WEBVTT 00:00:00.001 --> 00:00:04.960 Hello and welcome to Python Bytes, where we deliver Python news and headlines directly to your earbuds. 00:00:04.960 --> 00:00:09.800 This is episode 351, recorded September 5th, and I am Brian Okken. 00:00:09.800 --> 00:00:11.020 And I'm Michael Kennedy. 00:00:11.020 --> 00:00:14.920 And this episode is also sponsored by Sentry. Thank you, Sentry. 00:00:14.920 --> 00:00:19.420 And if you want to reach any of us or the show, either of us or the show, 00:00:19.420 --> 00:00:25.560 we're at mkennedy, at brianauken, and at pythonbytes, all at fostadon.org. 00:00:25.680 --> 00:00:31.220 And if you're listening later and you'd like to join the show live sometime, 00:00:31.220 --> 00:00:36.080 you can join us on YouTube at pythonbytes.fm/live. 00:00:36.080 --> 00:00:37.700 And we'd love to have you here. 00:00:37.700 --> 00:00:38.340 Indeed. 00:00:38.340 --> 00:00:39.820 What do you got for us first, Michael? 00:00:39.820 --> 00:00:42.600 Let's talk about multiprocessing. 00:00:42.600 --> 00:00:46.240 Empire, not pyre, but mpire, right? 00:00:46.240 --> 00:00:48.460 Pyre is the type checker from meta. 00:00:48.460 --> 00:00:53.220 Empire is something entirely different about multiprocessing. 00:00:53.220 --> 00:00:58.820 And so it's a Python package for easy multiprocessing that's faster than the built-in 00:00:58.820 --> 00:01:04.840 multiprocessing, but has a similar API, has really good error reporting and a bunch of 00:01:04.840 --> 00:01:07.840 other types of reporting, like how well did that session go? 00:01:07.840 --> 00:01:13.080 You know, like how well did you utilize the multiprocessing capabilities of your machine 00:01:13.080 --> 00:01:13.620 and so on? 00:01:13.620 --> 00:01:15.400 So yeah, let's dig into it. 00:01:15.400 --> 00:01:20.440 So the whole acronym for the name is multiprocessing is really easy, 00:01:20.440 --> 00:01:22.260 which is not what most people say, right? 00:01:22.600 --> 00:01:23.040 No. 00:01:23.040 --> 00:01:29.780 But it's a package that's faster than multiprocessing in most cases, has more features, 00:01:29.780 --> 00:01:34.020 and is generally more user-friendly than the default multiprocessing package. 00:01:34.020 --> 00:01:41.560 It has APIs like multiprocessing.pool, but it also has the benefits of things like copy, 00:01:41.560 --> 00:01:42.620 unwrite, shared objects. 00:01:42.620 --> 00:01:44.720 We're going to come back to that later as well. 00:01:45.200 --> 00:01:52.380 But also the ability to have like init and exit, set up teardown for your workers, some more 00:01:52.380 --> 00:01:53.960 state that you can use and so on. 00:01:53.960 --> 00:01:55.300 So pretty cool. 00:01:55.300 --> 00:01:56.920 It has a progress bar. 00:01:57.080 --> 00:02:02.660 It has TQDM progress built right into it across the multiple processes. 00:02:02.660 --> 00:02:05.720 So you can say things like, here is some work I want you to do. 00:02:05.720 --> 00:02:06.700 There's a hundred items. 00:02:06.700 --> 00:02:09.500 Split that across five cores. 00:02:10.020 --> 00:02:15.960 And as those different processes complete the work for individual elements, give me a unified 00:02:15.960 --> 00:02:18.260 progress bar, which is pretty awesome, right? 00:02:18.260 --> 00:02:18.820 Yeah. 00:02:18.820 --> 00:02:19.300 Yeah. 00:02:19.300 --> 00:02:19.760 Very cool. 00:02:19.880 --> 00:02:20.000 Yeah. 00:02:20.000 --> 00:02:20.540 Yeah. 00:02:20.540 --> 00:02:22.280 It's got a progress dashboard. 00:02:22.280 --> 00:02:23.580 Actually, I have no idea what that is. 00:02:23.580 --> 00:02:27.260 It has a worker insights that you can ask when it's done. 00:02:27.260 --> 00:02:31.020 Like how well did, you know, how efficient was that multiprocessing story? 00:02:31.020 --> 00:02:33.660 Graceful and user-friendly exception handling. 00:02:33.660 --> 00:02:34.680 It has timeouts. 00:02:34.680 --> 00:02:41.200 So you can say, I would like the execution of the work to not take more than three seconds. 00:02:41.200 --> 00:02:47.240 And actually you can even say things such as if the worker process itself takes 10 seconds 00:02:47.240 --> 00:02:50.860 or more to exit, maybe there's like some, something happening over there. 00:02:50.860 --> 00:02:55.140 That's like a hung connection on a database thing or who knows, right? 00:02:55.140 --> 00:02:55.780 Some network thing. 00:02:55.780 --> 00:02:59.320 You can actually set a different timeout for the process, which is pretty cool. 00:02:59.320 --> 00:03:01.680 It has automatic chunking. 00:03:01.680 --> 00:03:06.580 So instead of saying, I have a hundred things, let's go individually one at a time, hand them 00:03:06.580 --> 00:03:07.200 off to workers. 00:03:07.200 --> 00:03:13.460 It can break them up into bigger blocks, including numpy arrays, which is pretty cool. 00:03:13.740 --> 00:03:18.360 You can set the maximum number of tasks that are allowed to run at any given moment. 00:03:18.360 --> 00:03:21.780 I guess, you know, you can set the workers, but also like, if it does this chunking and 00:03:21.780 --> 00:03:24.620 you can say how many things can run to avoid memory problems. 00:03:24.620 --> 00:03:29.400 You could even say, four, I want to use five processes, but after. 00:03:29.400 --> 00:03:33.480 You know, 10, 10 bits of work on any given process. 00:03:33.660 --> 00:03:38.280 Give me a new worker and shut down the others in case there's like leaky memory or other 00:03:38.280 --> 00:03:39.480 things along those lines. 00:03:39.480 --> 00:03:42.880 you can create a pool of them through a daemon option. 00:03:42.880 --> 00:03:44.080 A whole bunch of stuff. 00:03:44.080 --> 00:03:51.000 It uses dill to serialize across, the multi-process processes, which is cool because it gives you 00:03:51.000 --> 00:04:00.000 more exotic, serialization options for say objects that are not pickleable lambdas functions and other 00:04:00.000 --> 00:04:01.980 things in IPython and Jupyter notebooks. 00:04:01.980 --> 00:04:03.420 So all that's pretty awesome. 00:04:03.420 --> 00:04:03.700 Right? 00:04:03.700 --> 00:04:04.260 Yeah. 00:04:04.260 --> 00:04:04.840 Yeah. 00:04:05.060 --> 00:04:11.180 So the API is super, super simple from empire import worker pool with worker pool jobs equal 00:04:11.180 --> 00:04:14.520 five, pull that map, some function, some data go. 00:04:14.520 --> 00:04:19.840 So this jobs here tells you how many processes to run basically for the progress bar. 00:04:19.840 --> 00:04:22.560 you just set progress bar equals true. 00:04:22.560 --> 00:04:23.620 That's not too bad. 00:04:23.620 --> 00:04:26.220 Another thing that's cool is you can have shared objects. 00:04:26.220 --> 00:04:33.500 So you can have, some shared data that's passed across without, basically using shared 00:04:33.500 --> 00:04:33.840 memory. 00:04:33.840 --> 00:04:37.860 I think is how that it works so that it's more efficient instead of trying to pickle it 00:04:37.860 --> 00:04:38.280 across. 00:04:38.280 --> 00:04:39.740 I think they have to be read only or something. 