.680 So you can use Docker scan to scan for vulnerabilities in your Docker containers. 00:19:35.200 --> 00:19:40.220 And this, Ivan thought, well, I'll look at some of the standard Python containers that are available. 00:19:40.220 --> 00:19:40.820 Right. 00:19:40.820 --> 00:19:49.180 Theoretically, some of the things that are nice is I can just go and say Docker or in my Docker container, I can say from Python colon three nine. 00:19:49.340 --> 00:19:51.460 And I don't have to think about how do I install Python? 00:19:51.460 --> 00:19:52.700 How do I keep it up to date? 00:19:52.700 --> 00:19:58.260 You know, make sure that pip is there and that I'll be able to, you know, pip install stuff that needs to do build things. 00:19:58.260 --> 00:19:59.680 All that stuff will be there, right? 00:19:59.680 --> 00:20:02.120 So it seems like, of course, this is what you want. 00:20:02.120 --> 00:20:02.680 Yeah. 00:20:02.680 --> 00:20:06.640 Well, and also, that's kind of one of the neat things about Docker. 00:20:06.640 --> 00:20:08.940 I can just say I have these standard parts. 00:20:08.940 --> 00:20:11.800 Now I just want to put my custom stuff on top of it. 00:20:11.800 --> 00:20:13.900 And it's great. 00:20:13.900 --> 00:20:15.880 So, well, what did he find? 00:20:16.080 --> 00:20:23.980 So he used, so Docker scan apparently uses a third party tool called Snake, S-N-Y-K container. 00:20:23.980 --> 00:20:30.600 And we've covered Snake before, not the container version, but we covered Snake in episode 227. 00:20:30.600 --> 00:20:35.560 But so it's looking for vulnerabilities, and that's a good thing. 00:20:35.560 --> 00:20:37.420 But he found them in everything. 00:20:37.420 --> 00:20:43.780 And he found them in all of the standard Python ones, except for Alpine, I guess. 00:20:44.860 --> 00:20:47.340 And so he didn't really know what to make of it, really. 00:20:47.340 --> 00:20:53.280 He was just sort of reporting his results that maybe Alpine is the only one with few vulnerabilities. 00:20:53.280 --> 00:20:59.740 But then this went out on Hacker News, and there was a big discussion around it. 00:20:59.740 --> 00:21:05.380 So he updated the article, which I appreciate, with some of the feedback that he got. 00:21:06.180 --> 00:21:11.660 And so some of the feedback was that these vulnerability checkers sometimes give you false positives. 00:21:11.660 --> 00:21:23.920 And I don't really have enough experience to know what that means, but I don't have enough experience to know if these really are false positives or if they're actual vulnerabilities or not. 00:21:24.580 --> 00:21:33.460 The other thing that maybe some people suggested that these standard ones really aren't updated very much. 00:21:33.460 --> 00:21:35.560 So I don't really know much about that either. 00:21:35.560 --> 00:21:39.840 And if they're not, that's kind of a bummer because I think people are relying on them. 00:21:40.260 --> 00:21:45.680 So I actually just kind of am left with a little bit of a confusion as to what to do. 00:21:45.680 --> 00:21:54.280 I want to also mention that the Alpine, in his current one, or his original article, he says Alpine's pretty good for vulnerabilities. 00:21:54.280 --> 00:21:57.660 But then his follow-up says it doesn't really... 00:21:57.660 --> 00:22:01.680 Well, there's a lot of applications that can't run on Alpine because of some issues or another. 00:22:01.680 --> 00:22:03.960 So anyway, I'm not sure what to make of it. 00:22:03.960 --> 00:22:06.220 So I was hoping Michael might give us some insight. 00:22:06.220 --> 00:22:08.580 I did some thinking about this morning. 00:22:08.580 --> 00:22:14.660 And in fact, I recently spoke a lot about this over on Talk Python. 00:22:14.660 --> 00:22:21.040 So I had Itamar on the show, and we talked about best practices for Docker packaging. 00:22:21.040 --> 00:22:24.660 And we talked a lot about both security and package size. 00:22:25.020 --> 00:22:28.180 So I can try to relay a couple of things from that. 00:22:28.180 --> 00:22:32.160 So we've got our official image over here, our Python official image. 00:22:32.160 --> 00:22:34.020 There's actually a bunch of options. 00:22:34.020 --> 00:22:42.040 As you can see, there's a few, like 310 beta 2 Buster or the 310 RC Buster. 00:22:42.040 --> 00:22:44.460 That sounds bad, but I think it's actually good. 00:22:44.460 --> 00:22:44.940 No, I'm just kidding. 00:22:44.940 --> 00:22:45.460 I know what it is. 00:22:45.460 --> 00:22:49.000 So these are by default based on Debian. 00:22:49.000 --> 00:22:51.880 And Buster is the latest version of Debian. 00:22:51.880 --> 00:23:02.340 And so you can do a Buster, which is like full Debian with 310, or you can do a 310 Slim Buster, which is like a slimmed down version of Debian Buster that supports Python 310. 00:23:02.340 --> 00:23:02.900 Okay. 00:23:02.980 --> 00:23:07.300 So there's a lot going on here in terms of the options. 00:23:07.300 --> 00:23:14.380 One of, so the article talks about how Alpine had the fewest security vulnerabilities. 00:23:14.380 --> 00:23:23.640 And I actually, so the Python latest, if you run the sneak package scanner thingy on it, it says there's 364 vulnerabilities. 00:23:23.640 --> 00:23:32.660 If you just do Python latest, three, nine, and 353 after you run apt update, apt upgrade. 00:23:32.660 --> 00:23:38.040 So if you try to get the container to update itself, there's still 353 in the, that one. 00:23:38.040 --> 00:23:39.160 I don't use that. 00:23:39.160 --> 00:23:39.880 I use Ubuntu. 00:23:39.880 --> 00:23:45.680 So I use the Ubuntu latest and the bare version of that one had 31 vulnerabilities. 