WEBVTT 00:00:00.020 --> 00:00:04.060 Hello and welcome to Python Bytes, where we deliver Python news and headlines directly 00:00:04.570 --> 00:00:05.180 to your earbuds. 00:00:05.580 --> 00:00:07.060 This is episode 482, 00:00:07.220 --> 00:00:08.700 recorded Monday, 00:00:09.130 --> 00:00:09.840 June 1st. 00:00:09.940 --> 00:00:10.420 I'm Michael Kennedy. 00:00:10.940 --> 00:00:11.820 And I'm Brian Okken. 00:00:12.000 --> 00:00:15.460 This episode is brought to you by us and all of the Python things that we're doing. 00:00:16.040 --> 00:00:20.480 Check out pythontest.com, get Brian's multiple courses over there, 00:00:20.760 --> 00:00:22.280 even something to do with 00:00:22.600 --> 00:00:23.820 Lean TDD, 00:00:24.090 --> 00:00:25.040 if you want to check that out. 00:00:25.130 --> 00:00:27.140 I hear that an audio book is underway, 00:00:27.340 --> 00:00:28.200 which is super cool. 00:00:28.500 --> 00:00:31.640 and got a couple new courses over at Talk Python, 00:00:32.200 --> 00:00:35.780 especially the new OWASP top 10 security plus agentic AI. 00:00:36.060 --> 00:00:38.720 Like how do you take Claude Code or something like that? 00:00:39.120 --> 00:00:42.080 Plus known recognized patterns like the OWASP top 10 00:00:42.220 --> 00:00:43.420 and make your code safe. 00:00:43.680 --> 00:00:44.660 That's what that course is about. 00:00:44.970 --> 00:00:46.060 So check out those kinds of things. 00:00:46.280 --> 00:00:48.080 That's what we're bringing you this week. 00:00:48.480 --> 00:00:48.700 Also, 00:00:49.240 --> 00:00:51.320 if you want to connect with us on socials, 00:00:51.800 --> 00:00:51.960 share, 00:00:52.400 --> 00:00:52.840 show ideas, 00:00:53.100 --> 00:00:53.700 whatever you can do. 00:00:53.700 --> 00:00:54.880 So on all the socials, 00:00:55.180 --> 00:00:55.920 sign up for the newsletter. 00:00:56.100 --> 00:00:58.360 We got awesome extras that come in through there. 00:00:58.460 --> 00:01:00.120 So just do that on the website. 00:01:00.740 --> 00:01:01.040 With that, 00:01:01.700 --> 00:01:01.880 Brian, 00:01:02.500 --> 00:01:03.320 I think it might be time. 00:01:03.640 --> 00:01:04.559 Should we kick it off? 00:01:04.739 --> 00:01:05.600 Speaking of security. 00:01:05.940 --> 00:01:07.180 We have a security issue, 00:01:07.520 --> 00:01:09.740 but the issue I think is resolved. 00:01:09.960 --> 00:01:14.940 We just, I'm highlighting an article that's about the maintainer. 00:01:15.000 --> 00:01:16.840 So this is interesting. 00:01:16.980 --> 00:01:18.960 So this is the CVE, 00:01:19.140 --> 00:01:20.440 I'm not going to, let's see. 00:01:20.680 --> 00:01:21.980 CVE 2026, 00:01:22.780 --> 00:01:25.280 48710, a maintainer's perspective. 00:01:25.640 --> 00:01:30.140 So this is from Marcelo Trilasinski. 00:01:30.640 --> 00:01:30.940 Cool name. 00:01:31.320 --> 00:01:37.740 Anyway, this is about a security advisory from Starlet. 00:01:38.200 --> 00:01:39.200 So even, 00:01:39.510 --> 00:01:40.380 so there's a, what, 00:01:40.580 --> 00:01:43.080 isn't FastAPI used Starlet, I think? 00:01:43.400 --> 00:01:44.040 I do believe so, 00:01:44.280 --> 00:01:44.340 yeah. 00:01:44.740 --> 00:01:46.020 And a bunch of other stuff too. 00:01:46.090 --> 00:01:49.460 So even if you don't think that you're using Starlet, you might be using Starlet. 