WEBVTT 1 00:00:00.080 --> 00:00:03.160 Welcome back to the Deep Dive. Today, we are focusing 2 00:00:03.240 --> 00:00:05.559 on a tool that I suspect a huge portion of 3 00:00:05.599 --> 00:00:09.160 our listeners use well every single day. Oh yeah, it's 4 00:00:09.240 --> 00:00:13.439 the Swiss Army Knife of coding. It powers everything from 5 00:00:14.160 --> 00:00:18.320 the little script that organizes your messy desktop all the 6 00:00:18.399 --> 00:00:21.920 way to neural networks running the most advanced AI. 7 00:00:22.480 --> 00:00:25.120 We are, of course talking about Python. It really is 8 00:00:25.160 --> 00:00:28.079 the language of the moment. You see it everywhere web development, 9 00:00:28.280 --> 00:00:31.719 data science, automation. I mean, it's just completely taken over. 10 00:00:31.839 --> 00:00:34.159 But here's the catch, and this is specifically why we 11 00:00:34.280 --> 00:00:38.200 chose today's source material. Because Python is so readable, so 12 00:00:38.399 --> 00:00:42.320 beginner friendly. There's this massive trap people fall into. You 13 00:00:42.359 --> 00:00:44.799 write a script, it runs, you get the output you wanted, 14 00:00:44.840 --> 00:00:47.520 and you think, great, I know Python. I'm a Python developer. 15 00:00:47.640 --> 00:00:50.200 Ah, the classic Dunning Krueger effect. It's so common with 16 00:00:50.240 --> 00:00:52.200 Python because that barrier to entry is just so. 17 00:00:52.280 --> 00:00:55.280 Low, exactly. And then you walk into that technical interview. 18 00:00:55.640 --> 00:00:58.759 The interviewer leans back and asks, so tell me about 19 00:00:58.759 --> 00:01:00.200 the global interpreter. 20 00:00:59.759 --> 00:01:04.439 Lock, or how does Python actually handle garbage collection in 21 00:01:04.480 --> 00:01:05.359 its private heap? 22 00:01:05.599 --> 00:01:08.640 And suddenly you realize, I don't know Python. I just 23 00:01:08.680 --> 00:01:10.079 know how to use Python syntax. 24 00:01:10.439 --> 00:01:13.799 That is a very, very painful distinction to realize in 25 00:01:13.920 --> 00:01:17.560 real time during an interview. It's that difference between scripting 26 00:01:17.599 --> 00:01:18.359 and engineering. 27 00:01:18.760 --> 00:01:21.200 So to save you from that pain, we're doing a 28 00:01:21.200 --> 00:01:26.359 deep dive into Python interview questions by Swati Saxena. This 29 00:01:26.519 --> 00:01:30.840 is a comprehensive guide from BPB Publications. Look, this isn't 30 00:01:30.879 --> 00:01:32.879 just some list of gotcha questions to memorize. 31 00:01:32.959 --> 00:01:35.439 No it's not. It's really a tour of the language's 32 00:01:35.439 --> 00:01:38.799 internal architecture. It bridges that gap between I can make 33 00:01:38.799 --> 00:01:40.519 it work and I understand how it works. 34 00:01:40.640 --> 00:01:40.840 Right. 35 00:01:40.920 --> 00:01:42.680 We're going to look at the memory model, the object 36 00:01:42.719 --> 00:01:46.840 oriented pillars, even gy development, and some of those specific 37 00:01:47.000 --> 00:01:51.319 performance bottlenecks like the dogpile effect that can crash your 38 00:01:51.359 --> 00:01:52.799 system if you're not careful. 39 00:01:52.959 --> 00:01:55.000 So if you're a developer looking to level up, or 40 00:01:55.079 --> 00:01:57.200 just someone who wants to know what's actually happening when 41 00:01:57.239 --> 00:01:59.280 you hit run, this one's for you. We'll start at 42 00:01:59.280 --> 00:02:02.519 the very beginning, the origin story. Okay, because I have 43 00:02:02.560 --> 00:02:05.000 to admit, for the longest time I looked at that 44 00:02:05.040 --> 00:02:10.240 Python logo the two snakes, and I just assumed, hmmm, well, snakes. 45 00:02:10.039 --> 00:02:13.199 It's the most reasonable assumption in the world. But as 46 00:02:13.280 --> 00:02:15.599 Vaccina points out right at the start, it's wrong. 47 00:02:15.479 --> 00:02:19.520 Totally wrong. Guido van Rossum, the creator, he released this 48 00:02:19.560 --> 00:02:22.199 back in nineteen ninety one. He wasn't into reptiles. He 49 00:02:22.240 --> 00:02:23.639 was into British comedy right. 