00:04:39.740 --> 00:04:40.720 And there's a whole bunch about it. 00:04:40.720 --> 00:04:41.240 Oh, interesting. 00:04:41.240 --> 00:04:43.640 But you pass it into the worker pool. 00:04:43.640 --> 00:04:44.140 Okay. 00:04:44.140 --> 00:04:44.740 Yeah. 00:04:44.740 --> 00:04:49.060 You say worker pool, these things, I want you to set them up in a way to be shared. 00:04:49.060 --> 00:04:54.740 And I think, like I said, in a read only way across, across all the processes instead 00:04:54.740 --> 00:04:55.900 of trying to copy them over. 00:04:55.900 --> 00:05:03.780 you have a setup and a teardown thing that you can do, to like prepare the worker, 00:05:03.780 --> 00:05:08.840 when it gets started, you can ask it for the insights, like I said, and then 00:05:08.840 --> 00:05:09.260 benchmarks. 00:05:09.260 --> 00:05:11.760 It shows it's significantly faster. 00:05:11.760 --> 00:05:16.080 Not just the compared, not just against multi-processing, but they say, here's how you do it. 00:05:16.080 --> 00:05:17.980 Here's what happens if you do it in a serial way. 00:05:17.980 --> 00:05:21.440 Here's what multi-processing and process pool executors look like. 00:05:21.520 --> 00:05:24.520 But it also compares against job lib, dask, and array. 00:05:24.520 --> 00:05:24.960 Mm. 00:05:24.960 --> 00:05:28.580 And it's, it's pretty much hanging there with the best of them, isn't it? 00:05:28.580 --> 00:05:29.000 Yeah. 00:05:29.000 --> 00:05:31.980 It does a titch faster than ray everywhere. 00:05:31.980 --> 00:05:32.460 Yeah. 00:05:32.460 --> 00:05:33.560 Just, yeah. 00:05:33.560 --> 00:05:34.560 Just, just a bit. 00:05:34.560 --> 00:05:35.840 One other thing. 00:05:35.960 --> 00:05:40.180 I don't remember where it was, in this huge, long list of things. 00:05:40.180 --> 00:05:47.500 but you can also pin the CPUs to CPU cores, which can be really valuable when you're thinking 00:05:47.500 --> 00:05:52.160 about, taking advantage of, you know, L1, L2 CPU caches. 00:05:52.300 --> 00:05:56.920 So if your processes are bouncing around back and forth, one's working with some data, then 00:05:56.920 --> 00:05:58.840 it switches to another core. 00:05:58.840 --> 00:06:05.340 And then it has to pull a new data into the, into the L2 cache, which is like hundreds 00:06:05.340 --> 00:06:06.800 of times slower than real memory. 00:06:06.800 --> 00:06:08.080 And that's how that slows it down. 00:06:08.080 --> 00:06:09.480 Then they switch back and they keep running. 00:06:09.480 --> 00:06:14.740 So you can say, you know, pin these workers to these CPUs and you know, you've got a better 00:06:14.740 --> 00:06:17.600 chance of them not redoing their cache all the time. 00:06:17.600 --> 00:06:18.260 So that's kind of cool. 00:06:18.260 --> 00:06:19.900 So there's just a bunch of neat little features in here. 00:06:19.900 --> 00:06:20.480 Yeah. 00:06:20.620 --> 00:06:23.400 If you're already using multiprocessing, you might check this out. 00:06:23.400 --> 00:06:26.800 Oh, if you care about performance for real, you know, 00:06:26.800 --> 00:06:30.740 why are you using multiprocessing if you don't care about performance? 00:06:30.740 --> 00:06:36.240 Well, I mean, you're, you're looking to pull out the final little bits of performance, I 00:06:36.240 --> 00:06:36.640 suppose. 00:06:36.640 --> 00:06:37.020 Yeah. 00:06:37.020 --> 00:06:37.220 Yeah. 00:06:37.220 --> 00:06:37.440 Yeah. 00:06:37.440 --> 00:06:38.100 Right. 00:06:38.100 --> 00:06:42.160 Like these benchmarks are cool, but they're doing, you know, computation on the workers. 00:06:42.160 --> 00:06:42.600 Right. 00:06:42.600 --> 00:06:47.380 Whereas a lot of what you're doing is like talking to queues and talking to networks and talking 00:06:47.380 --> 00:06:47.980 to databases. 00:06:47.980 --> 00:06:51.260 Like it doesn't matter what framework you used to do that long as you're doing something 00:06:51.260 --> 00:06:51.680 parallel. 00:06:51.680 --> 00:06:52.300 Right. 00:06:52.300 --> 00:06:52.820 Yeah. 00:06:52.820 --> 00:06:53.680 Well, yeah. 00:06:53.680 --> 00:06:54.320 Well, I don't know. 00:06:54.320 --> 00:06:55.760 That's why you have to do your own benchmarks. 00:06:55.760 --> 00:06:56.760 Yeah, for sure. 00:06:56.760 --> 00:07:00.740 And then there's that article over on medium by the creator as well. 00:07:01.160 --> 00:07:03.640 That gives you a whole lot of background on all this stuff. 00:07:03.640 --> 00:07:04.220 Oh, neat. 00:07:04.220 --> 00:07:04.800 Nice. 00:07:04.800 --> 00:07:05.140 Yeah. 00:07:05.140 --> 00:07:06.740 This is quite a long article. 00:07:06.740 --> 00:07:08.660 And I think it's actually more relevant. 00:07:08.660 --> 00:07:13.240 Like, for example, it's got screenshots where it shows, you know, if you use something like, 00:07:13.240 --> 00:07:14.760 let me read really quickly. 00:07:14.760 --> 00:07:19.000 Ray or job lib and you get some kind of exception. 00:07:19.000 --> 00:07:20.560 It just says exception occurred. 00:07:20.560 --> 00:07:21.160 Yikes. 00:07:21.540 --> 00:07:26.340 Whereas with this one, with empire, you get things like here's the call stack that goes 00:07:26.340 --> 00:07:28.520 back a little more correctly to the right function. 00:07:28.520 --> 00:07:30.900 And here's the arguments that were passed. 00:07:30.900 --> 00:07:31.960 Oh, interesting. 00:07:31.960 --> 00:07:32.620 Over. 00:07:32.620 --> 00:07:37.700 So when it crashes, you know, because you have like five processes, potentially all doing 00:07:37.700 --> 00:07:37.940 stuff. 00:07:37.940 --> 00:07:38.580 One of them crashed. 00:07:38.580 --> 00:07:39.780 Like what data did they have? 00:07:39.780 --> 00:07:40.480 I don't know. 00:07:40.480 --> 00:07:41.460 It's a parallel. 00:07:41.460 --> 00:07:41.960 It's hard. 00:07:41.960 --> 00:07:42.500 Right. 00:07:42.560 --> 00:07:46.600 So having the arguments like these are the ones that cause the error is pretty excellent. 00:07:46.600 --> 00:07:47.120 Yeah. 00:07:47.120 --> 00:07:47.580 Cool. 00:07:47.580 --> 00:07:49.400 Anyway, empire. 00:07:49.400 --> 00:07:50.560 That's what I got for number one. 00:07:50.560 --> 00:07:52.240 All right. 00:07:52.240 --> 00:07:52.500 Cool. 00:07:52.500 --> 00:07:56.740 I want to have, I have something else that starts with M. 00:07:56.740 --> 00:07:58.840 Mop up. 00:07:58.840 --> 00:08:05.240 So mop up is something that I learned about from an article by glyph. 00:08:05.240 --> 00:08:07.340 So I'll let me jump to the article first. 00:08:07.340 --> 00:08:12.040 So glyph wrote an article saying, get your Mac Python from python.org. 00:08:12.040 --> 00:08:13.440 That's what I already do. 00:08:13.440 --> 00:08:19.080 I've tried all the other stuff and I just like just the downloader from python.org. 00:08:19.080 --> 00:08:23.760 So this article talks about reasons why that's probably what you want. 00:08:23.760 --> 00:08:28.900 And that's probably what if you're writing a tutorial, it's probably what your users need 00:08:28.900 --> 00:08:29.440 to do, too. 00:08:29.440 --> 00:08:35.840 If they're if they're using a Mac and I won't go through all the details, but he he goes through 00:08:35.840 --> 00:08:42.020 reasons why you probably want this one and not things like what are the others? 