00:23:46.100 --> 00:23:58.100 But then if I either install Python through apt or, or build it through source and put it in the necessary foundational bits, like build essentials and stuff to build Python, it goes up to 35 total problems. 00:23:58.100 --> 00:23:59.900 We're at 28 of Merlot. 00:23:59.900 --> 00:24:01.600 So seven or medium, nothing major. 00:24:01.820 --> 00:24:13.500 One thing I thought was weird was I actually ran another step where I said, okay, let's uninstall those intermediate tools like GCC and Wget and stuff like that, that I needed to get stuff on the machine, but I'm not going to use again. 00:24:13.500 --> 00:24:15.040 And I took them away. 00:24:15.040 --> 00:24:19.300 And almost all those warnings were about those tools that I had apt uninstalled. 00:24:19.300 --> 00:24:21.660 So I don't know why sneak is still showing them. 00:24:21.660 --> 00:24:26.440 Cause if I go into the container, I type Wget, it says, Nope, this thing is not installed. 00:24:26.440 --> 00:24:26.800 Sorry. 00:24:27.220 --> 00:24:31.400 But it still says the warning is that Wget has a vulnerability in it, for example. 00:24:31.400 --> 00:24:31.740 Right. 00:24:31.740 --> 00:24:34.620 So there's like, there's like this over-reporting for sure. 00:24:34.620 --> 00:24:38.740 But I mean, the difference between 28 and 350 is not trivial. 00:24:38.740 --> 00:24:39.320 Right. 00:24:39.320 --> 00:24:39.920 Right. 00:24:39.920 --> 00:24:44.760 So like running an apt install Python three type of thing is not, you know, it's probably worth it. 00:24:44.760 --> 00:24:45.200 For example. 00:24:45.200 --> 00:24:54.660 When I switched from Python three, nine to Python three, nine slim buster, it went from 350 to 69. 00:24:54.660 --> 00:24:56.060 So that's a lot better. 00:24:56.060 --> 00:24:56.380 Right. 00:24:56.380 --> 00:24:56.860 Yeah. 00:24:57.160 --> 00:25:00.240 it's still not as good as a moon two, but it's a lot better. 00:25:00.240 --> 00:25:02.100 The it's still twice as many. 00:25:02.100 --> 00:25:09.220 I mean, you can't, it sounds better, but it could be like 359 low problems and then 69 critical ones. 00:25:09.220 --> 00:25:10.780 it totally could. 00:25:10.780 --> 00:25:11.480 It totally could. 00:25:11.480 --> 00:25:11.740 Yeah. 00:25:11.740 --> 00:25:19.300 Also if the reporting, like if the, if, if we can't trust snake necessarily, then like maybe, 00:25:19.300 --> 00:25:25.120 you know, if you can't trust your reporting system, then like maybe that, maybe none of this is means anything. 00:25:25.120 --> 00:25:25.580 Right. 00:25:25.580 --> 00:25:25.840 Yeah. 00:25:25.840 --> 00:25:26.640 Yeah. 00:25:26.640 --> 00:25:32.800 I think one of the things the article originally started out to address was if you have fewer subsystems, 00:25:32.800 --> 00:25:36.780 there's no chance the missing subsystem could get hacked because it's not there. 00:25:36.780 --> 00:25:37.260 Right. 00:25:37.260 --> 00:25:42.000 So if there's a vulnerability in SSH, but you literally don't install SSH, who cares? 00:25:42.260 --> 00:25:46.420 Whereas if you, you know, you're going to have a lot of things that you're going to have to do with your own. 00:25:46.420 --> 00:25:47.540 You're going to have to do with your own. 00:25:47.540 --> 00:25:48.540 You're going to have to do with your own. 00:25:48.540 --> 00:25:49.740 You're going to have to do with your own. 00:25:49.740 --> 00:25:49.900 Yeah. 00:25:49.900 --> 00:25:52.400 And then it went down this rattle of like, well, let me scan it. 00:25:52.400 --> 00:25:52.880 And so on. 00:25:52.880 --> 00:25:54.540 So I want to add one more thing. 00:25:54.540 --> 00:26:01.460 Like Alpine did result in the best outcome from the scanner, but there's a lot of issues with Alpine and Python. 00:26:01.460 --> 00:26:10.680 So for example, there's this PEP here, 656, that right now, if I try to pip install something on Alpine. 00:26:10.680 --> 00:26:15.840 So especially in the data science world where things are large and the compiling takes a lot of steps and so on. 00:26:15.840 --> 00:26:21.820 The wheels that are built for Linux are built for, is it GLib? 00:26:21.820 --> 00:26:22.720 GClib? 00:26:22.720 --> 00:26:23.420 I mean, hold on. 00:26:23.420 --> 00:26:24.600 I'll look over here. 00:26:24.600 --> 00:26:25.420 I wrote it down. 00:26:25.420 --> 00:26:25.960 So I know. 00:26:25.960 --> 00:26:27.000 No, I didn't write it down. 00:26:27.000 --> 00:26:27.140 Sorry. 00:26:27.140 --> 00:26:32.620 There's like, I think it's GLib or GClib, which is the C runtime on like Ubuntu and Debian. 00:26:32.620 --> 00:26:36.220 But there's one MUSL, Muscle, on Alpine. 00:26:36.220 --> 00:26:39.720 And the wheels are not built for Muscle. 00:26:39.820 --> 00:26:41.220 They're built for GClib. 00:26:41.220 --> 00:26:44.020 And so you can't pip install that. 00:26:44.020 --> 00:26:46.600 You've got to download everything and then compile it. 00:26:46.600 --> 00:26:53.460 And it's like compiling matplotlib and Jupyter from scratch can take a really long time versus just downloading the wheel. 00:26:53.460 --> 00:26:54.600 And it takes up a lot of space. 00:26:54.600 --> 00:26:59.680 And there's a bunch of issues and things around that that make it slightly not Python friendly. 00:26:59.680 --> 00:27:08.420 That's why there's this pep, pep656, to allow wheels to be tagged as supporting Muscle, not GClib. 00:27:08.960 --> 00:27:10.400 Is that more than you wanted, Brian? 