00:01:49.690 --> 00:01:55.260 So the short version is you probably ought to upgrade to 1.0.1 if you haven't already. 00:01:55.420 --> 00:02:00.680 um the that's one of the versions if if you don't care about the like the nitty-gritty of this just 00:02:00.710 --> 00:02:06.640 do that the other part is an interesting thing about basically that this is an interesting article 00:02:06.840 --> 00:02:11.760 it's kind of hard to cover but i wanted to bring it up because it does uh we do have more and more 00:02:11.960 --> 00:02:18.339 ai things like attack both both trying to attack websites and also so we have to keep up on security 00:02:18.360 --> 00:02:18.560 updates, 00:02:19.020 --> 00:02:25.600 but also a lot of these security vulnerability reports against projects. 00:02:25.900 --> 00:02:26.140 And 00:02:26.580 --> 00:02:29.200 a lot of these projects like Starlette are not, 00:02:29.360 --> 00:02:30.420 they're not a commercial enterprise. 00:02:30.760 --> 00:02:30.960 It's an 00:02:30.980 --> 00:02:33.300 open source project with volunteer maintainers. 00:02:33.620 --> 00:02:35.520 So that's really what we're talking about here 00:02:35.740 --> 00:02:41.140 is a little bit of that. And also the problem really wasn't with Starlet. And apparently 00:02:41.760 --> 00:02:46.340 they got some bad press in a couple of places because of a vulnerability. 00:02:46.780 --> 00:02:47.140 But this, 00:02:47.360 --> 00:02:47.720 I'm going 00:02:47.520 --> 00:02:52.740 read this interesting interesting thing the vulnerability came came from a pattern an 00:02:52.920 --> 00:02:58.300 application pattern and the deployment never from something starlet intended and um there's a 00:02:58.400 --> 00:03:05.120 description in the in the link about what the issue is and it i actually got lost um so it's 00:03:05.180 --> 00:03:10.300 something about like yeah i'm not even going to try to summarize it because i got lost here um 00:03:10.860 --> 00:03:15.120 But there's one description of the bug. 00:03:15.860 --> 00:03:19.440 It says from Ostif, I don't know what Ostif is, 00:03:19.680 --> 00:03:20.460 but an Ostif article, 00:03:20.660 --> 00:03:21.900 this bug is a classic 00:03:22.340 --> 00:03:25.120 responsibility gap where this maintainer didn't patch, 00:03:25.440 --> 00:03:27.520 where if this maintainer didn't patch, 00:03:27.820 --> 00:03:33.280 then thousands of exposed projects would have to individually secure their projects. 00:03:33.720 --> 00:03:39.159 In doing this work, they've voluntarily taken on the responsibility to protect the ecosystem 00:03:39.180 --> 00:03:41.000 from long-term systemic harm. 00:03:42.780 --> 00:03:44.040 As with all open source projects, 00:03:44.610 --> 00:03:45.700 they owed us nothing 00:03:45.850 --> 00:03:48.100 and could have left it for everyone else's problem 00:03:48.490 --> 00:03:48.960 and took, 00:03:49.460 --> 00:03:50.780 but they took the extraordinary steps 00:03:50.810 --> 00:03:52.000 of helping the ecosystem. 