00:08:42.020 --> 00:08:42.960 homebrew. 00:08:42.960 --> 00:08:46.580 You can brew install your Python, but he doesn't recommend it. 00:08:46.580 --> 00:08:48.200 And you can read all the pyenv. 00:08:48.200 --> 00:08:49.200 I've tried it. 00:08:49.200 --> 00:08:52.040 It like messes up other stuff for me. 00:08:52.040 --> 00:08:54.500 So I like the downloader from Python. 00:08:54.500 --> 00:09:01.720 But one of the things that I don't like is that if like if I had Python 311 4 installed and 00:09:01.720 --> 00:09:03.640 now Python 311 5 is out. 00:09:03.780 --> 00:09:05.480 How do I get that on my computer? 00:09:05.480 --> 00:09:06.660 Do I just reinstall it? 00:09:06.660 --> 00:09:07.620 Yes, you can. 00:09:07.620 --> 00:09:11.900 But Glyph made a new thing called mop up. 00:09:11.900 --> 00:09:16.720 So what mop up does is you you just pip install mop up. 00:09:16.840 --> 00:09:23.920 And it's like the only thing I install on my global my global Python versions like 311 pip install. 00:09:23.920 --> 00:09:26.980 I update pip and install this and that's it. 00:09:26.980 --> 00:09:28.820 Everything else goes into a virtual environment. 00:09:29.580 --> 00:09:32.000 Or pip x install this. 00:09:32.000 --> 00:09:32.560 Exactly. 00:09:32.560 --> 00:09:34.640 But mop up. 00:09:34.640 --> 00:09:35.360 What's the usage? 00:09:35.360 --> 00:09:36.680 So I just tried it this morning. 00:09:36.680 --> 00:09:38.780 I didn't pass it any flags. 00:09:38.780 --> 00:09:40.860 I just installed it and ran it. 00:09:40.860 --> 00:09:48.360 And it updated me from Python 311 4 to Python 311 5 without me having to re-download anything 00:09:48.360 --> 00:09:49.320 other than this. 00:09:49.580 --> 00:09:52.680 So I'm going to set up something that goes through. 00:09:52.680 --> 00:09:54.760 I've got a lot of versions on my computer. 00:09:54.760 --> 00:09:58.140 I've got I think well I've got 3.7 through 3.12 install. 00:09:58.140 --> 00:10:02.960 And and I want all of them to be on the latest bug fix release. 00:10:02.960 --> 00:10:12.760 So I'm just going to use probably use Brett Cannon's pi installer or Python installer pi on 00:10:12.760 --> 00:10:17.640 my Mac to go to each of the versions and run mop up on all of them to update it. 00:10:17.780 --> 00:10:19.920 So that's what I'd like to do. 00:10:19.920 --> 00:10:20.900 Anyway it's cool. 00:10:20.900 --> 00:10:25.940 I'm I'm really excited about this because this was like the one hole in using the install 00:10:25.940 --> 00:10:29.840 the the python.org installer is how they update it. 00:10:29.840 --> 00:10:31.260 So nice. 00:10:31.260 --> 00:10:31.520 Yep. 00:10:31.520 --> 00:10:32.260 Interesting. 00:10:32.260 --> 00:10:32.780 Interesting. 00:10:32.780 --> 00:10:37.460 I got to admit I'm still a brew install Python 3 sort of person. 00:10:37.460 --> 00:10:38.080 Okay. 00:10:38.080 --> 00:10:43.660 And the drawback the main drawback that glyph makes an argument for which is valid is you 00:10:43.660 --> 00:10:46.700 don't control necessarily the version of Python that you get. 00:10:46.820 --> 00:10:53.740 Because if you brew install I don't know some other you know YouTube downloader app or whatever 00:10:53.740 --> 00:10:59.880 rando thing it might say well I need a Python 3.12 and you only have 3.8 right. 00:10:59.880 --> 00:11:02.120 And it'll auto upgrade on you without you knowing. 00:11:02.120 --> 00:11:05.560 But I'm always running the absolute latest Python anyway. 00:11:05.560 --> 00:11:10.140 And so you know when it those other packages say greater than 3.10 like I don't care I already 00:11:10.140 --> 00:11:11.040 have greater than 3.10. 00:11:11.400 --> 00:11:15.600 So I don't know that's the world I'm living in now but that's okay for me. 00:11:15.600 --> 00:11:16.680 Oh okay. 00:11:16.680 --> 00:11:23.220 So yeah I'm I'm a package maintainer so I I have multiple versions on my box but it's 00:11:23.220 --> 00:11:28.100 but in a lot of people like pyenv for that reason but I yeah I don't. 00:11:28.320 --> 00:11:32.640 But anyway I've always I've had trouble with pyenv too especially around the Apple Silicon 00:11:32.640 --> 00:11:41.260 Rosetta compiler mismatch like there's just like it wouldn't install for me and so yeah I think the 00:11:41.260 --> 00:11:43.760 whole I think the python.org it's a good recommendation though. 00:11:43.760 --> 00:11:44.580 Okay cool. 00:11:44.580 --> 00:11:45.520 Yep yep. 00:11:45.520 --> 00:11:46.080 All right. 00:11:46.920 --> 00:11:49.740 before we move on to our next topic Brian. 00:11:49.740 --> 00:11:55.000 Well I'd like to thank Sentry for sponsoring this episode of Python Bytes. 00:11:55.000 --> 00:11:59.560 You know Sentry for their error tracking service but did you know that you can take take it all 00:11:59.560 --> 00:12:04.520 the way through your multi-tier and distributed app with their distributed tracing feature? 00:12:04.520 --> 00:12:05.400 How cool is that? 00:12:05.400 --> 00:12:10.720 Distributed tracing is a debugging technique that involves tracking your the request of your 00:12:10.720 --> 00:12:15.120 system starting from the very beginning like the user action all the way to the back end 00:12:15.120 --> 00:12:16.840 database and third-party services. 00:12:16.840 --> 00:12:24.180 This can help you identify if the cause of the an error is one project in one project is due to an 00:12:24.180 --> 00:12:25.920 error in the other another project. 00:12:25.920 --> 00:12:27.280 That's very useful. 00:12:27.280 --> 00:12:33.240 Every system can benefit from distributed tracing but they are useful for especially for microservices 00:12:33.240 --> 00:12:40.060 in microservice architecture logs won't give you the full picture so you can't debug every request 00:12:40.060 --> 00:12:45.000 in full by reading the logs but distributed tracing with a platform like Sentry 00:12:45.000 --> 00:12:50.060 can give you visual overview of which services were called during the execution of certain requests. 00:12:50.060 --> 00:12:56.060 Aside from debugging and visualizing architecture distributed tracing also helps you identify 00:12:56.060 --> 00:12:59.060 performance bottlenecks through a visual like Gantt chart. 