00:27:10.400 --> 00:27:10.840 Are you good? 00:27:10.840 --> 00:27:11.980 Okay. 00:27:11.980 --> 00:27:18.780 So the takeaway that I'm getting is probably not panic on some of these, but maybe at least pay attention to them. 00:27:18.780 --> 00:27:27.260 And it is good, like you said, to remove tools out of your Docker images that you're not using. 00:27:27.260 --> 00:27:30.140 If you're not using Wget in your application, take it off. 00:27:30.140 --> 00:27:30.900 Things like that. 00:27:31.180 --> 00:27:31.640 Yeah, exactly. 00:27:31.640 --> 00:27:33.280 I think Julia's point was great, right? 00:27:33.280 --> 00:27:35.840 It's if you it might be a false positive. 00:27:35.840 --> 00:27:43.780 But at the same time, if you're not going to use it again, because Docker, a lot of times you pip install all your stuff and then it's kind of ready to run. 00:27:43.780 --> 00:27:46.020 But you're not going to go and pip install something again. 00:27:46.020 --> 00:27:48.480 You're going to do a new Docker build from scratch. 00:27:48.480 --> 00:27:49.000 Right. 00:27:49.000 --> 00:27:55.500 Like one of the final lines could be remove, remove all those intermediate things that could have problems and make it larger and whatnot. 00:27:56.000 --> 00:28:02.000 Yeah, I thought so I've only thought about this from like package from like image size. 00:28:02.000 --> 00:28:02.400 Right. 00:28:02.400 --> 00:28:02.920 Like that. 00:28:02.920 --> 00:28:03.240 Yeah. 00:28:03.240 --> 00:28:07.160 That you want some more images just because it takes forever to get them around. 00:28:07.160 --> 00:28:10.700 But it's interesting to think about from the vulnerability perspective. 00:28:10.700 --> 00:28:17.220 And I've always seen it done as you do whatever installation you need and then you do all these like cleaning steps. 00:28:17.220 --> 00:28:23.240 But what you said, Michael, about like not ever putting certain things on your image was is interesting. 00:28:23.240 --> 00:28:24.960 I haven't heard of that before. 00:28:24.960 --> 00:28:25.860 Yeah, thanks. 00:28:25.860 --> 00:28:31.200 I also had Peter McKee from who works at Docker on Talk Python a little while, like six months ago or something. 00:28:31.200 --> 00:28:36.980 And he talks about having these multi-step builds, something to the effect of doesn't make as much sense with Python. 00:28:36.980 --> 00:28:38.100 I'll try to put it together. 00:28:38.100 --> 00:28:39.740 But like imagine you're building a Go library. 00:28:39.740 --> 00:28:44.560 You could put the Go runtime and build tools on a container, build your thing. 00:28:44.560 --> 00:28:48.320 But the thing you get from Go is an actual binary that's all self-contained. 00:28:48.320 --> 00:28:55.720 You could throw that container away and just copy the output of that into your actual container and never even put all those tools on the 00:28:55.720 --> 00:28:57.160 actual system that goes to production. 00:28:57.160 --> 00:29:04.260 With Python, that might look something like maybe using PEX to package up all the stuff inside of a virtual environment. 00:29:04.260 --> 00:29:08.420 And long as Python, the runtime is there, then you can like PEX run on your other machine. 00:29:08.420 --> 00:29:11.860 But you could potentially not even ever install those, which might be good. 00:29:11.860 --> 00:29:12.820 Yeah, that makes sense. 00:29:12.900 --> 00:29:17.020 There's a lot there that is sort of beyond my comfort level. 00:29:17.020 --> 00:29:20.280 But that's what I thought as I looked at this, Brian. 00:29:20.280 --> 00:29:21.700 Well, thanks for taking a look. 00:29:21.700 --> 00:29:22.320 There you bet. 00:29:22.320 --> 00:29:23.020 All right. 00:29:23.020 --> 00:29:26.680 We'd like to talk about GUIs on the show every now and then. 00:29:26.680 --> 00:29:32.000 And we want to talk about pandas and data frames and data science and all that. 00:29:32.000 --> 00:29:33.180 So let's put those together. 00:29:33.180 --> 00:29:36.980 There's this project over here called Pandas GUI. 00:29:37.280 --> 00:29:41.260 And the documentation is sparse, let's say. 00:29:41.260 --> 00:29:41.800 It's pretty easy. 00:29:41.800 --> 00:29:43.300 There's a couple of examples or two. 00:29:43.300 --> 00:29:46.880 So I could come down here and I could like do my panda stuff and create a data frame. 00:29:46.880 --> 00:29:49.920 And I could just import show from the pandas GUI. 00:29:49.920 --> 00:29:56.660 And within my notebook, it will pop open a separate window that it then allows me to cruise around and check it out. 00:29:56.660 --> 00:30:04.580 So it does, you know, you can print out the data frame in a notebook and you get kind of a static Excel grid looking thing. 00:30:04.580 --> 00:30:05.880 And that's nice. 00:30:05.880 --> 00:30:10.460 But with this, you get a interactive one that lets you sort and select. 00:30:10.460 --> 00:30:15.500 You can actually copy and paste chunks out of there as if it was Excel and then paste it in other places. 00:30:15.500 --> 00:30:18.400 It also has a plotting library with like pictures. 00:30:18.400 --> 00:30:20.840 So I'm going to go click on the bar graph picture. 00:30:20.840 --> 00:30:24.600 And then there's a list of all the columns and the things that the bar graph needs. 00:30:24.600 --> 00:30:29.060 And you can drag and drop this column is the X axis and this column is the Y axis. 00:30:29.060 --> 00:30:35.780 And I want to group by color and have, you know, group by color it by some other aspect of the data. 00:30:35.780 --> 00:30:40.500 And, you know, like group into multiple charts or multiple lines or plots on a chart. 00:30:40.500 --> 00:30:42.080 All sorts of cool stuff like that. 00:30:42.080 --> 00:30:43.520 There's a statistics section. 