00:03:53.040 --> 00:03:55.100 Please consider donating to Glutex, 00:03:55.130 --> 00:03:57.740 which is the person that wrote this article, 00:03:58.450 --> 00:03:59.320 which is an interesting And 00:03:59.320 --> 00:04:01.180 I believe the maintainer of Starlette as well. 00:04:01.940 --> 00:04:02.080 Okay, 00:04:02.230 --> 00:04:02.400 yeah, 00:04:02.560 --> 00:04:04.240 you're right. And also, 00:04:04.990 --> 00:04:08.740 so apparently Ars Technica even covered it. 00:04:08.840 --> 00:04:10.620 And we've referenced the articles a lot, 00:04:10.880 --> 00:04:11.700 you know, in the past, 00:04:12.460 --> 00:04:13.300 sometimes some good stuff. 00:04:13.520 --> 00:04:15.620 But apparently they reached out and said, 00:04:15.780 --> 00:04:16.000 you know, 00:04:17.060 --> 00:04:17.299 you know, 00:04:17.600 --> 00:04:18.820 this seems to be serious. 00:04:19.120 --> 00:04:19.859 Do you have any comments? 00:04:21.000 --> 00:04:21.680 How severe is it? 00:04:21.959 --> 00:04:22.580 And they didn't. 00:04:22.720 --> 00:04:25.600 They sent that request out two hours before publication. 00:04:25.960 --> 00:04:27.360 And since they didn't get a response, 00:04:27.580 --> 00:04:29.260 they posted that. 00:04:29.660 --> 00:04:30.040 They didn't. 00:04:30.220 --> 00:04:30.740 They asked, 00:04:30.900 --> 00:04:32.360 but didn't get a reply from Starlette. 00:04:32.680 --> 00:04:34.020 And two hours? 00:04:35.380 --> 00:04:35.640 Anyway. 00:04:36.160 --> 00:04:39.900 The article goes on to talk about the burden of triaging, 00:04:40.370 --> 00:04:42.200 how many security advisories come in, 00:04:42.430 --> 00:04:43.540 all the work that's involved. 00:04:43.670 --> 00:04:56.140 And also that places like Ars Technica and others treat open source projects as if they're a for-profit corporation and expect like a PR department and stuff. 00:04:56.210 --> 00:04:56.960 And there's not. 00:04:57.260 --> 00:04:58.020 It's just people. 00:04:58.500 --> 00:04:59.060 So anyway. 00:05:00.040 --> 00:05:00.160 Yeah. 00:05:00.460 --> 00:05:03.200 Do you have anything else to add to this or just interesting? 00:05:03.560 --> 00:05:04.380 Yeah, it's just interesting. 00:05:04.670 --> 00:05:05.180 I don't know. 00:05:06.080 --> 00:05:08.520 It doesn't look like a super dangerous vulnerability, 00:05:09.040 --> 00:05:09.600 I don't think, 00:05:09.920 --> 00:05:11.480 but definitely go out and patch. 00:05:11.840 --> 00:05:14.820 And it's a 1.0.0 to 1.0.1 bump. 00:05:15.260 --> 00:05:17.380 So it's not like you've got to jump a major version. 00:05:17.640 --> 00:05:18.840 Yeah, definitely patch. 00:05:18.900 --> 00:05:19.280 But also, 00:05:19.780 --> 00:05:20.280 I think it's, 00:05:21.020 --> 00:05:26.680 I don't know how to get out of this, but because we have a lot of these open source projects that big companies are relying on, 00:05:27.060 --> 00:05:30.860 and they want the reaction times of a paid product. 00:05:31.080 --> 00:05:31.760 And they just, 00:05:32.100 --> 00:05:32.180 yeah. 00:05:32.360 --> 00:05:32.700 And actually. 00:05:32.780 --> 00:05:33.160 You know what? 00:05:33.300 --> 00:05:37.720 They could do that by maybe putting the maintainer on a healthy retainer, 00:05:38.020 --> 00:05:40.420 making the project sustainable and say, 00:05:40.780 --> 00:05:44.120 look, we'll pay you $2,000 a month and we probably won't ask you anything for a long time. 00:05:44.210 --> 00:05:47.200 But if we need you, we need you to jump in and we'll then pay you out earlier. 