50 00:02:23.680 --> 00:02:26.960 He was reading scripts from Monty Python's Flying Circus while 51 00:02:26.960 --> 00:02:30.599 he was developing it. He needed a name that was short, unique, 52 00:02:30.840 --> 00:02:32.199 and a little bit mysterious. 53 00:02:32.280 --> 00:02:34.879 So Python it was, which explains a lot about the 54 00:02:34.919 --> 00:02:35.840 culture of the language. 55 00:02:35.840 --> 00:02:38.360 Actually it does. It's supposed to be fun, although spam 56 00:02:38.439 --> 00:02:42.719 and eggs, variables and tutorials instead of foo and bar exactly. 57 00:02:43.319 --> 00:02:46.479 But let's get serious about the definitions. The book defines 58 00:02:46.560 --> 00:02:52.879 Python as an interpreted high level object oriented language, almost 59 00:02:52.919 --> 00:02:56.680 up on that first word interpreted, because for someone coming 60 00:02:56.680 --> 00:02:59.120 from C or C plus plus a, this is a 61 00:02:59.159 --> 00:02:59.719 major shift. 62 00:03:00.039 --> 00:03:02.439 Really is in a compiled language L A C plus 63 00:03:02.560 --> 00:03:04.879 plus A. You write your code, you run it through 64 00:03:04.879 --> 00:03:07.719 a compiler, and it spits out machine code zeros and 65 00:03:07.800 --> 00:03:11.360 ones that the processor crunches directly, super fast, incredibly fast. 66 00:03:11.400 --> 00:03:14.840 But that compilation step takes time. Python doesn't really do that. 67 00:03:14.919 --> 00:03:17.479 It adds a layer when you run a script. You're 68 00:03:17.520 --> 00:03:21.080 actually running the interpreter, which reads your code line by 69 00:03:21.120 --> 00:03:22.680 line and executes it. 70 00:03:23.120 --> 00:03:26.039 Wait, I've seen these files of my project's folders. I 71 00:03:26.080 --> 00:03:28.800 write a dot pi file, but sometimes a dot pi 72 00:03:28.879 --> 00:03:31.520 file appears next to it, or in that pie cash folder. 73 00:03:31.919 --> 00:03:33.919 I usually just ignore them or add them to my 74 00:03:34.000 --> 00:03:36.360 dot j you'd ignore, but the author makes a point 75 00:03:36.360 --> 00:03:37.479 that these are actually important. 76 00:03:37.639 --> 00:03:40.319 They are and this is a nuance that catches people out. 77 00:03:40.960 --> 00:03:44.560 Python is interpreted, but there is a compilation step happening 78 00:03:44.599 --> 00:03:47.520 in the background. Okay, the interpreter doesn't just read your 79 00:03:47.520 --> 00:03:50.400 english looking text and guess what to do. It first 80 00:03:50.439 --> 00:03:53.000 compiles your dot PI file into something called bytecode. 81 00:03:53.120 --> 00:03:55.039 Ah, and that's the dot PI file. 82 00:03:55.159 --> 00:03:56.520 That is what's in the dot PI file. 83 00:03:56.560 --> 00:03:58.560 It's like an intermediate language exactly. 84 00:03:58.599 --> 00:04:02.280 It's a low level plus platform independent version of your code. 85 00:04:02.439 --> 00:04:06.719 The Python Virtual Machine, the PVM, reads that ltecode. Think 86 00:04:06.759 --> 00:04:09.759 of it like a chef prepping all the ingredients. 87 00:04:09.280 --> 00:04:11.560 Before cooking, right the meuson plas. 88 00:04:11.800 --> 00:04:15.479 Yes, the dot pike is the chopped vegetables. So next 89 00:04:15.479 --> 00:04:17.759 time you run the program, if the code hasn't changed, 90 00:04:18.000 --> 00:04:20.480 Python skips the prep and goes straight to cooking. It 91 00:04:20.560 --> 00:04:21.360 just starts faster. 92 00:04:21.480 --> 00:04:24.600 Okay, that makes sense, a hybrid approach for efficiency. Now, 93 00:04:24.639 --> 00:04:27.759 speaking of writing the code, let's talk about the thing 94 00:04:27.800 --> 00:04:30.720 that drives non Python developers absolutely crazy. 95 00:04:30.800 --> 00:04:32.600 Oh yes, the white space. 96 00:04:32.319 --> 00:04:35.920 War, indentation and other languages just you know, good manners. 