00:30:43.520 --> 00:30:49.020 There's you can export important export, I guess, import CSV files with drag and drop. 00:30:49.020 --> 00:30:50.860 And there's also search that you can do. 00:30:51.200 --> 00:30:54.860 So it's a pretty neat, quick way to explore pandas. 00:30:54.860 --> 00:30:55.520 Yeah. 00:30:55.520 --> 00:30:57.060 It's a neat idea. 00:30:57.060 --> 00:31:05.600 Like when you, when you first encounter a data frame, like you really want to, you really want to just be able to like look at it without any assumptions. 00:31:05.880 --> 00:31:15.100 And there's a lot of stuff that like kind of goes towards that with like the dot plot API and pandas and making that, making it really accessible to make plots really quickly. 00:31:15.100 --> 00:31:18.020 But this is like kind of like the step beyond that, right? 00:31:18.020 --> 00:31:20.560 Of just visualizing it immediately. 00:31:21.060 --> 00:31:21.160 Yeah. 00:31:21.160 --> 00:31:29.820 Like one thing you get when you view the data frame as, you know, like I said, it looks kind of just like printing DF in or just typing DF in the notebook. 00:31:29.820 --> 00:31:32.360 But then on the right, you can say, oh, I want to see the filters. 00:31:32.360 --> 00:31:38.320 And you could type in these filter expressions, these query expressions, and then turn them all, like pile them on. 00:31:38.320 --> 00:31:41.980 You can have little checkboxes to like optionally turn them off, but not delete them. 00:31:42.120 --> 00:31:45.360 And then of course you can sort within there like that. 00:31:45.360 --> 00:31:49.100 And the graphing, I think the support for the graphing part is really, really helpful. 00:31:49.100 --> 00:31:53.920 So the fact that you can just go and click and say, oh, I want a box plot. 00:31:53.920 --> 00:31:56.280 And then the box plot needs these things. 00:31:56.280 --> 00:32:02.140 You can just drag and drop from the column, from your data frame definition over, and it just live updates. 00:32:02.140 --> 00:32:02.740 Yeah. 00:32:02.740 --> 00:32:11.340 I think that really like lets people visualize the data in the way that they want to sometimes, rather than like the way they already know how in that plot loop. 00:32:11.660 --> 00:32:15.200 Which I think is what people end up doing, at least for exploratory stuff. 00:32:15.200 --> 00:32:16.100 Yeah, exactly. 00:32:16.100 --> 00:32:23.560 You could real quickly switch between a bar, a box, a scatter plot, back and forth without having to actually be familiar with how those works. 00:32:23.560 --> 00:32:29.540 Can you tell if there's a way to export the filters or is there any mechanism for that? 00:32:29.540 --> 00:32:31.440 There is, I don't think so. 00:32:31.440 --> 00:32:36.440 At least in the YouTube explainer video, there were some comments like, you know what would be awesome? 00:32:36.440 --> 00:32:40.920 Export this as code from here so that I can just turn it back into Python. 00:32:41.200 --> 00:32:43.540 I didn't see anything like that. 00:32:43.540 --> 00:32:44.240 Yeah. 00:32:44.240 --> 00:32:47.120 Sometimes GUIs are a little weird for me because of that. 00:32:47.120 --> 00:32:51.680 You know, like you end up in this GUI world and it's not, you can't reproduce anything. 00:32:51.680 --> 00:32:56.840 I clicked on a whole bunch of stuff and then it looked great, but don't touch it. 00:32:56.840 --> 00:32:57.700 Yeah, exactly. 00:32:57.700 --> 00:32:58.660 I can't do it again. 00:33:00.120 --> 00:33:13.240 Okay, but to be fair, it is a fairly quick way to look at the data and know what you, maybe you can't produce that exact plot again, but you know what the data looks like and you can use a different plotting mechanism to do that. 00:33:13.560 --> 00:33:14.040 Yeah. 00:33:14.040 --> 00:33:15.040 And the visually it's pretty clear. 00:33:15.040 --> 00:33:18.660 Like, okay, well, X is assigned to speed and we know it's a histogram. 00:33:18.660 --> 00:33:26.520 And so you could pretty quickly, you know, with some Googling stack overflowing, go, all right, how do I map plot live a histogram and get that going? 00:33:26.520 --> 00:33:26.840 You know? 00:33:26.840 --> 00:33:27.400 Right. 00:33:27.400 --> 00:33:28.560 That's a huge time saver. 00:33:29.240 --> 00:33:29.460 Yeah. 00:33:29.460 --> 00:33:34.320 But some, some, some sort of export of like, okay, give me the code to make this plot in my own code. 00:33:34.320 --> 00:33:34.960 That would be great. 00:33:34.960 --> 00:33:36.520 Yeah, absolutely. 00:33:36.520 --> 00:33:37.840 Absolutely. 00:33:37.840 --> 00:33:38.300 All right. 00:33:38.300 --> 00:33:39.200 On to the next. 00:33:39.200 --> 00:33:45.160 But before we get there, I do want to call out just a shout out by Pylang that FS spec is sweet. 00:33:45.160 --> 00:33:45.660 Good mention. 00:33:45.660 --> 00:33:46.020 Yeah. 00:33:46.020 --> 00:33:47.660 I like it as well. 00:33:47.660 --> 00:33:48.380 Cool. 00:33:48.380 --> 00:33:49.160 All right. 00:33:49.160 --> 00:33:49.840 X-Ray. 00:33:49.840 --> 00:33:50.760 X-Ray. 00:33:50.760 --> 00:33:51.660 Okay. 00:33:51.660 --> 00:33:57.220 So X-Ray is, it's my favorite library. 00:33:57.980 --> 00:33:59.540 It's a, it's like a pandas. 00:33:59.540 --> 00:34:04.520 So it's a pandas like API, but it's for N dimensional data. 00:34:04.520 --> 00:34:21.860 So if you have like a lot of times people talk about in like geospatial data where there's that long time and others, but also for image data where there's maybe a bunch of different bands from like satellite imagery or other disciplines where you just have labeled data. 00:34:21.860 --> 00:34:23.040 That's not tabular. 