00:12:59.060 --> 00:13:08.120 You can see if a particular span in your stack took longer than expected and how it could be causing slowdowns in other parts of your app. 00:13:08.120 --> 00:13:17.180 You can see if you can see the results of your app. 00:13:17.180 --> 00:13:17.180 You can see the results of your app. 00:13:17.180 --> 00:13:18.180 You can see the results of your app. 00:13:18.180 --> 00:13:19.180 You can see the results of your app. 00:13:19.180 --> 00:13:30.240 You can see the results of your app. 00:13:30.240 --> 00:13:41.300 You can see the results of your app. 00:13:41.300 --> 00:13:52.360 You can see the results of your app. 00:13:52.360 --> 00:14:03.420 You can see the results of your app. 00:14:03.420 --> 00:14:04.420 You can see the results of your app. 00:14:04.420 --> 00:14:16.480 You can see the results of your app. 00:14:16.480 --> 00:14:28.540 You can see the results of your app. 00:14:28.540 --> 00:14:29.540 You can see the results of your app. 00:14:29.540 --> 00:14:30.540 You can see the results of your app. 00:14:30.540 --> 00:14:31.540 You can see the results of your app. 00:14:31.540 --> 00:14:32.540 You can see the results of your app. 00:14:32.540 --> 00:14:33.540 You can see the results of your app. 00:14:33.540 --> 00:14:34.540 You can see the results of your app. 00:14:34.540 --> 00:14:35.540 You can see the results of your app. 00:14:35.540 --> 00:14:36.540 You can see the results of your app. 00:14:36.540 --> 00:14:37.540 You can see the results of your app. 00:14:37.540 --> 00:14:38.540 You can see the results of your app. 00:14:38.540 --> 00:14:39.540 You can see the results of your app. 00:14:39.540 --> 00:14:40.540 You can see the results of your app. 00:14:40.540 --> 00:14:41.540 You can see the results of your app. 00:14:41.540 --> 00:14:42.540 You can see the results of your app. 00:14:42.540 --> 00:14:43.540 You can see the results of your app. 00:14:43.540 --> 00:14:44.540 You can see the results of your app. 00:14:44.540 --> 00:14:45.540 You can see the results of your app. 00:14:45.540 --> 00:14:46.540 You can see the results of your app. 00:14:46.540 --> 00:14:47.540 You can see the results of your app. 00:14:47.540 --> 00:14:48.540 You can see the results of your app. 00:14:48.540 --> 00:14:49.540 You can see the results of your app. 00:14:49.540 --> 00:14:50.540 You can see the results of your app. 00:14:50.540 --> 00:14:51.540 You can see the results of your app. 00:14:51.540 --> 00:14:52.540 You can see the results of your app. 00:14:52.540 --> 00:14:53.540 You can see the results of your app. 00:14:53.540 --> 00:14:54.540 You can see the results of your app. 00:14:54.540 --> 00:14:55.540 You can see the results of your app. 00:14:55.540 --> 00:14:56.540 You can see the results of your app. 00:14:56.540 --> 00:14:57.540 You can see the results of your app. 00:14:57.540 --> 00:14:58.540 You can see the results of your app. 00:14:58.540 --> 00:14:59.540 You can see the results of your app. 00:14:59.540 --> 00:15:00.540 You can see the results of your app. 00:15:00.540 --> 00:15:01.540 You can see the results of your app. 00:15:01.540 --> 00:15:02.540 You can see the results of your app. 00:15:02.540 --> 00:15:03.540 You can see the results of your app. 00:15:03.540 --> 00:15:04.540 You can see the results of your app. 00:15:04.540 --> 00:15:05.540 You can see the results of your app. 00:15:05.540 --> 00:15:06.540 You can see the results of your app. 00:15:06.540 --> 00:15:07.540 You can see the results of your app. 00:15:07.540 --> 00:15:08.540 You can see the results of your app. 00:15:08.540 --> 00:15:09.540 You can see the results of your app. 00:15:09.540 --> 00:15:10.540 You can see the results of your app. 00:15:10.540 --> 00:15:11.540 You can see the results of your app. 00:15:11.540 --> 00:15:12.540 You can see the results of your app. 00:15:12.540 --> 00:15:13.540 You can see the results of your app. 00:15:13.540 --> 00:15:19.540 So if you have x equals a string, and then I say y equals x, it goes up to that thing 00:15:19.540 --> 00:15:22.540 and says plus plus, you know, plus equals one on the reference count. 00:15:22.540 --> 00:15:26.540 And when y goes away, then it minus minuses it, right? 00:15:26.540 --> 00:15:27.540 When that number hits zero, it gets cleaned up. 00:15:27.540 --> 00:15:31.540 There's also stuff on the object for cycles and garbage collection. 00:15:31.540 --> 00:15:33.540 So there's a lot of stuff that's happening there, right? 00:15:33.540 --> 00:15:34.540 Yeah. 00:15:34.540 --> 00:15:41.540 And so what they're doing is they're running a lot of Django for Instagram, which is pretty awesome. 00:15:41.540 --> 00:15:49.540 However, what they're trying to take advantage of is the fact that there's a lot of similar data, similar memory usage. 00:15:49.540 --> 00:15:58.540 When I load up Python, so if I write type Python on the terminal, and then I open up a new terminal type Python, it's gone through exactly the same startup process, right? 00:15:58.540 --> 00:16:01.540 So it's loaded the same shared libraries or DLLs. 00:16:01.540 --> 00:16:02.540 It's created. 00:16:02.540 --> 00:16:09.540 It's, it's, you know, negative 255 to 255 flywheel numbers. 00:16:09.540 --> 00:16:13.540 There's going to reuse instead of when you say the number seven, it doesn't always create a new seven. 00:16:13.540 --> 00:16:17.540 You always have the seven that was created at startup exceptions, those kinds of things, right? 00:16:17.540 --> 00:16:32.540 Well, if you have a web server that's got 10 or 20 or a hundred worker processes that all went through the same startup for a Python app, you would want to have things like that number seven or some exception type or whatever modules, right? 00:16:32.540 --> 00:16:33.540 Core modules that are loaded. 00:16:33.540 --> 00:16:40.540 You would like to have one copy of those in memory on Linux, and then have a copy on right thing for the stuff that actually changes. 00:16:40.540 --> 00:16:42.540 But those other pieces, you want them to stay the same. 00:16:42.540 --> 00:16:43.540 Yeah. 00:16:43.540 --> 00:16:44.540 Yeah. 00:16:44.540 --> 00:16:49.540 Like there's no point in having like a different representation of the number four for every process. 00:16:49.540 --> 00:16:52.540 If there's some way to share that memory that was created at startup. 00:16:52.540 --> 00:16:55.540 And we don't need reference counts updated and all that stuff. 00:16:55.540 --> 00:16:56.540 Because exactly exactly. 00:16:56.540 --> 00:17:07.540 So the, I, what they found was, while many of their Python objects are practically or effectively immutable, they didn't actually over time behave that way. 00:17:07.540 --> 00:17:10.540 So they have graphs of private memory and shared memory. 00:17:10.540 --> 00:17:16.540 And what you would hope is that the shared memory stays pretty stable over time, or maybe even grows. 00:17:16.540 --> 00:17:17.540 Maybe you're doing new stuff. 00:17:17.540 --> 00:17:20.540 That's like pulled in similar things, but that's not what happens in practice. 00:17:20.540 --> 00:17:22.540 I'm current Python. 00:17:22.540 --> 00:17:32.540 The shared memory goes down and down and down, because even though that object, or let's say that flywheel number that got created to be shared, it's still got his reference count number change. 00:17:32.540 --> 00:17:36.540 So throughout the behavior of one app, it might go, well, four was used for a long time. 00:17:36.540 --> 00:17:39.540 Four was used 300 times here and 280 over there. 00:17:39.540 --> 00:17:41.540 So those are not the same four. 00:17:41.540 --> 00:17:45.540 Cause on the reference count, they have 281 and 301 or whatever it is. 00:17:45.540 --> 00:17:46.540 Right. 00:17:46.540 --> 00:18:02.540 And so that, that shared memory is falling down because the garbage collector and, just the interacting with the ref count is very in very meaningless and small ways, changing pieces of the shared memory to make them fall out of the shared state. 