97 00:04:36.160 --> 00:04:38.519 You do it to your coworkers, don't hate you, but 98 00:04:38.639 --> 00:04:39.639 the computer doesn't care. 99 00:04:39.879 --> 00:04:42.920 Right in Java or C plus plus war, you use 100 00:04:43.000 --> 00:04:46.000 curly braces to tell the computer this loop starts here 101 00:04:46.079 --> 00:04:49.000 and it ends here. Python says, no, we don't do that. 102 00:04:49.040 --> 00:04:52.879 Here, we do spaces in Python. Indentation is the syntax. 103 00:04:53.680 --> 00:04:55.920 The book is very strict about this. It says it 104 00:04:55.959 --> 00:04:57.680 specifies a block of code. 105 00:04:57.759 --> 00:05:00.839 It forces you to write readable code. If you're spacing 106 00:05:00.920 --> 00:05:03.079 is off. The interpreter doesn't just think it's ugly. It 107 00:05:03.120 --> 00:05:06.360 throws an error. It literally won't understand where a loop ends. 108 00:05:06.720 --> 00:05:10.279 It's a design choice that prioritizes human readability over almost 109 00:05:10.319 --> 00:05:10.879 everything else. 110 00:05:11.000 --> 00:05:12.879 Let's go deeper. Let's talk about this stuff. We can't 111 00:05:12.920 --> 00:05:16.480 see memory management. This is where the interview questions start 112 00:05:16.480 --> 00:05:21.639 getting really technical. The source mentions the Python private heap. 113 00:05:21.839 --> 00:05:24.480 This is a critical concept in a language like C. 114 00:05:24.959 --> 00:05:28.040 You're often manually grabbing memory from the system and then 115 00:05:28.120 --> 00:05:30.480 releasing it. It's like renting a hotel room. You check in, 116 00:05:30.600 --> 00:05:31.720 you check out, and if. 117 00:05:31.639 --> 00:05:33.480 You forget to check out, you've got a memory leak. 118 00:05:33.519 --> 00:05:36.160 You've got a memory leak. Python acts more like a 119 00:05:36.240 --> 00:05:38.879 high end travel agent. It handles the bookings for you. 120 00:05:39.079 --> 00:05:42.079 All your objects, your numbers, your lists, they all live 121 00:05:42.120 --> 00:05:45.839 in this private heap. The key phrase Saxena uses is 122 00:05:45.879 --> 00:05:47.839 that the programmer has no access to. 123 00:05:47.800 --> 00:05:50.800 This, so it's a gated community, strictly gated. 124 00:05:50.920 --> 00:05:54.480 The Python interpreter manages the whole thing. You create a variable, 125 00:05:54.720 --> 00:05:57.680 it finds space, you stop using it, it cleans it up. 126 00:05:57.879 --> 00:06:00.879 So how does it know when I'm done with something? 127 00:06:00.920 --> 00:06:03.279 I mean, if I create a huge list of data 128 00:06:03.680 --> 00:06:05.600 and then I just move on, how does Python know 129 00:06:05.720 --> 00:06:10.079 it can delete that list? The book talks about reference counting, right. 130 00:06:09.920 --> 00:06:13.279 This is the mechanism. Imagine every single object you create 131 00:06:13.360 --> 00:06:15.240 has a little counter attached to it. So you say 132 00:06:15.439 --> 00:06:18.600 x gos one, two, three, That list now has one reference. 133 00:06:18.680 --> 00:06:22.199 The count is one. If you then say why equals x, 134 00:06:22.319 --> 00:06:25.240 Now two variables point to it, the count is two. 135 00:06:25.560 --> 00:06:27.399 It's tracking popularity exactly. 136 00:06:27.839 --> 00:06:30.600 Now if you reassign X and say ex eques Hello, 137 00:06:31.199 --> 00:06:33.920 X isn't pointing to the list anymore. The list's count 138 00:06:34.000 --> 00:06:36.680 drops back to one. If you then delete, why, the 139 00:06:36.720 --> 00:06:37.319 count hits. 140 00:06:37.240 --> 00:06:39.120 Zero and zero means death. 141 00:06:39.480 --> 00:06:43.639 Zero means the garbage collector swoops in immediately. It sees 142 00:06:43.720 --> 00:06:47.240 the zero, reclaims that memory, and scrubs the spot for 143 00:06:47.279 --> 00:06:49.959 the next object. It's very efficient for most cases because 144 00:06:49.959 --> 00:06:51.639 it all just happens. You don't have to think about it. 145 00:06:51.759 --> 00:06:53.240 I like that it's like a rum, but that just 146 00:06:53.360 --> 00:06:56.759 runs constantly. Yeah, okay, let's move to data structures. This 147 00:06:56.839 --> 00:07:00.199 is the bread and butter of coding interviews, specifically the 148 00:07:00.240 --> 00:07:02.120 war between lists and tupples. 