00:05:47.300 --> 00:05:48.200 I don't know, something like that, right? 00:05:48.360 --> 00:05:48.600 That would, 00:05:48.850 --> 00:05:50.340 if you get five companies doing that, 00:05:50.530 --> 00:05:52.220 I bet you can get some response time out of that. 00:05:52.580 --> 00:05:52.760 Yeah. 00:05:53.120 --> 00:05:58.300 But the ironic reverse thing is actual paid products are really slow to patch anything. 00:05:58.820 --> 00:06:02.560 So we actually get usually faster response times from open source projects. 00:06:02.920 --> 00:06:03.500 So anyway. 00:06:04.340 --> 00:06:04.740 Indeed, 00:06:05.120 --> 00:06:05.440 indeed. 00:06:05.920 --> 00:06:11.540 All right, let's go and chat a bit about the Daily Stars Explorer. 00:06:11.680 --> 00:06:12.460 This is a fun project, 00:06:12.640 --> 00:06:12.780 Brian. 00:06:13.020 --> 00:06:17.220 So what this is, this is a web app that's open source on GitHub. 00:06:17.640 --> 00:06:18.180 What is it built with? 00:06:18.600 --> 00:06:18.760 Scrolling. 00:06:19.120 --> 00:06:20.460 It's built with JavaScript and Go. 00:06:20.860 --> 00:06:20.960 So, 00:06:21.360 --> 00:06:23.580 but it doesn't matter because it's not about the internals. 00:06:23.680 --> 00:06:25.700 It's about the information you get about different projects, 00:06:26.060 --> 00:06:27.440 including Python ones. 00:06:27.740 --> 00:06:31.120 So it's an app that you can put in a repository, 00:06:31.670 --> 00:06:37.800 and it will give you just basically historical information about the stars sliced and diced in all these statistical ways. 00:06:38.200 --> 00:06:38.380 Ooh, 00:06:38.600 --> 00:06:38.720 neat. 00:06:39.280 --> 00:06:41.160 So it has full star history, 00:06:41.480 --> 00:06:42.220 hourly stars. 00:06:42.290 --> 00:06:44.400 You can compare side-by-side repos, 00:06:44.780 --> 00:06:47.740 activity timelines of like show me the commits over time, 00:06:48.050 --> 00:06:48.880 which is kind of interesting. 00:06:49.340 --> 00:06:51.300 You can favorite them and so on. 00:06:51.380 --> 00:06:54.680 You can see when the repos were mentioned on Hacker News, 00:06:54.860 --> 00:06:55.040 Reddit, 00:06:55.200 --> 00:06:55.380 YouTube, 00:06:55.720 --> 00:06:56.220 and GitHub, 00:06:56.860 --> 00:07:00.860 and you download the data for feeding to your AI or feeding to your data science stack. 00:07:01.310 --> 00:07:02.340 So I think this is pretty cool. 00:07:02.580 --> 00:07:05.140 And let's just do this by, 00:07:05.320 --> 00:07:08.020 let me first find the Flask repo. 00:07:08.250 --> 00:07:08.780 One second. 00:07:09.260 --> 00:07:11.380 All right, so let's go and open up the little demo. 00:07:11.800 --> 00:07:14.280 So it's a web app, which means you have to self-host it. 00:07:14.360 --> 00:07:14.620 However, 00:07:15.240 --> 00:07:21.540 if you just want to try it out, I think just use the demo and put whatever you want in there. 00:07:21.690 --> 00:07:23.000 Unless you want to save your history and stuff. 00:07:23.100 --> 00:07:24.280 But let's go over and put Flask in. 00:07:24.620 --> 00:07:25.480 And it thinks for a moment. 00:07:25.970 --> 00:07:28.340 The first time you get one in there, I put a repo in there. 00:07:28.380 --> 00:07:30.580 It takes a minute to download all the data from GitHub. 