Name: Pythonによるソケットチュートリアル 3 パート 2 - データのバッファリングとストリーミング (Sockets Tutorial with Python 3 part 2 - buffering and streaming data)
Uploaded: 2021-01-14T10:43:21.000Z
Duration: 14 min 46 s
Description: VoiceTubeの動画で発音を聞きながら英語表現を覚えよう！学べる英語：

Everybody welcome back to another sockets with python tutorial video.

程度の副詞

In this video, what we're really talking about is basically just building on the last video where we reached the point where we realized things were still little clunky when we have messages that are bigger than, um, the buffer size that we selected.

But also, how do we maintain this stream of data?

So because we're finding that the only way that we can finally get our full message is to issue a close which closes the socket and ends the client so we'd have to rerun the client, I say doesn't work.

So we did cover everything you need to send and receive data via socket.

So I'm sure lots of people are gonna leave us behind me like I've got everything I need.

But you probably don't because it's gonna get a lot more complicated than this, and or you're gonna be missing out on a ton of stuff that sockets have to offer you, um, with basically any time you want to send things bigger than your buffer size s o most of the socket tutorials that I've ever seen Don't even talk about how you build the message.

They use a buffer of, like 2048 bites, and they don't talk about what happens when you exceed that.

So anyways, not gonna do that to you guys.

How do you handle sockets that exceed your buffer?

But you don't want to have to close this connection.

You want to keep a stream open and all that generally, what you're gonna do is use a header.

Now, there's a 1,000,000 ways that we could do something with a header tool in.

Basically, a header is gonna notify your program How long is your message and maybe give some other information about your message.

But in this case, we really just want to inform the program.

Hey, I'm about to send you a message that it's 50 characters long.

And then once the program receives that first initial header there, like Okay, I'm gonna wait for till I have 50 characters of data once I've got that boom.

Now I'm gonna be waiting for a header again.

And then the process repeats so there's lots of ways that we could build a header.

You could make a header, have some sort of fixed pattern, like begin hitter, you know, And then maybe end, you know, and message.

You do something like that until and it would work fine.

Until people learned that that's what you used.

And then they would start adding that to their messages just to have a good old time.

So you just always have a header that, you know, proceeds every new message.

And it says, How long is that message going to be?

And then that way your program knows exactly how much the way for then, when it's all done, it re looks for that fixed length header.

You know, nobody can really mess with that.

Well, enter string formatting and python.

So you got a message message and that will be equal to our welcome to the server.

Um, now what we want to do is get the length of that and issue that in a fixed length header.

So you can't just say land message because that might be, you know, if it's under 100 it could be two characters, one character.

If it's greater than 103 characters, we needed to be a fixed length.

So they were gonna do that is first, you have to kind of decide Well, what is, like, the longest message we could ever receive.

So if you're thinking you're going to get a message longer than one billion characters do something even bigger.

So what we're gonna say is Prince, and then this will be f so these f string here and then we're gonna say Len message.

And then, um and this will be colon less, Stan.

And then whatever that header length is in this case, we're gonna say it's 10 So now we'll print that out and you see it says 22.

But what you don't see is the extra characters.

The next thing we could do is we could actually append to it.

And now you can see here it proceeds the message.

Now, if we said 20 it would, you know, proceed with 20 characters.

So again it could be as many as you need.

So you can understand that eventually you you could come up with a number and it's not gonna like, you know, require a huge like your buffer will always be more than enough to handle your header size.

Well, I guess I can't say always, but it should be Okay, Great.

So now well, we want to do is implement this.

Luckily, we have basically done it so at least server side.

So now I'm going to come down here cut, come down here a paste that we'll have that bad boy right on over message will equal the header plus the message.

Don't forget the plus message bit otherwise you're gonna have a bad day.

And also let me changes back to 10 thing.

Other thing I'd like to do is implement a constant that is just going to behead her size.

Um, and then here, we're gonna pass that with another.

Never needed to embed a variable inside of another variable.

Um, but yeah, you can just embed more brackets inside more brackets.

Come over to our client and we're ready to implement this on our client.

So the first thing I would do is share the header size constant.

And now we are gonna continue with a new wow loop.

So again, while true full message, we're gonna tab this over not only when we have full message, but we also want to track new message as a flag.

So the very first message we get is definitely the start of a new message.

We'll continue receiving in small amounts, but eight is way too small.

字幕リスト動画再生

Pythonによるソケットチュートリアル 3 パート 2 - データのバッファリングとストリーミング (Sockets Tutorial with Python 3 part 2 - buffering and streaming data)

awesome

constant

basically

matter