149 00:07:02.319 --> 00:07:05.160 You will get asked this, guaranteed. It's probably the most 150 00:07:05.160 --> 00:07:07.199 standard warm up technical question there is. 151 00:07:07.319 --> 00:07:09.639 On the surface, they look almost the same. A list 152 00:07:09.720 --> 00:07:11.199 is a bunch of things in order. A tuple is 153 00:07:11.199 --> 00:07:14.360 a bunch of things in order. But the book emphasizes mutability. 154 00:07:14.560 --> 00:07:18.279 That's the divider. A list defined with square brackets is mutable. 155 00:07:18.319 --> 00:07:20.199 You can change it, you can add to it, delete 156 00:07:20.240 --> 00:07:22.480 from it, swap items around. It's dynamic. 157 00:07:22.560 --> 00:07:23.439 And the tupole. 158 00:07:23.360 --> 00:07:28.040 Defined with corientheses and it is immutable sealed. Once you 159 00:07:28.079 --> 00:07:30.399 create it, you cannot change a single thing about it. 160 00:07:30.480 --> 00:07:33.240 Okay, here's where I play Dell's advocate, and I imagine 161 00:07:33.240 --> 00:07:35.879 a listener asking this too. If a list can do 162 00:07:36.079 --> 00:07:40.079 everything a tuple can do, plus be changed, why would 163 00:07:40.079 --> 00:07:41.959 I ever use a tuple? It just sounds like a 164 00:07:42.000 --> 00:07:43.319 list with fewer features. 165 00:07:43.480 --> 00:07:45.839 That's a great question, and it's the follow up an 166 00:07:45.959 --> 00:07:48.720 interviewer will ask to see if you understand system design. 167 00:07:49.000 --> 00:07:53.680 There are two main reasons, safety and performance. Okay, safety wise, 168 00:07:53.959 --> 00:07:57.160 think of map coordinates or server settings. You don't want 169 00:07:57.160 --> 00:07:59.720 those to change by accident. Using a tuple is a 170 00:07:59.759 --> 00:08:01.639 way of saying this is constant, don't touch. 171 00:08:01.839 --> 00:08:03.959 It's a right protection sticker exactly. 172 00:08:04.319 --> 00:08:08.279 And then there's performance. Because tuples are static, Python can 173 00:08:08.360 --> 00:08:11.759 optimize them better. They use less memory, and they're faster 174 00:08:11.839 --> 00:08:14.720 to iterate over than lists. So for a massive data set, 175 00:08:14.759 --> 00:08:18.279 you're only reading from a tupol is the smarter, faster choice. 176 00:08:18.360 --> 00:08:20.879 Okay, that clicks. Now, there's a function in the book 177 00:08:20.920 --> 00:08:22.639 that I wanted to highlight because I actually use it 178 00:08:22.639 --> 00:08:24.199 all the time and it just makes you look like 179 00:08:24.240 --> 00:08:25.279 a pro DoD zip. 180 00:08:25.319 --> 00:08:28.639 Zip is fantastic. It's one of those really pythonic tools 181 00:08:28.839 --> 00:08:31.319 that saves you from writing a clunky four loop. 182 00:08:31.399 --> 00:08:34.559 The book describes it as taking multiple iterables and mapping 183 00:08:34.559 --> 00:08:35.080 them together. 184 00:08:35.279 --> 00:08:38.279 The zipper analogy is perfect. Imagine you have the left 185 00:08:38.320 --> 00:08:42.080 side of a zipper that's a list of names, Alice, Bob, Charlie. 186 00:08:42.320 --> 00:08:45.200 The right side is a list of ages twenty five, 187 00:08:45.360 --> 00:08:47.759 thirty thirty five, and you just zip them up. You 188 00:08:47.799 --> 00:08:51.159 pull the slider and the teeth lock. Alice with twenty five, 189 00:08:51.559 --> 00:08:54.600 Bob with thirty, Charlie with thirty five. It creates a 190 00:08:54.600 --> 00:08:55.960 new object made of those bears. 191 00:08:56.080 --> 00:08:58.399 But what if my lists are messy? What if I 192 00:08:58.399 --> 00:09:01.919 have five names but only three eight ages? Does it crash? 193 00:09:02.080 --> 00:09:04.799 It doesn't crash, but it does something very specific. You 194 00:09:04.840 --> 00:09:07.240 need to know. The zipper stops at the shorder list. 195 00:09:07.360 --> 00:09:09.960 Those extra two names are just left hanging. They get ignored. 