00:18:02.540 --> 00:18:06.540 So this whole path, this whole idea is we're going to make those types of things. 00:18:06.540 --> 00:18:08.540 so that their reference count can't change. 00:18:08.540 --> 00:18:10.540 And their GC structures can't change. 00:18:10.540 --> 00:18:11.540 They cannot be changed. 00:18:11.540 --> 00:18:17.540 They're just always set to some magic number for like this thing's reference count is unchanged. 00:18:17.540 --> 00:18:18.540 Right. 00:18:18.540 --> 00:18:26.540 So if you look at like the object header, it's got a GC header reference count object type, and then the actual data for the ones that don't change. 00:18:26.540 --> 00:18:28.540 Now these new ones can be set. 00:18:28.540 --> 00:18:31.540 So even their GC header and the reference counts don't change. 00:18:31.540 --> 00:18:32.540 Cool. 00:18:32.540 --> 00:18:32.540 Right. 00:18:32.540 --> 00:18:33.540 Yeah. 00:18:33.540 --> 00:18:33.540 Yeah. 00:18:33.540 --> 00:18:38.540 And what that means is if you come down here, it says there's some challenges. 00:18:38.540 --> 00:18:41.540 First, they had to make sure that applications wouldn't crash. 00:18:41.540 --> 00:18:44.540 If some objects suddenly had different ref counts. 00:18:44.540 --> 00:18:55.540 Second, it changes the core memory representation of a Python object, which if you work in the C level, just directly with the memory, that's, you know, pointers to the object. 00:18:55.540 --> 00:18:56.540 That can be tricky. 00:18:56.540 --> 00:19:08.540 And finally, the core implementation relies on adding checks explicitly to the increment and decrement, the, you know, add ref, remove ref, decrement ref, which are two of the hottest bits of code in all of Python. 00:19:08.540 --> 00:19:09.540 Yeah. 00:19:09.540 --> 00:19:10.540 As in the most performance critical. 00:19:10.540 --> 00:19:15.540 So if you make a change to it, or you make all the Python slower for this, that's bad. 00:19:15.540 --> 00:19:18.540 And they did make Python slower, but only 2%. 00:19:18.540 --> 00:19:22.540 And they believe that the benefits they get is actually worth it. 00:19:22.540 --> 00:19:26.540 Cause you bring in, you know, for like heavy workloads, you get actually better performance. 00:19:26.540 --> 00:19:28.540 So it's a trade off, but there it is. 00:19:28.540 --> 00:19:30.540 One of the things that was reading this article. 00:19:30.540 --> 00:19:40.540 And one of the things that confused me was, is there, is this just something internal to Python that, that it's going to happen under the hood or do I need to change my syntax in any way? 00:19:40.540 --> 00:19:41.540 Yes. 00:19:41.540 --> 00:19:42.540 I was looking for that as well. 00:19:42.540 --> 00:19:53.540 And every single thing about, I went and read the PEP and everything I remember from reading the PEP and maybe I missed something, but everything I got from the PEP was it, the C layer. 00:19:53.540 --> 00:19:59.540 It was, you know, here's the pie immortal, you know, call that you make in the C API. 00:19:59.540 --> 00:20:04.540 So what I would like to see is something where you set a decorate, like kind of like a data class. 00:20:04.540 --> 00:20:06.540 Like this thing is outside of garbage collection. 00:20:06.540 --> 00:20:14.540 This class is out or this, I don't know, in some way to say in Python, this thing is a moral for now, at least. 00:20:14.540 --> 00:20:15.540 Yeah. 00:20:15.540 --> 00:20:16.540 But I didn't see it either. 00:20:16.540 --> 00:20:22.540 It also would be good even if we could just do like, like, that would be kind of like a constant then also. 00:20:22.540 --> 00:20:28.540 We could set up some, some constants in your system that, that are immortal or something. 00:20:28.540 --> 00:20:29.540 Yeah. 00:20:29.540 --> 00:20:30.540 Okay. 00:20:30.540 --> 00:20:31.540 Yeah. 00:20:31.540 --> 00:20:41.540 Like the dictionary of a module that loads up, if you're not dynamically changing it, which you almost never do, unless you're like mocking something out, like, let it, let it be, you know, it's just tell it it's the same. 00:20:41.540 --> 00:20:42.540 Don't, don't reference combat. 00:20:42.540 --> 00:20:43.540 Yeah. 00:20:43.540 --> 00:20:53.540 I'd be curious to see in this implement, as they're implementing it, it does seem like parts of the system are going to go a little bit slower, but also parts of it are going to go faster because you don't have to do all that work, but. 00:20:53.540 --> 00:20:54.540 Exactly. 00:20:54.540 --> 00:20:55.540 Right. 00:20:55.540 --> 00:20:56.540 Yeah. 00:20:56.540 --> 00:20:57.540 You don't have to, you don't have to do a lot of stuff. 00:20:57.540 --> 00:20:58.540 Okay. 00:20:58.540 --> 00:21:00.540 Like the garbage collection cycles that happen over time, right? 00:21:00.540 --> 00:21:01.540 Yeah. 00:21:01.540 --> 00:21:04.540 These things will just be excluded from garbage collection entirely. 00:21:04.540 --> 00:21:05.540 So that's cool. 00:21:05.540 --> 00:21:10.540 So they have some graphs of what happened afterwards and the before and after and on the after in the shared memory. 00:21:10.540 --> 00:21:13.540 Well, sorry, the before it went almost to zero. 00:21:13.540 --> 00:21:20.540 Like it went from, you know, Y axis with no numbers really high to Y axis low with no numbers. 00:21:20.540 --> 00:21:22.540 But I don't know exactly what this is. 00:21:22.540 --> 00:21:33.540 Maybe a percent, but like I said, it doesn't really, really say, but after processing as few as 300 requests, it was like a 10th of the original shared memory was left. 00:21:33.540 --> 00:21:39.540 And that was it now after it's down to it's 75, 80% still shared, which is pretty excellent. 00:21:39.540 --> 00:21:40.540 Okay, cool. 00:21:40.540 --> 00:21:46.540 But as you said, this is like one of the internal core things from what I can tell. 00:21:46.540 --> 00:21:47.540 Yeah. 00:21:47.540 --> 00:21:58.540 They do say that this is foundational work for the per interpreter Gil that six, eight, four, as well as making the global interpreter lock optional and see Python seven or three. 00:21:58.540 --> 00:21:59.540 So that's why I'm not going to be able to do that. 00:21:59.540 --> 00:22:00.540 And that's why I'm not going to be able to do that. 00:22:00.540 --> 00:22:01.540 And that's why I'm not going to be able to do that. 00:22:01.540 --> 00:22:02.540 And that's why I'm not going to be able to do that. 00:22:02.540 --> 00:22:03.540 And that's why I'm not going to be able to do that. 00:22:03.540 --> 00:22:04.540 And that's why I'm not going to be able to do that. 00:22:04.540 --> 00:22:05.540 Well, I'm not going to be able to do that. 00:22:05.540 --> 00:22:06.540 I'm not going to be able to do that. 00:22:06.540 --> 00:22:07.540 I'm not going to be able to do that. 00:22:07.540 --> 00:22:08.540 I'm not going to be able to do that. 00:22:08.540 --> 00:22:09.540 I'm not going to be able to do that. 00:22:09.540 --> 00:22:10.540 I'm not going to be able to do that. 00:22:10.540 --> 00:22:11.540 I'm not going to be able to do that. 00:22:11.540 --> 00:22:12.540 I'm not going to be able to do that. 00:22:12.540 --> 00:22:13.540 I'm not going to be able to do that. 00:22:13.540 --> 00:22:14.540 I'm not going to be able to do that. 00:22:14.540 --> 00:22:15.540 I'm not going to be able to do that. 00:22:15.540 --> 00:22:16.540 I'm not going to be able to do that. 00:22:16.540 --> 00:22:17.540 I'm not going to be able to do that. 00:22:17.540 --> 00:22:18.540 I'm not going to be able to do that. 00:22:18.540 --> 00:22:19.540 I'm not going to be able to do that. 00:22:19.540 --> 00:22:20.540 I'm not going to be able to do that. 00:22:20.540 --> 00:22:22.540 So that's why it's there to support. 