00:34:23.040 --> 00:34:28.600 So the axes like mean something, but there's not just one or two of them. 00:34:28.980 --> 00:34:39.720 Then X-Ray is like great for that because it lets you do things like you can select a certain subset of time or a certain subset of whatever your dimension is. 00:34:39.720 --> 00:34:42.880 And you can also aggregate across different dimensions. 00:34:43.620 --> 00:34:45.780 And you can use the labels directly. 00:34:45.780 --> 00:35:04.860 So if you don't have a tool like this, I see people doing this a lot with like machine learning workflows where they'll be, they'll have like separate, like a list of all their, they'll have like a list of all their labels and then they'll have their data and they'll do some manipulation and they'll try to like reattach them at the end. 00:35:04.860 --> 00:35:08.540 And it's just, it just turns into a mess. 00:35:08.540 --> 00:35:12.300 And it's actually just like takes care of that all for you. 00:35:12.300 --> 00:35:13.940 It's pretty great. 00:35:13.940 --> 00:35:18.740 And I think that it has applications that have not been fully realized yet. 00:35:18.740 --> 00:35:23.020 And it's starting to like take off in other spaces, but it really comes from this geospatial world. 00:35:23.020 --> 00:35:25.560 But I think it could be useful for all sorts of people. 00:35:25.560 --> 00:35:26.120 Right. 00:35:26.120 --> 00:35:29.660 Because in geospatial, sometimes you have three dimensions, not just two. 00:35:29.660 --> 00:35:30.440 Yeah. 00:35:30.460 --> 00:35:31.980 You almost always have three. 00:35:31.980 --> 00:35:32.520 Right. 00:35:32.520 --> 00:35:33.440 Sorry, Brian. 00:35:33.440 --> 00:35:34.020 Go ahead. 00:35:34.020 --> 00:35:35.560 No, the documentation looks great too. 00:35:35.560 --> 00:35:42.680 The documentation has like getting started guides and tutorials and videos and galleries and stuff. 00:35:42.680 --> 00:35:44.340 So definitely check out the documentation. 00:35:44.340 --> 00:35:45.160 Yeah. 00:35:45.160 --> 00:35:49.920 I think it got a major, it seems like I looked at it for this too, and it seems like it got a major facelift. 00:35:49.920 --> 00:35:51.640 So it looks really nice. 00:35:52.060 --> 00:35:55.000 It also has like plotting. 00:35:55.000 --> 00:36:01.300 It supports the dot plot API or some different version of it that's like the pandas version. 00:36:01.300 --> 00:36:06.320 But you can plot in different, you know, three dimensions or aggregate and then plot. 00:36:06.320 --> 00:36:10.080 And so that's like a really nice way to get the visuals quickly. 00:36:11.080 --> 00:36:23.100 And then the last thing that I wanted to say about it is that it's normally backed by NumPy arrays, but it can also be backed by Dask arrays or Sparse arrays or all sorts of different arrays natively. 00:36:23.100 --> 00:36:32.520 So it's a, it's a really cool, it's another one of these like building block things where you can have X arrays like you're labeling and you're indexing and all the like nice stuff. 00:36:32.520 --> 00:36:38.100 And then down inside it can be NumPy or QPy or Dask. 00:36:38.100 --> 00:36:39.060 How interesting. 00:36:39.060 --> 00:36:45.680 So it's, it can do that juggling and piecing back together that other people are manually doing and you just have this simple API. 00:36:45.680 --> 00:36:47.680 And if it has to do that, it'll figure it out. 00:36:47.680 --> 00:36:48.440 Yeah. 00:36:48.440 --> 00:36:48.820 Yeah. 00:36:48.820 --> 00:36:49.740 That's pretty cool. 00:36:49.740 --> 00:36:50.020 Nice. 00:36:50.020 --> 00:36:52.860 And you talked about QPy and Dask. 00:36:52.860 --> 00:36:55.600 Like those are some pretty interesting backends for this. 00:36:55.600 --> 00:36:56.260 Yeah. 00:36:56.260 --> 00:36:56.820 Yeah. 00:36:56.820 --> 00:37:00.840 The Dask one is, I said QPy. 00:37:00.840 --> 00:37:04.060 And now I'm wondering if maybe it's just like Dask and then QPy. 00:37:04.060 --> 00:37:05.560 So don't quote me on that. 00:37:05.560 --> 00:37:11.380 But, but yeah, the Dask one is, is like really integrated with X-Ray code. 00:37:11.380 --> 00:37:16.580 So you do like, they do just do some special things to make it so that it works with paralyzing and things. 00:37:16.580 --> 00:37:19.100 But, but from the user experience, it's the same. 00:37:19.100 --> 00:37:19.920 Yeah. 00:37:19.920 --> 00:37:20.380 Fantastic. 00:37:20.380 --> 00:37:23.220 And then also noticed it requires Python 3.7. 00:37:23.220 --> 00:37:28.480 Really nice to see tools sort of keeping up with the latest, not, not, not really old stuff. 00:37:28.480 --> 00:37:30.420 Well, hopefully it's 3.7 and above. 00:37:30.860 --> 00:37:31.160 Well, yeah. 00:37:31.160 --> 00:37:31.620 Yeah. 00:37:31.620 --> 00:37:32.420 Greater than or equal to. 00:37:32.420 --> 00:37:35.100 Well, I mean, I ran into a library. 00:37:35.100 --> 00:37:38.960 It was an internal thing that, that was only 3.7. 00:37:38.960 --> 00:37:44.480 So I tried it on, I'm like, I assumed or above and I tried it on 3.9 and it like fell over. 00:37:44.480 --> 00:37:45.440 Like what's going on? 00:37:45.440 --> 00:37:46.960 It was only 3.7. 00:37:46.960 --> 00:37:47.720 It's weird. 00:37:47.720 --> 00:37:49.540 Okay. 00:37:49.540 --> 00:37:50.100 That is weird. 00:37:50.100 --> 00:37:55.760 That'd be interesting to think about what special features of 3.7 there, depending on that broken 3.8. 00:37:56.160 --> 00:37:56.440 Yeah. 00:37:56.440 --> 00:37:57.040 That's what I was thinking. 