196 00:09:10.120 --> 00:09:14.000 That's a good gotcha to remember. It truncates one more 197 00:09:14.039 --> 00:09:17.279 structure before we move on sets. The book mentioned set 198 00:09:17.360 --> 00:09:18.159 and frozen set. 199 00:09:18.399 --> 00:09:20.679 A set is just like the mathematical concept. It's an 200 00:09:20.720 --> 00:09:24.360 unordered collection, and the key thing is no duplicates. If 201 00:09:24.399 --> 00:09:26.399 you have a list with ten apples and you convert 202 00:09:26.440 --> 00:09:28.440 it to a set, you just get one apple. It's 203 00:09:28.480 --> 00:09:30.879 the fastest way to get unique items from data. 204 00:09:30.919 --> 00:09:31.799 And a frozen set just. 205 00:09:31.799 --> 00:09:34.039 Apply the logical use for tupls. A frozen set is 206 00:09:34.039 --> 00:09:37.279 an immutable set. You create it and then you can't 207 00:09:37.279 --> 00:09:40.240 add or remove anything. It's frozen. This is useful because 208 00:09:40.320 --> 00:09:43.320 only immutable objects can be used as keys in a dictionary. 209 00:09:43.399 --> 00:09:47.080 Ah I see Okay. Moving into object oriented programming, the 210 00:09:47.159 --> 00:09:51.080 source makes a bold claim. In Python, everything is an object. 211 00:09:51.200 --> 00:09:54.039 It really means everything. In some languages you have primitive 212 00:09:54.120 --> 00:09:57.639 types like integers that are just raw data. In Python, 213 00:09:57.879 --> 00:10:01.320 the number five is an object. It has methods. A 214 00:10:01.360 --> 00:10:04.200 string is an object. Even a function is an object. 215 00:10:04.240 --> 00:10:06.080 That is a real head scratcher if you're not used 216 00:10:06.080 --> 00:10:10.200 to it. And with objects come classes. The book details 217 00:10:10.200 --> 00:10:13.519 the life cycle methods specifically in it and Dela in 218 00:10:13.559 --> 00:10:14.120 No Right. 219 00:10:14.600 --> 00:10:17.399 In it is the constructor. It's the first thing that 220 00:10:17.480 --> 00:10:19.679 runs when you create an instance of a class. It 221 00:10:19.720 --> 00:10:22.440 sets the stage, gives the object its initial state, and 222 00:10:22.519 --> 00:10:25.039 dell the destructor. It runs when the object is all 223 00:10:25.120 --> 00:10:25.759 to be destroyed. 224 00:10:25.919 --> 00:10:28.080 But the expert note in the book suggests we shouldn't 225 00:10:28.080 --> 00:10:29.159 really rely on dell it. 226 00:10:29.240 --> 00:10:32.000 Why is that because of that automatic garbage collection we 227 00:10:32.039 --> 00:10:34.759 talked about, you don't always know exactly when the garbage 228 00:10:34.799 --> 00:10:36.840 collector is going to run. So if you put critical 229 00:10:36.879 --> 00:10:41.480 logic in del like close this database connection, it might 230 00:10:41.519 --> 00:10:44.200 not run when you expect it to. Not reliable, not 231 00:10:44.279 --> 00:10:47.559 reliable enough. It's better to handle that kind of cleanup explicitly. 232 00:10:47.799 --> 00:10:50.639 That's a pro tip. Don't trust the garbage man to 233 00:10:50.639 --> 00:10:55.039 lock your front door. Okay. Inheritance, the book talks about 234 00:10:55.159 --> 00:10:56.399 using the super command. 235 00:10:56.759 --> 00:11:00.200 Super is how you access the parent class. So if 236 00:11:00.200 --> 00:11:02.480 you have a generic class called shape and you build 237 00:11:02.480 --> 00:11:05.759 a specific class called square, you might want to reuse 238 00:11:05.799 --> 00:11:08.960 logic that already exists in shape. Use super to call 239 00:11:09.039 --> 00:11:11.279 up to the parent so you're not rewriting cut exactly. 240 00:11:11.320 --> 00:11:14.200 It keeps things dr don't repeat yourself. 241 00:11:14.240 --> 00:11:16.440 Now here's a question that I think I have failed 242 00:11:16.480 --> 00:11:21.519 in a mock interview before. The difference between is and agle. 243 00:11:22.480 --> 00:11:24.840 They both sound like they're checking if things are equal. 244 00:11:25.000 --> 00:11:29.240 This is the ultimate trap. In English, is and equals 245 00:11:29.279 --> 00:11:32.639 are synonyms. In Python, they are totally different operations. 