Thanks for all the great content, I have always been enjoying taking them with my breakfast haha. I have a question on something could be quite trivial that is not even the main point of this video. At 16:48, you rewrite your `if` statement into `try except`, why did you choose `try` over `if` in this particular case? I have always been confused about when I should use `try` and when `if`, and a lot of times I find myself using `if` without the need of using `try`, and I can argue that a lot of times I could have used `try` instead. What is your general rule of picking one if the situation allows you to use either?
I wonder - How do you come up and understand the internals of python so good? What do you do rather than read and write a lot of code. It's been a year since I watched this video last and I still can't make up a singleton code from my mind even though I work as software engineer.How do you know when to use cls.inst and when super().__new__(cls) - it feel to me that I could memorize the idea and code from your videos, but I would never know how to use cls for dunder functions such as __new__ or less familiar and used ones like it in order to solve the singleton design pattern of other ideas...
mind=pancake This is some confusing stuff. Seeing you write that metaclass taught me alot about how pythons classes work behind-the-scenes, I'm gonna do some more research on that. I like python for it's "hackability", you can break the language right down and do all sorts of magic with it's internals.
For me singelton or global state is indication that i made something wrong with architecture ot thought process, i avoid it with any cost! But grate video!
I really enjoy your content. Your love for programming is very evident and it's contagious. Thank you so so much.
Watching you code just leaves me at awe... Truly an inspiration!
You are a rare gem. I have encountered such problems before, thanks alot for this video. Love from Nigeria.
glad it was helfpul!
I am very much enjoying your videos. Keep up the good work Anthony!
Thanks for all the great content, I have always been enjoying taking them with my breakfast haha.
I have a question on something could be quite trivial that is not even the main point of this video. At 16:48, you rewrite your `if` statement into `try except`, why did you choose `try` over `if` in this particular case?
I have always been confused about when I should use `try` and when `if`, and a lot of times I find myself using `if` without the need of using `try`, and I can argue that a lot of times I could have used `try` instead. What is your general rule of picking one if the situation allows you to use either?
in that case it's for performance -- make the common path fast. the pattern is called EAFP
@@anthonywritescode Thanks for your reply!
Awesome video. Great tricks. Thank you
Here's how i would do it:
from weakref import WeakValueDictionary
class Singleton(type):
_instances = WeakValueDictionary()
def __call__(cls, *args, **kwargs):
try:
return type(cls)._instances[cls]
except KeyError:
obj = type(cls)._instances[cls] = super(Singleton, cls).__call__(*args, **kwargs)
return obj
@classmethod
def cache_size(mcs):
return len(mcs._instances)
class Foo(metaclass=Singleton):
def __init__(self):
self.x = 1
def do_stuff(self):
return self.x + 1
obj1 = Foo()
obj2 = Foo()
print(obj1 is obj2) # True
print(f"Objects in cache: {Singleton.cache_size()}") # 1
del obj1
del obj2
print(f"Objects in cache: {Singleton.cache_size()}") # 0
What does "mcs" stands for?
@@alexandrugheorghe5610metaclass
great video as always. saved me lots of headaches.
Is there a difference between type(self) and self.__class__?
heheheh th-cam.com/video/6rAIttnm3Fs/w-d-xo.html
Sir plz share more about design patterns in python
Well if this wasn't a really great commentary of python's data model!
More patterns/idioms, please!
I wonder - How do you come up and understand the internals of python so good? What do you do rather than read and write a lot of code. It's been a year since I watched this video last and I still can't make up a singleton code from my mind even though I work as software engineer.How do you know when to use cls.inst and when super().__new__(cls) - it feel to me that I could memorize the idea and code from your videos, but I would never know how to use cls for dunder functions such as __new__ or less familiar and used ones like it in order to solve the singleton design pattern of other ideas...
mostly just trying a lot of things
what about a practical, real-life-code-problem-solving, design patterns series?
I've done those in the past, they get really poor reception on youtube. my twitch streams are basically that though so maybe check those out!
mind=pancake
This is some confusing stuff.
Seeing you write that metaclass taught me alot about how pythons classes work behind-the-scenes, I'm gonna do some more research on that.
I like python for it's "hackability", you can break the language right down and do all sorts of magic with it's internals.
exactly :D python lets you hook into its internals pretty easily when needed
More patterns and design infrastructure please
Singleton is so cursed that at the end it broke anthony scene switcher! hahah
For me singelton or global state is indication that i made something wrong with architecture ot thought process, i avoid it with any cost! But grate video!
class Singleton:
__instance = None
def __new__(cls):
if Singleton.__instance == None:
Singleton.__instance = super().__new__(cls)
return Singleton.__instance
database = Singleton()
database.x = 8
print(database.x)
database2 = Singleton()
print(database2.x)
😮😮😮
TLDR do not ever use singletons.