I enjoy using it for short lines of code in dictionaries. I mean if you convolute it, of course it might become less readable. But the same thing goes for list comprehensions in the end. For now I suffered no readability problems with the first approach, but I will have to update you in 6 months when I look back at it. If I can read it without any comments, I'll know for sure 😅

The thing is: it's sugar syntax.

I agree. I remember reading there is no n++ operator b/c Guido never mixes expressions and assignments, but I guess that is passé. Moreover, which order is the dict construction evaluated? I guess left to right, but that is non-pythonic code imho. Also: what if you assign to word more than once in the dict constructor? Sounds buggy af.

People used to look at my Perl code and instantly say, "You're a C programmer, aren't you?" YES! Because I like very easy to read code. I like declaring everything clearly. I like for each statement to clearly ahve a purpose. I prefer to not obfuscate the code with incidental assignments or side-effects or "cool tricks" of any kind. Sure, I use extra lines. Here's a secret: The compiler does not care. Here's another reason to not use the walrus operator - if you are that 1 in 1000 python eveloper that actually uses the debugger *gasp* instead of prints everywhere it makes things harder to debug.

I am the walrus

you have to adopt changes Else: you'll be left behind

Just think of all the new operators you'll have in Python 10.0!

@@joseoncrack if im still alive yeh lol

@@KeithKazamaFlick Get ready for the double-Walrus operator: =:= (no it won't be a cat smiley, but an operator). This one will allow bidirectional assignment. Yes, this is groundbreaking. Don't thank me.

Python was originally written to not not have all these handy but complex feature like previous languages. After 30 years people realize the value of those "old" constructs and are adding them back.

This is a really good point.

Oh damn that's a cool one

you should pass x to do_things()....just to show that you actually need it.

Not necessarily. You may use x only next time your program reaches this snippet to check whether x has been changed since last call.

@@epicanthus idk. I make all function pure, so this use case I would avoid. Maybe in a method.

OHH wait it does work! It needs extra parentheses for some reason: print(f'{(absolute := abs(a - b))}')

You need to search for something called "Ligatures"

@@Indently Isn't that the thing you put on instruments to keep the reed on?

@@hiredfiredtired Not in the crime documentaries that I've watched LOL

That's actually my next video coincidentally

I prefer collections.defaultdict, which is different...but usually more appropriate. just won't throw a KeyError.

Koo Koo kachoo!

and I have been doing py since 2002, and the arrow is new to me. As far as Im concerned it's for babies addicted to their IDEs to tell them how to write code. I mean I usually go with new python stuff, but this seems down right unpythonic. Java dev hand holding.

​@@DrDeuteronare you talking about the "->" in the function signature?

@@IManu96I the whole package. I mean when I see: name: str = "Monty" that is not pythonic at all.

@@DrDeuteron There is not any package imported in the line you wrote. Do you mean the use of "typing" package and type hints? What you wrote is redundant, the "str" has no effect there in that specific case. It is assumed that this is a str. And what do you mean with "no pythonic"? Its zen says "better explicit than implicit". Technically adding str is more explicit. In addition, the typing provides 1) extra IDE autocompletion (as the IDE knows the type of the variable), 2) new way of documenting code and 3) tools that can scan your code to verify that there is not anything breaking through static type checking, which is a way of static testing. When developing code for companies, it is a must. Many of them enforce it. It gives you confidence that your code will not break when running it or after a big refactor

well that explains it.

except i'd recommend not naming it f since your function is already named f :D

Out of curiousity, where do you see it overused? In 2 years I've seen it (only barely) a couple of times being used.

@@Indently I was being a little hyperbolic. By overused, I mean using it even once is too much. I just personally don’t like it because in every use case I can think of it reduces readability for the sake of saving one additional line. Edit: Fyi it wasn’t meant to be a dig, I do enjoy your channel. Just wanted to share my opinion, as at the time I saw many comments from beginners that wanted to try using this operator.

I agree. I use it in if statements where I need the value on some branches but not all..esp if I only need it to throw errors with messages. I mean if I need the value for clean execution, I'm happy to devote a line of code.

Because in the example text is the length of the string, not the string itself (that's user_input), so in the output you would need to rewrite print(len(user_input), "emoji"). Just a poorly named variable, should've been called something like "string_length".

@@gregoriopescucci4997 oh yeah, I've just realized it was the string length. Still I don't think it's worth the trouble to create a separate variable, just to avoid mentioning len() twice.

I reject 2 function calls, and would do: print(user_input, 'up' if Len(user_input) > 5 else 'down') no, I would have:\ RESPONSE = {True: 'up', False: 'down'} defined in the appropriate namespace and have 1 execution line: print(ui, RESPONSE[len(ui) > 5]) because I prefer static complexity over dynamic complexity...within reason.

Or Haskell

As far as I personally use it, the whole purpose of Walrus Operator is actually to reduce some annoying template typing. For example, when you use coroutines (implemented by generators), you always have to type over and over again like x=getCoro(); next(x); y=getCoro(); next(y) Or when performing regular matching, m=re.match(...); if m is not None: return ...; m=re.match(...); if m is not None: return . ..; these templates are really intrusive and interfere with readability. The Walrus Operator helps with readability as long as you maintain consistency throughout your source code.

that is nice

But the ternary operator isn’t the walrus operator, unless you’re just talking about the ternary operator with no correlation to the video.

You’re every marketing teams worst nightmare for a business 🤣

:= was used for assignment in Pascal. Maybe others.

Yeah, Python was never used again by anyone after this sugar syntax

Hey there! If you consider the video a waste of time, imagine how others will see your comment. I'm sorry you saw it as a waste of time, maybe you'd like to share a better example so more people can learn from your comment? I mean, the real shame is not being constructive in the comments section to help people :)

@@Indently I've always felt that the best way to explain 'abstract' concepts like the walrus was through the use of examples.