246 00:11:32.679 --> 00:11:34.519 Break it down for me, Okay. 247 00:11:34.440 --> 00:11:38.200 Checks for value, it asks does the thing in box 248 00:11:38.279 --> 00:11:40.960 A look exactly like the thing in box B. If 249 00:11:41.000 --> 00:11:43.919 you have two lists both containing one, two, three, will 250 00:11:43.919 --> 00:11:45.080 say yes, these are equal. 251 00:11:45.120 --> 00:11:45.840 Okay, that makes sense. 252 00:11:45.879 --> 00:11:48.679 It's checks for identity. It asks is this actually the 253 00:11:48.679 --> 00:11:51.039 same box. It's checking the memory address. 254 00:11:51.159 --> 00:11:54.440 So going back to those two lists with one, two three, right. 255 00:11:54.320 --> 00:11:56.720 So list A is true, but list A is List 256 00:11:56.759 --> 00:11:59.559 B is false. Ah, they're twins, but they are two 257 00:11:59.600 --> 00:12:01.559 different people living in two different houses. 258 00:12:01.639 --> 00:12:03.919 I love that analogy. Twins are equal, but they are 259 00:12:03.960 --> 00:12:04.279 not the. 260 00:12:04.200 --> 00:12:07.039 Same person precisely. Unless you do list B list A, 261 00:12:07.519 --> 00:12:10.440 then both variables point to the same object. Then is 262 00:12:10.519 --> 00:12:11.399 returns true. 263 00:12:11.679 --> 00:12:16.360 Got it crucial? Okay, let's pivot. A huge chunk of 264 00:12:16.399 --> 00:12:19.000 this book is dedicated to something we can't really show, 265 00:12:19.080 --> 00:12:22.080 but we can visualize GUI programming with. 266 00:12:22.120 --> 00:12:25.440 Kinter right Tinker is the standard GUI toolkit. It comes 267 00:12:25.519 --> 00:12:27.799 bundled with Python, which is why it's so popular for 268 00:12:27.879 --> 00:12:28.919 quick desktop apps. 269 00:12:29.000 --> 00:12:31.919 The book walks through building windows and adding widgets. We 270 00:12:31.960 --> 00:12:35.639 have buttons labels, but it makes a distinction between radio 271 00:12:35.679 --> 00:12:37.200 buttons and check buttons. 272 00:12:37.519 --> 00:12:40.759 Think of filling out a form. A radio button is exclusive, 273 00:12:41.360 --> 00:12:43.600 like a car radio. You can only pick one station 274 00:12:44.000 --> 00:12:47.399 if you click am FMD selects and the check button 275 00:12:47.519 --> 00:12:51.159 that's select all that apply like pizza toppings. You can 276 00:12:51.200 --> 00:12:53.159 have pepperonia and d mushrooms. 277 00:12:53.200 --> 00:12:56.639 Now place these on screen. The source gets into geometry 278 00:12:56.679 --> 00:13:00.440 management and list this very specific string format three hundred 279 00:13:00.440 --> 00:13:02.559 by three hundred plus two hundred plus two hundred. 280 00:13:02.639 --> 00:13:04.840 It looks like a math equation, but it's a coordinate map. 281 00:13:04.879 --> 00:13:07.440 The first part three hundred by three hundred is within height, 282 00:13:07.759 --> 00:13:10.360 the second part plus two hundred plus two hundred is 283 00:13:10.360 --> 00:13:13.519 the position two hundred pixels from the left to the screen, 284 00:13:13.720 --> 00:13:14.799 two hundred from the top. 285 00:13:14.919 --> 00:13:18.000 And then there's resizable zero, which just locks the window. 286 00:13:18.080 --> 00:13:21.000 The zero's act is false, so you're saying no resizing 287 00:13:21.000 --> 00:13:23.279 the wits no resizing the height. The user can't drag 288 00:13:23.320 --> 00:13:23.799 the corners. 289 00:13:23.840 --> 00:13:26.039 Now, making the buttons actually do something, you use the 290 00:13:26.080 --> 00:13:26.879 command parameter. 291 00:13:27.039 --> 00:13:29.600 Right, you link the button click to a function command 292 00:13:29.600 --> 00:13:30.000 my function. 293 00:13:30.120 --> 00:13:33.000 But here's the problem the book solves. Suppose my function 294 00:13:33.080 --> 00:13:36.759 needs arguments like add five to ten. If I write 295 00:13:36.799 --> 00:13:40.000 command add five to ten in the code, Python runs 296 00:13:40.000 --> 00:13:42.159 that function immediately when the program starts. 297 00:13:42.440 --> 00:13:46.120 Yes, exactly, because you put the parentheses, Python executes it right. 298 00:13:46.159 --> 00:13:47.720 Then it doesn't wait for the click. 