00:07:30.940 --> 00:07:32.500 But because I put that in there before, 00:07:32.700 --> 00:07:34.260 it kind of goes more quickly. 00:07:34.900 --> 00:07:38.540 So you can see right now Flask has 71,598 stars. 00:07:39.200 --> 00:07:40.780 And you can have different themes, 00:07:40.980 --> 00:07:42.280 of course, for the thing. 00:07:42.620 --> 00:07:45.100 It shows you the total stars over time. 00:07:45.660 --> 00:07:47.480 So you've got this graph, 00:07:47.640 --> 00:07:49.760 this orange line, if you pull it up, 00:07:50.120 --> 00:07:53.820 is the integral or something like that. 00:07:53.820 --> 00:07:56.280 And here's the stars per day, 00:07:56.850 --> 00:07:57.280 like 13, 00:07:57.460 --> 00:07:57.680 15, 00:07:57.900 --> 00:07:57.980 whatever. 00:07:58.210 --> 00:07:59.660 And then the scales are different. 00:08:00.370 --> 00:08:02.320 One goes up to 80,000, another goes to 30. 00:08:02.760 --> 00:08:04.980 But you take the integral of the top curve, you get the bottom curve. 00:08:05.460 --> 00:08:06.700 Sum up all the stars. 00:08:07.200 --> 00:08:08.280 It's very cool. 00:08:08.280 --> 00:08:10.120 You can break it down by over five years, 00:08:10.360 --> 00:08:10.900 quarterly, 00:08:11.140 --> 00:08:11.800 years a day. 00:08:12.020 --> 00:08:15.560 you can go and say, I want to transform this by, I want to just look at trend lines, 00:08:16.240 --> 00:08:19.180 or I want to look at monthly bend stars, 00:08:19.700 --> 00:08:22.440 or maybe I want to normalize the stars across 00:08:22.780 --> 00:08:23.460 and have like a little, 00:08:24.320 --> 00:08:24.500 you know, 00:08:24.820 --> 00:08:25.060 lines. 00:08:25.340 --> 00:08:25.580 What else? 00:08:25.740 --> 00:08:28.800 We've got the running median and week over week growth. 00:08:29.740 --> 00:08:30.000 Neat, right? 00:08:30.320 --> 00:08:31.740 Shows you how old the repository is. 00:08:31.800 --> 00:08:32.740 16 years, one month, 00:08:32.940 --> 00:08:33.620 26 days, 00:08:34.200 --> 00:08:36.080 68 stars the last day. 00:08:36.460 --> 00:08:36.880 So I don't know. 00:08:36.979 --> 00:08:43.919 A lot of times you maybe want to think about what project do I actually want to work on or base my project upon? 00:08:44.440 --> 00:08:46.160 And obviously stars are not everything, 00:08:46.710 --> 00:08:48.640 but it's certainly something. 00:08:49.080 --> 00:08:50.680 Could we pull up here? Come on now, get away. 00:08:51.500 --> 00:08:53.160 You're right about this auto-hide stuff, 00:08:53.320 --> 00:08:53.400 right? 00:08:53.700 --> 00:08:56.160 So where can I find? 00:08:56.530 --> 00:09:00.580 Yeah, I don't know where to pull up the commits from this one right off the top of my head. 00:09:00.590 --> 00:09:05.440 But anyway, it'd be cool to see the commits because that would let you answer some interesting questions as well, 00:09:05.820 --> 00:09:05.920 right? 00:09:05.980 --> 00:09:09.040 Like if you see the commits just come to a stop two years ago, you're like, 00:09:09.200 --> 00:09:09.340 oh, 00:09:09.610 --> 00:09:09.940 I see. 00:09:10.180 --> 00:09:11.740 So where's the work happening? 