00:37:57.040 --> 00:38:01.360 Like, how do you do that without just checking for equal, equal 3.7 on version? 00:38:01.360 --> 00:38:02.060 Yeah. 00:38:02.060 --> 00:38:03.260 So anyway. 00:38:03.260 --> 00:38:03.840 Yeah. 00:38:03.840 --> 00:38:04.440 All right. 00:38:04.440 --> 00:38:06.300 Well, that's it for our six main topics. 00:38:06.300 --> 00:38:08.340 Brian, you got anything else you want to throw out there quickly? 00:38:09.220 --> 00:38:09.980 Yeah, actually. 00:38:09.980 --> 00:38:22.480 So I didn't have this up, but there was a, on Twitter, somebody like reacted to me with an emoji and I didn't, didn't know what they meant. 00:38:22.480 --> 00:38:26.440 So I looked up, let me, let me pop this up. 00:38:28.380 --> 00:38:37.800 And it was helpful and you can just, you can just copy and paste the emoji that somebody uses in there and it tells you what it means. 00:38:37.800 --> 00:38:44.320 And the, you know, kind of not just what it's supposed to mean, but also what people are using it for. 00:38:44.320 --> 00:38:49.340 Anyway, for somebody that's sort of an old, old guy that is out of touch sometimes, this was helpful. 00:38:49.340 --> 00:38:50.860 Anyway. 00:38:51.120 --> 00:38:51.340 Yeah. 00:38:51.340 --> 00:38:54.900 I mean, sometimes it's obvious, like a heart, we know what a heart means. 00:38:54.900 --> 00:38:55.280 Right. 00:38:55.280 --> 00:39:00.900 But, you know, like hands together, it's not necessarily that that's like a thank you sort of bow type of thing. 00:39:00.900 --> 00:39:03.040 I mean, there's certain ones where you're like, ah, what does that mean? 00:39:03.040 --> 00:39:06.520 It was like a hands together with like arrows coming out of the top. 00:39:06.520 --> 00:39:10.680 And I'm like, I don't know what this is, but apparently it's just raising hands. 00:39:10.680 --> 00:39:12.740 Like, like you're saying hooray for somebody. 00:39:12.740 --> 00:39:13.540 Oh, okay. 00:39:13.540 --> 00:39:14.400 That's nice. 00:39:14.400 --> 00:39:14.880 So. 00:39:14.880 --> 00:39:15.200 Okay. 00:39:15.200 --> 00:39:15.760 It's good. 00:39:15.760 --> 00:39:18.820 I use Emojipedia all the time, but I think I use it in the opposite way. 00:39:18.820 --> 00:39:24.100 Like I use it to get an emoji to like put somewhere because I don't have like an emoji keyboard or whatever. 00:39:24.100 --> 00:39:25.000 Oh yeah. 00:39:25.000 --> 00:39:25.700 That would be good too. 00:39:25.700 --> 00:39:38.040 The other thing I wanted to bring up is I hopefully have some cool news to share tomorrow about the pytestBook and the news will show up on a revamped pytestBook site. 00:39:38.040 --> 00:39:48.300 So if you go to pytestBook.com, you get redirected to this Pythontest.com page where I'll talk about the second edition. 00:39:48.680 --> 00:39:51.740 Hopefully there'll be news about the second edition coming out tomorrow. 00:39:51.740 --> 00:39:54.600 Is your new static site magic? 00:39:54.600 --> 00:39:55.700 Yeah. 00:39:55.700 --> 00:39:56.160 Yeah. 00:39:56.160 --> 00:39:56.800 Static site. 00:39:56.800 --> 00:39:58.640 And I totally, and it goes dark and light. 00:39:58.640 --> 00:40:01.720 But I totally stole from Pragyun. 00:40:01.720 --> 00:40:06.280 So Pragyun has the same, he's got a really nice site. 00:40:06.280 --> 00:40:08.540 So it's a bunch of great, great. 00:40:08.540 --> 00:40:09.220 It looked great. 00:40:09.220 --> 00:40:10.260 And I'm like, that'll work. 00:40:10.260 --> 00:40:11.320 I'll just do what he's doing. 00:40:11.320 --> 00:40:12.680 So that's what I did. 00:40:12.680 --> 00:40:13.140 Yeah. 00:40:13.140 --> 00:40:13.660 Yeah. 00:40:13.660 --> 00:40:14.000 Very cool. 00:40:14.000 --> 00:40:17.800 I think we have exactly the same stack for our Saturn Cloud site now. 00:40:17.800 --> 00:40:18.900 Oh, how neat. 00:40:18.900 --> 00:40:19.360 So it's cool. 00:40:19.360 --> 00:40:20.380 Awesome. 00:40:20.380 --> 00:40:21.140 How about you, Julie? 00:40:21.140 --> 00:40:22.560 Anything else you want to give a shout out to? 00:40:22.560 --> 00:40:26.480 Well, I've been really into entry points recently. 00:40:26.480 --> 00:40:29.600 Just like the concept of them is very cool. 00:40:29.680 --> 00:40:34.220 As in like Python packages, you can give them almost like CLI command type entry points? 00:40:34.220 --> 00:40:34.900 Yeah. 00:40:34.900 --> 00:40:39.480 But the thing that I think is really cool is like, like, like Matplotlib. 00:40:39.480 --> 00:40:43.520 This is an example that, that made me first realize about entry points is Matplotlib has 00:40:43.520 --> 00:40:44.200 this .plot. 00:40:44.200 --> 00:40:45.940 I think I mentioned this three times now. 00:40:45.940 --> 00:40:47.820 But you can swap out the backend. 00:40:47.820 --> 00:40:49.080 So you don't have to have Matplotlib. 00:40:49.080 --> 00:40:50.660 You can use other backends. 00:40:50.660 --> 00:40:57.180 And all the logic for that is in the other visualization libraries themselves, not in 00:40:57.180 --> 00:40:57.780 Pandas. 00:40:57.780 --> 00:41:02.320 So it's, it's just like, you can swap out other things. 00:41:02.320 --> 00:41:03.720 It's not just for CLIs. 00:41:03.720 --> 00:41:04.820 Okay. 00:41:04.820 --> 00:41:05.400 Yeah. 00:41:05.400 --> 00:41:05.860 How neat. 00:41:05.860 --> 00:41:06.260 All right. 00:41:06.260 --> 00:41:06.440 Yeah. 00:41:06.440 --> 00:41:09.340 I learned about entry points a year, year and a half ago. 00:41:09.340 --> 00:41:11.080 And ever since I'm like, oh yeah, this is awesome. 00:41:11.080 --> 00:41:14.100 I can now create these little commands that'll be part of just my shell. 