299 00:13:47.879 --> 00:13:50.759 So how do we fix that. The book suggests something 300 00:13:50.799 --> 00:13:51.919 called partial. 301 00:13:51.879 --> 00:13:54.919 Yes from the Funk tools library. This is a life saver. 302 00:13:55.320 --> 00:13:58.879 Partial basically freezes the function along with its arguments. It 303 00:13:58.919 --> 00:14:03.000 wraps add five ten in a package and says, don't 304 00:14:03.039 --> 00:14:05.159 open this until the button is clicked. 305 00:14:05.279 --> 00:14:07.960 So functals dot partial is the secret sauce for passing 306 00:14:08.120 --> 00:14:09.200 arguments in callbacks. 307 00:14:09.240 --> 00:14:12.960 It's pretty much essential for any complex gy programming. 308 00:14:13.039 --> 00:14:16.960 Okay, we're entering the final stretch advanced concepts. This is 309 00:14:16.960 --> 00:14:19.600 where the seniors get separated from the juniors. Let's talk 310 00:14:19.600 --> 00:14:20.320 about speed. 311 00:14:20.480 --> 00:14:21.799 Everyone wants faster code. 312 00:14:21.919 --> 00:14:24.519 So logically, if I have a powerful computer with eight cores, 313 00:14:24.639 --> 00:14:26.879 as you just use multi threading right split the work up. 314 00:14:26.960 --> 00:14:29.399 You would think so that works beautifully in Java or 315 00:14:29.440 --> 00:14:32.480 C plus plus mating, But the source drops a major 316 00:14:32.519 --> 00:14:36.120 reality check here. For Python, it introduces the GIL. The 317 00:14:36.159 --> 00:14:37.360 Global interpreter lock. 318 00:14:37.639 --> 00:14:38.480 Sounds ominous. 319 00:14:38.840 --> 00:14:42.240 It is if you're trying for parallel processing. The GIL 320 00:14:42.679 --> 00:14:45.759 is a lock that prevents multiple native threads from executing 321 00:14:45.799 --> 00:14:47.519 Python bytecodes at the same time. 322 00:14:47.759 --> 00:14:50.120 Wait, so even if I have multiple threads, they aren't 323 00:14:50.159 --> 00:14:51.120 running at the same time. 324 00:14:51.279 --> 00:14:54.960 Now the GIL forces them to take turns. It happens 325 00:14:54.960 --> 00:14:57.840 so fast that it looks parallel, but the CPU is 326 00:14:57.919 --> 00:15:00.399 only ever doing one thing at a time. It's a 327 00:15:00.440 --> 00:15:01.559 single lane highway. 328 00:15:02.200 --> 00:15:06.519 So for CPU heavy tasks like crunching massive numbers, threading 329 00:15:06.759 --> 00:15:08.120 might actually make things slower. 330 00:15:08.360 --> 00:15:10.960 Yes, because of the overhead of switching back and forth 331 00:15:10.960 --> 00:15:13.759 between the threads, you spend more time managing them than 332 00:15:13.840 --> 00:15:15.039 doing actual work. 333 00:15:15.159 --> 00:15:18.399 That is a massive gotcha. Yeah, So when should we 334 00:15:18.519 --> 00:15:19.720 use threads in Python? 335 00:15:20.000 --> 00:15:22.919 For io tasks waiting for a web page to load, 336 00:15:23.200 --> 00:15:26.320 writing to a file anytime, the program is just waiting. 337 00:15:26.399 --> 00:15:29.000 In those cases, the GIL releases the lock while it 338 00:15:29.039 --> 00:15:32.159 waits so other threads can run. But for raw computation, 339 00:15:32.639 --> 00:15:34.039 the GIL is a bottleneck. 340 00:15:34.200 --> 00:15:36.159 Speaking of bottlenecks, there's a term in the book I 341 00:15:36.200 --> 00:15:37.559 Love the dogpile effect. 342 00:15:37.840 --> 00:15:39.360 It's a vivid image, isn't it? 343 00:15:39.360 --> 00:15:40.159 It is? What is it? 344 00:15:40.320 --> 00:15:42.480 So? Imagine you have a website that shows the weather. 345 00:15:42.639 --> 00:15:45.799 It's a complex calculation that takes say five seconds to run. 346 00:15:45.960 --> 00:15:47.639 You don't want to run it for every user, so 347 00:15:47.679 --> 00:15:49.519 you calculate at once and save it in. 348 00:15:49.480 --> 00:15:53.080 The cash smart Everyone just reads the saved copy, but. 349 00:15:53.080 --> 00:15:57.440 That cash expires. Now, imagine at the exact second it expires. 350 00:15:57.559 --> 00:15:59.759 One hundred users hit the website. 