00:22:22.540 --> 00:22:26.540 And that's why it's relevant for some of these parallelism peps. 00:22:26.540 --> 00:22:28.540 So anyway, pretty cool. 00:22:28.540 --> 00:22:31.540 This is coming in 3.12, I guess. 00:22:31.540 --> 00:22:32.540 Nice. Cool. 00:22:32.540 --> 00:22:37.540 Well, I'd like to talk about something that I don't really think about that much in that 00:22:37.540 --> 00:22:40.540 that is doc strings for doc string formats. 00:22:40.540 --> 00:22:47.540 And I just ran across this article and I'm covering it partly just as a question to the audience. 00:22:47.540 --> 00:22:54.540 So the article is from Scott Robinson and it's called Common Doc String Formats in Python. 00:22:54.540 --> 00:22:58.540 And doc strings, people forget what they are. 00:22:58.540 --> 00:23:01.540 Like, let's say you have a function called add numbers or something. 00:23:01.540 --> 00:23:06.540 You can do, you can really do any kind of quote, but the first string in a function, 00:23:06.540 --> 00:23:09.540 if it's not assigned to a variable, is the doc string. 00:23:09.540 --> 00:23:11.540 And it's the first element. 00:23:11.540 --> 00:23:16.540 Anyway, the first line is a little, it's usually one line. 00:23:16.540 --> 00:23:19.540 one line and then maybe a space and then some other stuff. 00:23:19.540 --> 00:23:24.540 And there apparently there's several format, common formats of this. 00:23:24.540 --> 00:23:29.540 You can also, you can get access to it by the underscore doc attribute of something. 00:23:29.540 --> 00:23:37.540 So if you have a, a, a reference to a function, you can say dot dunder doc, and you can see the doc string. 00:23:37.540 --> 00:23:42.540 And a lot of like ID is use this to pop up hints and stuff. 00:23:42.540 --> 00:23:51.540 That's one of the reasons why you want to have like the first, at least the first line be an explanation that is good for somebody to see if it pops up on them and stuff. 00:23:51.540 --> 00:23:54.540 So anyway, which formats should this be? 00:23:54.540 --> 00:23:57.540 So it covers a handful of different formats. 00:23:57.540 --> 00:24:02.540 There's a restructured, restructured text doc stream format. 00:24:02.540 --> 00:24:08.540 So you've got all this like descriptions of parameters and the types and stuff. 00:24:08.540 --> 00:24:10.540 this is scary looking to me. 00:24:10.540 --> 00:24:15.540 there's the, the, let's go through next, Google doc stream format. 00:24:15.540 --> 00:24:18.540 This one makes a little more sense, but again, I don't know. 00:24:18.540 --> 00:24:21.540 it says, and it talks about the different parameters. 00:24:21.540 --> 00:24:26.540 And if, if you really have to describe them, this is probably one of my favorites is what looks pretty good. 00:24:26.540 --> 00:24:27.540 return. 00:24:27.540 --> 00:24:30.540 What, what, what, what, what is the information that it returns? 00:24:30.540 --> 00:24:36.540 what are the arguments and why some like one liner explanations, not too bad. 00:24:36.540 --> 00:24:39.540 There's a numpy scipy doc format. 00:24:39.540 --> 00:24:40.540 This is also pretty clear. 00:24:40.540 --> 00:24:44.540 maybe a little, let's compare the two. 00:24:44.540 --> 00:24:50.540 I guess this is it got an extra line because you're doing the underscore line, which is, I guess. 00:24:50.540 --> 00:24:51.540 Okay. 00:24:51.540 --> 00:24:53.540 It looks sort of, I don't know. 00:24:53.540 --> 00:24:58.540 This is a lot of space, but I would, I'm just curious if people are really using this. 00:24:58.540 --> 00:25:08.540 Looking at this, I can see the benefit of describing if it's not clear, from the name of your function, describing stuff. 00:25:08.540 --> 00:25:09.540 And I also like type hints. 00:25:09.540 --> 00:25:13.540 So this seems like a great argument for type hints because the types we, 00:25:13.540 --> 00:25:14.540 the types would be great. 00:25:14.540 --> 00:25:15.540 Just right in the right. 00:25:15.540 --> 00:25:16.540 Right. 00:25:16.540 --> 00:25:16.540 Right. 00:25:16.540 --> 00:25:17.540 Exactly. 00:25:17.540 --> 00:25:18.540 The parameters. 00:25:18.540 --> 00:25:25.540 and then if you don't have to describe the type, maybe just have variable names that are more clear. 00:25:25.540 --> 00:25:34.540 So, I, my personal preference really is, use type hints and then also, have a description. 00:25:34.540 --> 00:25:40.540 If you're going to do a doc string and it's not obvious from the name of the function, then have a description of what the function does. 00:25:40.540 --> 00:25:41.540 And that's it. 00:25:41.540 --> 00:25:48.540 and then if it's unclear to about what really what the stuff is, the behavior of different parameters, then add that. 00:25:48.540 --> 00:25:57.540 But, I, again, I'd love to hear back from people, go ahead and, send me a message on, @brianokken at faucet.org. 00:25:57.540 --> 00:25:59.540 This worked great last week. 00:25:59.540 --> 00:26:00.540 I got some great feedback. 00:26:00.540 --> 00:26:04.540 and so I'd love to hear what people are doing for their doc string formats. 00:26:04.540 --> 00:26:06.540 Do you, do you use doc string formats, Michael? 00:26:06.540 --> 00:26:09.540 I'm familiar with doc string formats and I've played with them. 00:26:09.540 --> 00:26:12.540 I like the Google one best, I think, but I'm with you. 00:26:12.540 --> 00:26:16.540 Like if you have good variable names, do you need the parameter information? 00:26:16.540 --> 00:26:19.540 If you use type hints, do you need the parameter information to say the type? 00:26:19.540 --> 00:26:23.540 If you have a return declaration with a type, do you need to have the returns? 00:26:23.540 --> 00:26:31.540 The function has a good name, like get user, you know, area angle bracket, optional user, like, oh, well, it returns a user or it returns none. 00:26:31.540 --> 00:26:33.540 How much more do you need to say about what it returns? 00:26:33.540 --> 00:26:34.540 You know? 00:26:34.540 --> 00:26:34.540 Right. 00:26:34.540 --> 00:26:49.540 Like there's a lot of it's, it's a little bit of a, a case study and yes, you want to be very thorough, but also good naming goes a really long way to like limit the amount of comments and docs you got to put onto a thing. 00:26:49.540 --> 00:26:50.540 There are times when it makes sense though. 