00:41:14.100 --> 00:41:14.740 I love it. 00:41:14.740 --> 00:41:15.320 Yeah. 00:41:15.320 --> 00:41:18.280 The other thing I wanted to say was GitHub CLI is really cool. 00:41:18.280 --> 00:41:22.360 I think that's standalone, but it's, I've been using it a lot. 00:41:22.360 --> 00:41:27.020 I'm sure people know the Git CLI, but what's the story of the GitHub CLI? 00:41:27.020 --> 00:41:34.160 Oh, well, the GitHub CLI is, makes it, so if you have ever tried to check out a branch on 00:41:34.160 --> 00:41:38.960 someone else's fork, like if you want to like evaluate a PR that someone has put on a fork. 00:41:38.960 --> 00:41:39.400 Yeah, exactly. 00:41:39.400 --> 00:41:39.760 Yeah. 00:41:39.760 --> 00:41:44.380 That is the situation where the GitHub CLI is really great because you can just do like 00:41:44.380 --> 00:41:51.140 GH checkout PR or a GH PR checkout, whatever the number is, and that you're just on their 00:41:51.140 --> 00:41:51.780 branch then. 00:41:51.780 --> 00:41:55.840 And if you can push, if you have push access to their branch, if you're a maintainer and 00:41:55.840 --> 00:41:58.080 they've allowed it, you can just push directly. 00:41:58.080 --> 00:42:03.840 And you don't, I mean, I was always looking at that sequence of commands before, like I 00:42:03.840 --> 00:42:08.120 know people have like Git aliases and stuff, but yeah, I'd really recommend checking it 00:42:08.120 --> 00:42:09.420 out if you do a lot of GitHub stuff. 00:42:09.420 --> 00:42:09.940 Okay. 00:42:09.940 --> 00:42:10.260 Awesome. 00:42:10.460 --> 00:42:10.580 Yeah. 00:42:10.580 --> 00:42:11.340 That's great advice. 00:42:11.340 --> 00:42:11.900 Yeah. 00:42:11.900 --> 00:42:15.640 I often want to like check out some, so a pull request, I want to be able to like play with 00:42:15.640 --> 00:42:16.520 it and run their code. 00:42:16.520 --> 00:42:17.100 And yeah. 00:42:17.100 --> 00:42:18.420 And so, yeah. 00:42:18.420 --> 00:42:19.320 It's the best. 00:42:19.320 --> 00:42:20.360 Yeah. 00:42:20.360 --> 00:42:20.800 Awesome. 00:42:20.800 --> 00:42:21.340 All right. 00:42:21.340 --> 00:42:24.800 I got a couple of things to add, by the way, first of all, just that first practical SQL 00:42:24.800 --> 00:42:26.260 analysis that you talked about. 00:42:26.260 --> 00:42:29.780 It also is a similar theme that you were talking about, Brian. 00:42:29.780 --> 00:42:33.140 One of the things I thought was cool though, as you scroll through it, it has a progress bar 00:42:33.140 --> 00:42:34.280 for reading at the top. 00:42:34.280 --> 00:42:35.440 And that just made me so happy. 00:42:35.440 --> 00:42:38.000 I don't know why that was, that was really neat. 00:42:38.300 --> 00:42:38.500 All right. 00:42:38.500 --> 00:42:40.600 But I have a bunch of hear all about it sort of things. 00:42:40.600 --> 00:42:44.540 So really quick, Python, B2, I just got the center. 00:42:44.540 --> 00:42:44.740 Yeah. 00:42:44.740 --> 00:42:45.100 Okay. 00:42:45.100 --> 00:42:46.360 Live update. 00:42:46.360 --> 00:42:50.520 Python 310 beta 2 is out if people want to check that out. 00:42:50.520 --> 00:42:52.560 And you can go download that. 00:42:52.560 --> 00:42:59.060 It also highlights all the major features like the pipe operator for writing unions and type 00:42:59.060 --> 00:43:03.240 specifications and a bunch of other stuff that people might care about. 00:43:03.240 --> 00:43:04.620 Structure pattern matching. 00:43:04.620 --> 00:43:05.760 It's probably a big one. 00:43:06.120 --> 00:43:06.260 Yeah. 00:43:06.260 --> 00:43:07.940 Go to the completely different down. 00:43:07.940 --> 00:43:09.200 Is that on here? 00:43:09.200 --> 00:43:10.840 And now for something completely different. 00:43:10.840 --> 00:43:11.540 I love that part. 00:43:11.540 --> 00:43:12.700 So right above the files. 00:43:12.700 --> 00:43:14.200 Yeah. 00:43:14.200 --> 00:43:16.040 Oh, interesting. 00:43:16.040 --> 00:43:21.960 The Aaron Fest paradox concerns the rotation of a rigid disk in the theory of relativity. 00:43:21.960 --> 00:43:24.840 It's original 1909 formulation presented by. 00:43:24.840 --> 00:43:25.280 Yeah. 00:43:25.280 --> 00:43:25.500 Okay. 00:43:25.500 --> 00:43:27.760 That is unexpected, but very cool. 00:43:27.760 --> 00:43:29.660 And completely different and irrelevant. 00:43:29.660 --> 00:43:30.160 Yeah. 00:43:30.160 --> 00:43:30.720 Yeah. 00:43:30.720 --> 00:43:31.060 Awesome. 00:43:31.180 --> 00:43:31.380 Okay. 00:43:31.380 --> 00:43:34.340 So takeaway 310 beta 2 is out. 00:43:34.340 --> 00:43:35.160 People can check that out. 00:43:35.160 --> 00:43:37.840 There's also some security patches for Django. 00:43:37.840 --> 00:43:38.980 So be sure to check that out. 00:43:38.980 --> 00:43:45.120 One thing that surprised me is the Microsoft install Python from the Windows store is already 00:43:45.120 --> 00:43:49.140 like has a 310 beta store install. 00:43:49.140 --> 00:43:50.280 So, okay. 00:43:50.280 --> 00:43:52.660 That's pretty cool that they're keeping that up to date. 00:43:52.660 --> 00:43:54.160 And it's rated E for everyone. 00:43:54.540 --> 00:43:54.760 Yeah. 00:43:54.760 --> 00:43:56.700 Even kids can pip install. 00:43:56.700 --> 00:43:57.120 Awesome. 00:43:57.120 --> 00:44:03.900 So Frederick Bankston sent a message in response to our last show where we talked about the 00:44:03.900 --> 00:44:05.760 method overloading by type. 00:44:05.760 --> 00:44:08.520 Like if it takes an int or a string, it calls different functions. 00:44:08.520 --> 00:44:13.020 It's also pointed us towards this multi-method other library that is similar. 