00:09:12.120 --> 00:09:12.660 Like where is it? 00:09:12.820 --> 00:09:16.300 Is it getting this data from in the browser or where's, 00:09:16.480 --> 00:09:16.960 when's the? 00:09:17.500 --> 00:09:17.740 There's a, 00:09:17.920 --> 00:09:21.140 there's a Go based web app running on someone's server that these, 00:09:21.520 --> 00:09:21.680 you know, 00:09:21.710 --> 00:09:23.200 that they've got up here that I believe. 00:09:23.760 --> 00:09:24.260 And then it just, 00:09:24.610 --> 00:09:25.400 it happens over there. 00:09:25.750 --> 00:09:30.640 If I pulled up a different one, you would see it grind away for maybe 10, 15 seconds. 00:09:31.880 --> 00:09:33.680 Well, that's a lot of data and a lot of graphs. 00:09:34.340 --> 00:09:34.900 I'm just, 00:09:35.280 --> 00:09:38.660 the pickiness in me, I'm like, well, what graph doesn't it have? 00:09:38.930 --> 00:09:41.160 I don't see any candlestick bar graphs. 00:09:42.900 --> 00:09:43.180 Sorry. 00:09:43.360 --> 00:09:44.140 Yeah, I know. 00:09:44.280 --> 00:09:44.960 It's pretty neat, though, 00:09:45.170 --> 00:09:46.700 if you want to explore some repos. 00:09:47.000 --> 00:09:47.400 It's cool. 00:09:48.640 --> 00:09:53.000 I haven't seen people include YouTube references before and something like that. 00:09:53.720 --> 00:09:55.620 Oh, yeah, and there's all this side stuff over here. 00:09:55.750 --> 00:09:56.440 You can compare. 00:09:56.660 --> 00:09:57.300 Here's the commits. 00:09:57.840 --> 00:09:58.740 So give it a moment. 00:09:58.900 --> 00:10:01.940 See, now it's grinding because I didn't ask for the commits on Flask earlier today. 00:10:02.060 --> 00:10:03.740 You can see PRs, issues, 00:10:03.940 --> 00:10:04.380 forks. 00:10:04.940 --> 00:10:05.600 Like, that's pretty neat. 00:10:05.740 --> 00:10:07.680 It's kind of hidden in this little hamburger menu on the left, 00:10:07.800 --> 00:10:08.820 but if you expand it out. 00:10:09.140 --> 00:10:09.340 Yeah. 00:10:09.560 --> 00:10:11.220 But it'll probably take, 00:10:11.400 --> 00:10:12.660 this should be interesting, I think. 00:10:12.720 --> 00:10:14.160 I've got some ideas. 00:10:14.660 --> 00:10:16.660 But I don't know, it'll take maybe 30 seconds. 00:10:16.800 --> 00:10:17.120 There you go. 00:10:17.800 --> 00:10:21.020 So year over, or all time, you can see there's quite a few commits 00:10:21.240 --> 00:10:22.340 and trending downward. 00:10:22.560 --> 00:10:23.540 It's kind of a, honestly, 00:10:23.730 --> 00:10:24.500 kind of a done product, 00:10:24.990 --> 00:10:25.200 you know. 00:10:25.850 --> 00:10:27.580 I don't mean that in a positive or negative way. 00:10:27.820 --> 00:10:29.820 There's a few things left that I think Flask could do. 00:10:30.300 --> 00:10:30.640 Primarily, 00:10:31.060 --> 00:10:31.920 I think one of the, 00:10:32.280 --> 00:10:33.840 there's probably two things to work on. 00:10:33.940 --> 00:10:37.500 I think primarily you could work on maybe some performance stuff to make Flask better. 00:10:37.840 --> 00:10:45.040 And you could unify the true async court and the sort of pseudo async, regular Flask async functionality. 00:10:45.500 --> 00:10:49.240 I know David and Phillip are like working towards making those just one thing. 00:10:49.540 --> 00:10:51.200 But you know, that could be a lot of work right there. 00:10:51.520 --> 00:10:52.120 That could be happening. 