00:26:50.540 --> 00:26:55.540 Like if, you're talking about, range or. 00:26:55.540 --> 00:26:56.540 something like that. 00:26:56.540 --> 00:27:00.540 is it inclusive of both, both numbers? 00:27:00.540 --> 00:27:05.540 and if I say one to 10, do I get one, two, three, four up to 10 or I get one, two, three, four up to nine. 00:27:05.540 --> 00:27:05.540 Right. 00:27:05.540 --> 00:27:12.540 Like those, those situations where you might need to say the non-inclusive upper bound of the rain. 00:27:12.540 --> 00:27:13.540 I don't know, whatever. 00:27:13.540 --> 00:27:14.540 Something like that. 00:27:14.540 --> 00:27:14.540 Right. 00:27:14.540 --> 00:27:15.540 Yeah. 00:27:15.540 --> 00:27:15.540 Yeah. 00:27:15.540 --> 00:27:19.540 I do like, an explanation of what's returned though. 00:27:19.540 --> 00:27:20.540 often it's not obvious. 00:27:20.540 --> 00:27:31.540 and it's, even if you are doing a type hint and you can get the type of what's returned, what's the meaning of what's returned is if that's not obvious, please put that in a doc string. 00:27:31.540 --> 00:27:32.540 But yeah, anyway, cool. 00:27:32.540 --> 00:27:35.540 I wonder, I don't, I, this is an honest question. 00:27:35.540 --> 00:27:41.540 I have no idea if you express it in any of this documentation or if the editors consume it. 00:27:41.540 --> 00:27:48.540 But what would be really awesome is if there was a way to express all possible exception types and the entire call stack, right? 00:27:48.540 --> 00:27:57.540 Like you could get a value error, you could get a database connection error, or you could get a uniqueness constraint exception with any of those three. 00:27:57.540 --> 00:28:00.540 Then you could have editors where you just hit like alt enter, right? 00:28:00.540 --> 00:28:02.540 The error handling goes, bam, bam, bam. 00:28:02.540 --> 00:28:05.540 Here's the three types of things you might catch. 00:28:05.540 --> 00:28:05.540 Right. 00:28:05.540 --> 00:28:06.540 That would be awesome. 00:28:06.540 --> 00:28:11.540 But I don't know if you can express the possible range of exceptions in there or not. 00:28:11.540 --> 00:28:13.540 Or unless you've, yeah. 00:28:13.540 --> 00:28:17.540 And especially if you're calling any extra functions within, within a function. 00:28:17.540 --> 00:28:18.540 Yeah. 00:28:18.540 --> 00:28:21.540 You don't know if it's going to raise an exception possibly, but. 00:28:21.540 --> 00:28:22.540 Possibly. 00:28:22.540 --> 00:28:23.540 Yeah. 00:28:23.540 --> 00:28:28.540 Anyway, that's something I would see actually really useful there that you don't express in like the type information or the name or any of those things. 00:28:28.540 --> 00:28:29.540 Yeah. 00:28:29.540 --> 00:28:30.540 Cool. 00:28:30.540 --> 00:28:31.540 Yeah. 00:28:31.540 --> 00:28:32.540 Cool. 00:28:32.540 --> 00:28:33.540 Well, those are our items. 00:28:33.540 --> 00:28:34.540 Michael, do you have any extras for us? 00:28:34.540 --> 00:28:37.540 I have an extra for you in particular. 00:28:37.540 --> 00:28:38.540 How about that? 00:28:38.540 --> 00:28:39.540 Okay. 00:28:39.540 --> 00:28:40.540 Let's start with that one then. 00:28:40.540 --> 00:28:44.540 So last week you asked about GitHub releases. 00:28:44.540 --> 00:28:45.540 Who uses these? 00:28:45.540 --> 00:28:46.540 Should I be bothered? 00:28:46.540 --> 00:28:47.540 Yeah. 00:28:47.540 --> 00:28:49.540 There's this person that seems to be telling everyone on GitHub. 00:28:49.540 --> 00:28:51.540 They should use releases if they're not. 00:28:51.540 --> 00:28:52.540 Do I care? 00:28:52.540 --> 00:28:58.540 And Rhett Turnbull, who's been on talk Python to talk about building Mac apps or with Python. 00:28:58.540 --> 00:29:01.540 GitHub said there said GitHub releases questions. 00:29:01.540 --> 00:29:06.540 I use them and I like them so people can subscribe to be notified of new releases. 00:29:06.540 --> 00:29:10.540 I use GH, the GitHub command line. 00:29:10.540 --> 00:29:14.540 GitHub release create to create one of the command line every time I pushed up IPI. 00:29:14.540 --> 00:29:18.540 I'm sure this can be done as an action, but I don't push that often. 00:29:18.540 --> 00:29:19.540 So it's fine with me. 00:29:19.540 --> 00:29:21.540 Anyway, there's some feedback for you. 00:29:21.540 --> 00:29:22.540 Thanks. 00:29:22.540 --> 00:29:27.540 Yeah, I actually got quite a few people reaching out and I really appreciate it. 00:29:27.540 --> 00:29:38.540 And it did convince me that I'm going to start trying to figure it out using GitHub releases, but I also want to make sure that it's automated as much as possible. 00:29:38.540 --> 00:29:40.540 I don't want to add redundant work just for the heck of it. 00:29:40.540 --> 00:29:45.540 So you're going to set up some automation to go around and tell everyone on GitHub who doesn't have releases going yet? 00:29:45.540 --> 00:29:46.540 No. 00:29:46.540 --> 00:29:47.540 They should do releases. 00:29:47.540 --> 00:29:50.540 Think of all the contributor badges you're going to get. 00:29:50.540 --> 00:29:52.540 Yeah. 00:29:52.540 --> 00:29:53.540 I'm just kidding. 00:29:53.540 --> 00:29:57.540 All right. 00:29:57.540 --> 00:29:58.540 Let's talk about one more thing. 00:29:58.540 --> 00:30:09.540 We've heard about IPI issues where people are uploading malicious packages and a lot of times it's crypto kitties and other idiots who are doing that or researchers to like just prove a concept that it's not going to be. 00:30:09.540 --> 00:30:10.540 Prove of concept that it can be done. 00:30:10.540 --> 00:30:14.540 But Lazarus hackers who are I'm pretty sure. 00:30:14.540 --> 00:30:14.540 Yeah. 00:30:14.540 --> 00:30:25.540 North Korean state sponsored hacking group is uploaded a fake VMware VM connect library targeting it professionals. 00:30:25.540 --> 00:30:29.540 So it only had 237 downloads. 00:30:29.540 --> 00:30:38.540 But when you start to think about state actor hacking level of stuff getting installed onto your machine, that's like a at minimum format. 00:30:38.540 --> 00:30:41.540 The OS maybe just throw it in the trash. 00:30:41.540 --> 00:30:42.540 I don't know. 00:30:42.540 --> 00:30:44.540 It's like pretty bad level of being infected. 00:30:44.540 --> 00:30:45.540 So I don't know. 00:30:45.540 --> 00:30:47.540 That's I have no action or further thoughts. 00:30:47.540 --> 00:30:49.540 Just like a, Hey, that's worth checking out. 00:30:49.540 --> 00:30:50.540 Yeah. 00:30:50.540 --> 00:30:54.540 And maybe we do need to care about our, our, you know, pipeline and whatever. 00:30:54.540 --> 00:31:01.540 But yeah, the supply supply chain, but we do have the new security person, Mike, that was hired. 00:31:01.540 --> 00:31:02.540 Right. 00:31:02.540 --> 00:31:03.540 So that's excellent. 00:31:03.540 --> 00:31:04.540 Yes. 00:31:04.540 --> 00:31:05.540 Yeah. 00:31:05.540 --> 00:31:06.540 He was in the audience. 00:31:06.540 --> 00:31:07.540 So that was great. 00:31:07.540 --> 00:31:08.540 I believe it was Mike. 00:31:08.540 --> 00:31:09.540 Right. 00:31:09.540 --> 00:31:10.540 Hopefully I got the name, right? 