00:44:13.020 --> 00:44:14.120 So people can check that out. 00:44:14.120 --> 00:44:14.520 That's cool. 00:44:14.520 --> 00:44:14.720 Yeah. 00:44:14.720 --> 00:44:15.200 Neat. 00:44:15.400 --> 00:44:22.120 Speaking of the GitHub stuff, I've been starting to use PyCharm 2021 to early access 00:44:22.120 --> 00:44:24.180 version, early access program version one. 00:44:24.180 --> 00:44:25.260 And it's been working fine. 00:44:25.260 --> 00:44:28.180 So if people want to try out the new features, there's a bunch of cool stuff. 00:44:28.180 --> 00:44:32.100 You have support for Python 310 and new stuff for pytest. 00:44:32.100 --> 00:44:38.080 I don't remember if this came in here, but one thing that I did learn about that recently 00:44:38.080 --> 00:44:44.500 that's in there that's super cool is they have in PyCharm, if you log in PyCharm into 00:44:44.500 --> 00:44:48.280 your GitHub account, there's a pull request section and you can just click it and it'll 00:44:48.280 --> 00:44:50.480 do those same steps that Julia was talking about. 00:44:50.480 --> 00:44:55.200 Like right there in PyCharm, just go, I want to try that PR before I accept it and just click 00:44:55.200 --> 00:44:55.700 that and go. 00:44:55.700 --> 00:44:57.260 You can even have comments. 00:44:57.260 --> 00:44:59.300 You see the conversation inside there and everything. 00:44:59.300 --> 00:44:59.680 It's cool. 00:44:59.680 --> 00:45:01.760 Never go to GitHub again. 00:45:01.760 --> 00:45:02.860 Exactly. 00:45:02.860 --> 00:45:05.040 And don't just forget how to use it basically. 00:45:05.040 --> 00:45:05.780 All right. 00:45:05.780 --> 00:45:07.200 That's it. 00:45:07.320 --> 00:45:08.340 That's all the items I got. 00:45:08.340 --> 00:45:11.280 So yeah, I've got other stuff that's just hanging around from before. 00:45:11.280 --> 00:45:11.780 Cool. 00:45:11.780 --> 00:45:12.680 All right. 00:45:12.680 --> 00:45:14.200 Well, you want to close it out with a joke? 00:45:14.200 --> 00:45:14.760 Yeah. 00:45:14.760 --> 00:45:15.700 A couple of jokes. 00:45:15.700 --> 00:45:16.340 Always. 00:45:16.340 --> 00:45:16.800 All right. 00:45:16.800 --> 00:45:22.080 So over at upjoke.com slash programmer to ask jokes, you'll find many bad jokes. 00:45:22.080 --> 00:45:26.340 Some even that are not very appropriate or whatever, but there's a few that are funny. 00:45:26.340 --> 00:45:27.940 So I pulled out three here. 00:45:27.940 --> 00:45:29.580 I'll do the first one. 00:45:29.580 --> 00:45:31.300 Brian, you can do the second. 00:45:31.300 --> 00:45:33.120 Julie, you can do the third, I guess, if you're up for it. 00:45:33.120 --> 00:45:33.580 Okay. 00:45:33.580 --> 00:45:36.520 So this one we should have saved for six months from now. 00:45:36.560 --> 00:45:39.580 But I asked a programmer what her new year's resolution would be. 00:45:39.580 --> 00:45:41.480 She answered 1920 by 1080. 00:45:41.480 --> 00:45:42.820 That's so bad. 00:45:42.820 --> 00:45:44.460 No, that's awesome. 00:45:44.460 --> 00:45:45.560 It's really bad. 00:45:45.560 --> 00:45:45.780 All right. 00:45:45.780 --> 00:45:46.920 Well, you got to do the next one. 00:45:46.920 --> 00:45:51.940 How does a programmer confuse a mathematician? 00:45:51.940 --> 00:45:53.220 I don't know how. 00:45:53.220 --> 00:45:55.580 Just saying that X equals X plus one. 00:45:55.580 --> 00:45:59.700 All right, Julia. 00:46:00.180 --> 00:46:00.620 Okay. 00:46:00.620 --> 00:46:03.960 Why do Python programmers have low self-esteem? 00:46:03.960 --> 00:46:06.880 They're constantly comparing their self to other. 00:46:06.880 --> 00:46:10.780 Also bad. 00:46:10.780 --> 00:46:11.720 Probably the worst. 00:46:11.720 --> 00:46:12.660 Sorry we gave you that one. 00:46:12.660 --> 00:46:13.860 That's okay. 00:46:13.860 --> 00:46:15.500 I saw this. 00:46:15.500 --> 00:46:19.240 I saw the one that Brian did and I was like, oh, it should be X plus equals one. 00:46:19.300 --> 00:46:20.680 And I was like, no, that ruins the joke. 00:46:20.680 --> 00:46:22.820 Exactly. 00:46:22.820 --> 00:46:24.160 Yeah. 00:46:24.160 --> 00:46:25.880 Yeah. 00:46:25.880 --> 00:46:30.800 I actually often do the slow way or the non-obvious way. 00:46:30.800 --> 00:46:30.900 The proposed way. 00:46:30.900 --> 00:46:31.160 Yeah. 00:46:31.160 --> 00:46:35.920 X equals X plus one just to make it more obvious to people reading it sometimes. 00:46:35.920 --> 00:46:36.260 Yeah. 00:46:36.260 --> 00:46:36.840 Yeah. 00:46:36.840 --> 00:46:37.640 No, I agree. 00:46:37.640 --> 00:46:38.320 Yeah. 00:46:38.320 --> 00:46:41.740 At least it's not C++ with X, plus plus X. 00:46:41.740 --> 00:46:43.540 I love that. 00:46:43.540 --> 00:46:44.640 No, no. 00:46:44.900 --> 00:46:46.280 We should have that. 00:46:46.280 --> 00:46:50.800 I'm okay with X plus plus, but not that also plus plus X. 00:46:50.800 --> 00:46:51.680 Oh, the pre-increment. 00:46:51.680 --> 00:46:52.240 Yeah. 00:46:52.240 --> 00:46:53.180 The pre-increment. 00:46:53.180 --> 00:46:53.800 The slight. 00:46:53.800 --> 00:46:54.380 That's weird. 00:46:54.380 --> 00:46:55.140 Yes. 00:46:55.140 --> 00:46:55.840 Exactly. 00:46:55.840 --> 00:46:56.540 Exactly. 00:46:56.540 --> 00:46:57.280 But I could go for it. 00:46:57.280 --> 00:46:57.880 X plus plus. 00:46:57.880 --> 00:46:58.220 Come on. 00:46:58.220 --> 00:46:58.820 All right. 00:46:58.820 --> 00:47:01.200 Well, Julia, thanks for joining us this week. 00:47:01.200 --> 00:47:02.420 And Brian, thanks as always. 00:47:02.420 --> 00:47:03.280 Oh, it was a pleasure. 00:47:03.280 --> 00:47:03.920 Thanks, Julia. 00:47:03.920 --> 00:47:04.500 Yeah. 00:47:04.500 --> 00:47:04.600 Bye. 00:47:04.600 --> 00:47:04.740 Bye.