00:10:52.760 --> 00:10:55.220 But other than that, I kind of feel like Flask is sort of done. 00:10:55.480 --> 00:10:55.680 Yeah. 00:10:55.860 --> 00:10:56.020 Anyway, 00:10:56.400 --> 00:10:58.120 this is not about Flask. 00:10:58.340 --> 00:10:59.220 It's Flask as an example. 00:10:59.440 --> 00:11:01.840 It's about the Daily Stars Explorer, which I think is pretty cool. 00:11:02.080 --> 00:11:04.140 Just a link to us from last week. 00:11:05.140 --> 00:11:09.020 No project's ever done because you have to test on new Python versions every year. 00:11:09.700 --> 00:11:09.780 Yeah. 00:11:10.020 --> 00:11:13.140 And if it completely gets read as done, then it starts to get really badly. 00:11:13.480 --> 00:11:15.040 Because it's dust and stuff. 00:11:15.340 --> 00:11:15.580 All right. 00:11:15.800 --> 00:11:16.040 Exactly. 00:11:16.180 --> 00:11:17.360 It gets read as if it were dead. 00:11:18.360 --> 00:11:19.720 How about something to markdown? 00:11:20.000 --> 00:11:20.740 Maybe document it? 00:11:20.990 --> 00:11:21.700 Write a book about it? 00:11:22.140 --> 00:11:22.500 Yeah. 00:11:23.030 --> 00:11:25.880 So I like a lot of markdown. 00:11:26.820 --> 00:11:28.340 I use it a lot of all the time. 00:11:28.540 --> 00:11:29.340 I'm actually writing, 00:11:29.780 --> 00:11:33.440 I wrote the Python testing with pytest in Markdown. 00:11:33.760 --> 00:11:35.540 I'm writing LeanTDD in Markdown. 00:11:35.840 --> 00:11:36.740 Use it for blog posts. 00:11:36.840 --> 00:11:37.380 Use it for everything. 00:11:37.840 --> 00:11:39.980 So what's new about this? 00:11:40.280 --> 00:11:43.280 What's new is I just learned about a tool called Types. 00:11:44.260 --> 00:11:45.380 I think it's type. 00:11:46.500 --> 00:11:48.940 It even says how to pronounce it, and I didn't look. 00:11:49.180 --> 00:11:50.600 So T-Y-P-S-T. 00:11:51.560 --> 00:11:56.900 So I needed a new formatting thing because I'm trying to self-publish LeanTDD, 00:11:56.920 --> 00:11:58.640 and I didn't like the book, 00:11:59.060 --> 00:12:02.700 the print book versions of output that I have available. 00:12:02.890 --> 00:12:04.000 So I wanted to customize that. 00:12:04.070 --> 00:12:07.840 So I reached out to all of our good friend, 00:12:08.440 --> 00:12:08.860 Matt Harrison, 00:12:09.060 --> 00:12:10.240 because he self-published and 00:12:10.870 --> 00:12:12.200 said, Matt, what do you use? 00:12:12.580 --> 00:12:14.740 And he's using types now. 00:12:14.890 --> 00:12:16.740 So I tried to check it out. 00:12:17.020 --> 00:12:21.340 So the gist is that Pandoc converts to PDF, 00:12:21.800 --> 00:12:23.980 but it does it through LaTeX. 00:12:24.110 --> 00:12:25.880 And I don't really like LaTeX. 00:12:26.900 --> 00:12:30.500 And you have to download something extra anyway to get Pandoc to convert it. 00:12:30.900 --> 00:12:31.180 So, 00:12:31.880 --> 00:12:32.140 and then, 00:12:32.440 --> 00:12:33.060 but there's types. 00:12:33.300 --> 00:12:34.040 It's actually quite a pain. 00:12:34.280 --> 00:12:36.240 It's not just a, it's like three or four libraries. 00:12:36.500 --> 00:12:36.960 They're all, 00:12:37.340 --> 00:12:37.520 I don't know. 00:12:37.520 --> 00:12:38.420 They're just, it's weird. 00:12:38.540 --> 00:12:38.920 It's not great. 