351 00:16:00.000 --> 00:16:01.720 They all see the cash as empty, and. 352 00:16:01.639 --> 00:16:03.919 They all try to run that five second calculation at 353 00:16:03.960 --> 00:16:06.799 the same time, and the server just melts. That's the dogpile, 354 00:16:06.919 --> 00:16:09.879 everyone jumping on the server at once. The CPU spikes 355 00:16:09.879 --> 00:16:12.519 to one hundred percent and the site goes down. 356 00:16:12.840 --> 00:16:16.679 The source suggests using a semaphore lock to fix this. 357 00:16:17.039 --> 00:16:17.840 How does that work. 358 00:16:18.000 --> 00:16:20.399 It's like a bathroom key at a gas station. The 359 00:16:20.440 --> 00:16:23.480 first process sees the cash is empty, grabs the key, 360 00:16:23.679 --> 00:16:26.720 locks the door, and does the calculation. The other ninety 361 00:16:26.759 --> 00:16:29.240 nine processes see the door is locked, and they just wait. 362 00:16:29.600 --> 00:16:31.360 They don't try to calculate it themselves. 363 00:16:31.440 --> 00:16:34.279 Nope, they queue up. When the first one finishes, it 364 00:16:34.360 --> 00:16:37.279 refills the cash, unlocks the door, and everyone else gets 365 00:16:37.279 --> 00:16:38.399 the new data instantly. 366 00:16:38.519 --> 00:16:41.360 That is brilliant. It's not just coding, it's architecture. 367 00:16:41.440 --> 00:16:42.919 That's what senior interviews are about. 368 00:16:42.919 --> 00:16:45.399 And we have just a minute left. Two quick definitions 369 00:16:45.960 --> 00:16:47.039 Lambda versus. 370 00:16:46.879 --> 00:16:51.480 Deaf simple distinction. Def creates a standard function with a name. 371 00:16:51.879 --> 00:16:55.440 It can have multiple lines, loops whatever. Lambda is an 372 00:16:55.440 --> 00:16:58.679 anonymous function, no name, and it can only evaluate a 373 00:16:58.720 --> 00:17:01.960 single expression. It's for quick throwaway logic. 374 00:17:01.879 --> 00:17:04.680 And finally Chathon with a Jay. 375 00:17:04.880 --> 00:17:08.759 It's Python implemented in Java. It compiles Python code into 376 00:17:08.880 --> 00:17:10.680 Java byte code so it can run on the Java 377 00:17:10.759 --> 00:17:14.039 virtual machine. Why would you ever do that interoperability if 378 00:17:14.079 --> 00:17:16.599 you're in a big corporate environment running on a javastack 379 00:17:17.000 --> 00:17:19.759 gython lets you write clean Python scripts that can natively 380 00:17:19.880 --> 00:17:23.559 use Java classes. It's a bridge between worlds. Wow, a 381 00:17:23.680 --> 00:17:24.559 very useful one. 382 00:17:24.599 --> 00:17:26.799 We have covered a massive amount of ground today. We 383 00:17:26.839 --> 00:17:29.160 went from Monty Python to the private heap, learned that 384 00:17:29.200 --> 00:17:32.039 tupples are the safer twins of lists, and learned why 385 00:17:32.079 --> 00:17:34.279 the GIL keeps Python single minded. 386 00:17:34.440 --> 00:17:36.960 It really shows that knowing the syntax is just step one. 387 00:17:37.400 --> 00:17:41.680 Mastering Python means understanding the memory, the interpreter and its limitations. 388 00:17:41.960 --> 00:17:44.559 And that's the takeaway for you, whether you are prepping 389 00:17:44.599 --> 00:17:47.240 for an interview or just building your next app. Don't 390 00:17:47.279 --> 00:17:50.799 just ask does it work? Ask where's this stored in memory? 391 00:17:51.200 --> 00:17:53.480 And what happens if one hundred people try to do 392 00:17:53.519 --> 00:17:54.000 this at once? 393 00:17:54.240 --> 00:17:56.519 That curiosity is what makes a senior engineer. 394 00:17:56.720 --> 00:17:58.640 And I want to leave you with one final thought 395 00:17:58.759 --> 00:18:01.440 based on that dog pile of We live in a 396 00:18:01.480 --> 00:18:04.759 world that demands instant data. We treat our caches in 397 00:18:04.839 --> 00:18:08.759 databases like magic. But as we learn today, when the 398 00:18:08.799 --> 00:18:12.839 magic stops even for a second, the dogpile happens. So 399 00:18:12.920 --> 00:18:15.759 ask yourself, is your code ready for the moment the 400 00:18:15.799 --> 00:18:16.640 cash expires? 401 00:18:16.839 --> 00:18:18.400