00:31:10.540 --> 00:31:11.540 Yeah. 00:31:11.540 --> 00:31:12.540 Over to you. 00:31:12.540 --> 00:31:13.540 That's what I got. 00:31:13.540 --> 00:31:20.540 It's actually this, the, I don't know how to pronounce that J and Y J and Y, on PI, the PI bytes slack. 00:31:20.540 --> 00:31:26.540 we were, talking about using, talking about using TRS 80 computers. 00:31:26.540 --> 00:31:35.540 And I said, Hey, I remember typing in lunar lander on my TRS 80 way back when, copied it out of the back of a magazine. 00:31:35.540 --> 00:31:41.540 And he's, he said, Oh, I've got a copy, copy of, of, lunar lander that works on Python. 00:31:41.540 --> 00:31:42.540 I'm like, Oh, I want to try it. 00:31:42.540 --> 00:31:48.540 And, and I still can't, I'm going to get back to him, but, I can't get his to work. 00:31:48.540 --> 00:31:51.540 And then I looked around and there was this other cool one. 00:31:51.540 --> 00:31:54.540 lunar lander Python. 00:31:54.540 --> 00:32:01.540 I found that's a four years old and apparently it was done as part of a fundamentals of computing course. 00:32:01.540 --> 00:32:03.540 which is, which is pretty impressive. 00:32:03.540 --> 00:32:06.540 I couldn't get it to work, but their website looks great. 00:32:06.540 --> 00:32:08.540 So they have a website with it. 00:32:08.540 --> 00:32:13.540 You got attached to it with, with like a screenshots and it shows good fonts too. 00:32:13.540 --> 00:32:14.540 Yeah. 00:32:14.540 --> 00:32:15.540 Yeah. 00:32:15.540 --> 00:32:18.540 And it looks exactly like the lunar lander that I typed into my TRS 80. 00:32:18.540 --> 00:32:20.540 So I'm pretty excited about that. 00:32:20.540 --> 00:32:22.540 anyway, but I can't get that to work either. 00:32:22.540 --> 00:32:32.540 So if anybody's got like a lunar lander, copy or something that works on, works with modern Python, I would love to, to play with it. 00:32:32.540 --> 00:32:35.540 I also want to like hack with it with my daughter and stuff. 00:32:35.540 --> 00:32:40.540 So anyway, that's the only extra thing I got is, bring on the lunar lander. 00:32:40.540 --> 00:32:41.540 That's really cool. 00:32:41.540 --> 00:32:42.540 I like it. 00:32:42.540 --> 00:32:43.540 Yeah. 00:32:43.540 --> 00:32:44.540 Mike, Mike Felder is here. 00:32:44.540 --> 00:32:45.540 Fidel is here. 00:32:45.540 --> 00:32:50.540 the security guy and people are thinking him and stuff for all the security work. 00:32:50.540 --> 00:32:53.540 So just getting started, but yeah, it's, it's not an easy job. 00:32:53.540 --> 00:32:54.540 I'm sure. 00:32:54.540 --> 00:32:55.540 Yeah. 00:32:55.540 --> 00:32:56.540 And we're pretty excited. 00:32:56.540 --> 00:32:58.540 I can't think of a better person to do this job. 00:32:58.540 --> 00:32:59.540 So indeed. 00:32:59.540 --> 00:33:01.540 So shall we play some bingo? 00:33:01.540 --> 00:33:02.540 Sure. 00:33:02.540 --> 00:33:03.540 All right. 00:33:03.540 --> 00:33:04.540 This is our joke. 00:33:04.540 --> 00:33:05.540 Program or bingo. 00:33:05.540 --> 00:33:06.540 I love it. 00:33:06.540 --> 00:33:07.540 So, you know, bingo works. 00:33:07.540 --> 00:33:10.540 Everybody gets a different card with different options. 00:33:10.540 --> 00:33:16.540 Typically it's numbers, but in this case it's program or actions or statements you call out or have happened. 00:33:16.540 --> 00:33:23.540 And as they get called out, you mark them off and whoever completes a row or column or I dunno, something diagonal. 00:33:23.540 --> 00:33:25.540 I don't play that much bingo, but you win. 00:33:25.540 --> 00:33:26.540 Right. 00:33:26.540 --> 00:33:29.540 And so this is, this is a possible program or bingo card. 00:33:29.540 --> 00:33:32.540 We should come up with one, a whole bunch of them. 00:33:32.540 --> 00:33:34.540 So I'll just read you some of the options out of this card. 00:33:34.540 --> 00:33:35.540 Okay. 00:33:35.540 --> 00:33:36.540 Ryan. 00:33:36.540 --> 00:33:39.540 So we've got number one written code without comments. 00:33:39.540 --> 00:33:43.540 Everybody could check that one off for all of the C inspired language. 00:33:43.540 --> 00:33:46.540 People forgot a semi-colon at the end of a line. 00:33:46.540 --> 00:33:47.540 That's good. 00:33:47.540 --> 00:33:51.540 I can certainly relate with number three, close 12 tabs after fixing an issue. 00:33:51.540 --> 00:33:52.540 Oh yeah. 00:33:52.540 --> 00:33:53.540 Oh yeah. 00:33:53.540 --> 00:33:54.540 Also related. 00:33:54.540 --> 00:33:57.540 Number four, 20 warnings, zero errors. 00:33:57.540 --> 00:33:59.540 Works on my machine, man. 00:33:59.540 --> 00:34:00.540 Yeah, exactly. 00:34:00.540 --> 00:34:02.540 The number five is program. 00:34:02.540 --> 00:34:04.540 Didn't run on someone else's computer. 00:34:04.540 --> 00:34:05.540 Yeah. 00:34:05.540 --> 00:34:08.540 And instantiation of the works on my machine problem. 00:34:08.540 --> 00:34:11.540 And then this number six to do list greater than completed tasks. 00:34:11.540 --> 00:34:13.540 Number seven copied code from stack overflow. 00:34:13.540 --> 00:34:15.540 I'm pretty sure we can all check that one off. 00:34:15.540 --> 00:34:17.540 Close program without saving it. 00:34:17.540 --> 00:34:18.540 Okay. 00:34:18.540 --> 00:34:21.540 Number nine, ask to fix a laptop because you're a programmer. 00:34:21.540 --> 00:34:22.540 I have a problem with my computer. 00:34:22.540 --> 00:34:24.540 Like, please don't, please don't. 00:34:24.540 --> 00:34:26.540 Number 10 turned your bug into a feature. 00:34:26.540 --> 00:34:29.540 11 deleted block of code and regretted it later. 00:34:29.540 --> 00:34:32.540 Finally learned a new programming language, but never used it. 00:34:32.540 --> 00:34:33.540 Hello TypeScript. 00:34:33.540 --> 00:34:35.540 We could come up with so many of these. 00:34:35.540 --> 00:34:36.540 We should, we should totally do. 00:34:36.540 --> 00:34:37.540 We should do more. 00:34:37.540 --> 00:34:38.540 So good. 00:34:38.540 --> 00:34:39.540 Aren't they? 00:34:39.540 --> 00:34:40.540 You can just go on and on. 00:34:40.540 --> 00:34:41.540 Yeah. 00:34:41.540 --> 00:34:42.540 Yeah. 00:34:42.540 --> 00:34:47.540 You can do a backup copy of your code repository, even though it's like a hub. 00:34:47.540 --> 00:34:48.540 Yeah. 00:34:48.540 --> 00:34:50.540 Zip is my source control. 00:34:50.540 --> 00:34:51.540 Yeah. 00:34:51.540 --> 00:34:57.540 And then the, the, the, there's usually a free one in the middle that could just be a need 00:34:57.540 --> 00:34:59.540 to need to update pip. 00:34:59.540 --> 00:35:01.540 That's exactly. 00:35:01.540 --> 00:35:02.540 Pip is out of date. 00:35:02.540 --> 00:35:03.540 Yeah. 00:35:03.540 --> 00:35:04.540 Awesome. 00:35:04.540 --> 00:35:11.540 Well, as usual, pleasure to talk with you, Michael, and thank you so much Sentry for sponsoring 00:35:11.540 --> 00:35:12.540 this episode. 00:35:12.540 --> 00:35:16.540 Again, everybody check out Sentry and go to, what was that link again? 00:35:16.540 --> 00:35:17.540 Pythonbytes.fm/sentry. 00:35:17.540 --> 00:35:18.540 Thanks Brian. 00:35:18.540 --> 00:35:19.540 Thank you. 00:35:19.540 --> 00:35:20.540 Bye. 00:35:20.540 --> 00:35:21.540 See y'all. 00:35:21.540 --> 00:35:22.240 Thank you. Bye.