00:12:39.220 --> 00:12:39.400 Okay. 00:12:39.600 --> 00:12:44.120 So I'll jump to the punchline is I've got a convert script now that does, 00:12:44.760 --> 00:12:45.400 converts 00:12:45.600 --> 00:12:47.240 Pandoc to type, 00:12:47.540 --> 00:12:47.940 TYP. 00:12:48.460 --> 00:12:52.240 And I don't know what the official extension is, but I just named it. 00:12:52.340 --> 00:12:56.720 So you say two types in for Pandoc, but it's a fairly recent Pandoc that, 00:12:56.840 --> 00:12:58.360 There's been updates recently, 00:12:58.540 --> 00:13:00.280 so you'll want to upgrade your Pandoc first, 00:13:00.760 --> 00:13:02.620 then convert your markdown, 00:13:02.780 --> 00:13:05.940 and then take the type TYP output, 00:13:06.400 --> 00:13:08.440 and you do types to compile, 00:13:08.840 --> 00:13:09.940 and then the output. 00:13:10.360 --> 00:13:11.960 And I did a little example, 00:13:12.160 --> 00:13:17.420 so it's just a normal markdown, and it just converts it to some nice PDF. 00:13:18.020 --> 00:13:18.980 So there's a whole bunch of cool... 00:13:18.980 --> 00:13:20.840 Oh, and it has syntax highlighting as well? 00:13:21.220 --> 00:13:21.320 Oh, 00:13:21.500 --> 00:13:21.580 yeah. 00:13:21.800 --> 00:13:23.880 It does syntax highlighting, 00:13:24.360 --> 00:13:26.660 and there's so much that it does. 00:13:26.760 --> 00:13:28.520 It has mathematical symbols. 00:13:28.720 --> 00:13:30.460 It is a replacement for LaTeX, 00:13:30.910 --> 00:13:33.120 for one, but it's very much simpler. 00:13:33.420 --> 00:13:34.700 And I, 00:13:35.020 --> 00:13:43.460 in like an hour of twiddling with it, I have a fairly decent book template for the new book. 00:13:44.740 --> 00:13:47.640 And yeah, so I'm pretty happy with it. 00:13:49.579 --> 00:13:52.840 And this has been around for a while, and I just didn't know about it. 00:13:53.040 --> 00:13:54.780 So that's why I'm bringing it up. 00:13:55.040 --> 00:13:56.140 And installing wise, 00:13:56.640 --> 00:13:56.960 just brew. 00:13:57.420 --> 00:13:58.460 For a Mac, brew. 00:13:59.200 --> 00:13:59.640 On Windows, 00:13:59.880 --> 00:14:01.580 there was some other process I did, 00:14:01.740 --> 00:14:06.100 but it's so much easier than the LaTeX chain to install. 00:14:06.480 --> 00:14:09.720 So I'm pretty happy with this tool. 00:14:09.960 --> 00:14:10.480 So thanks. 00:14:10.940 --> 00:14:12.220 So I was so happy with it. 00:14:12.220 --> 00:14:14.820 I did a little blog post to share with people. 00:14:16.080 --> 00:14:16.380 Excellent. 00:14:17.120 --> 00:14:17.460 Excellent. 00:14:17.840 --> 00:14:19.520 That is really interesting. 00:14:19.600 --> 00:14:24.820 I might start using that because right now I'm doing just straight markdown 00:14:24.840 --> 00:14:27.240 to Pandoc for Talk Python in production. 00:14:27.760 --> 00:14:29.760 And it's really good for EPUB, 00:14:29.980 --> 00:14:31.640 but it's kind of janky for PDF 00:14:31.880 --> 00:14:33.560 and it's kind of janky for Kindle 00:14:33.740 --> 00:14:36.060 because Kindle can't read it quite right. 00:14:36.060 --> 00:14:37.360 So I have to do like remove 00:14:37.540 --> 00:14:39.220 a few little niceties for Kindle. 00:14:39.380 --> 00:14:39.800 So anyway, 00:14:40.440 --> 00:14:41.440 certainly worth considering. 00:14:41.520 --> 00:14:41.700 Really? 00:14:41.940 --> 00:14:42.120