Top 10 Python One Liners YOU MUST KNOW!
HTML-код
- Опубликовано: 14 ноя 2022
- Top 10 Python One Liners YOU MUST KNOW!
Find Python jobs: pythonengineer.pallet.com
Get my Free NumPy Handbook:
www.python-engineer.com/numpy...
✅ Write cleaner code with Sourcery, instant refactoring suggestions in VS Code & PyCharm: sourcery.ai/?... *
⭐ Join Our Discord : / discord
📓 ML Notebooks available on Patreon:
/ patrickloeber
If you enjoyed this video, please subscribe to the channel:
▶️ : / @patloeber
~~~~~~~~~~~~~~~ CONNECT ~~~~~~~~~~~~~~~
🖥️ Website: www.python-engineer.com
🐦 Twitter - / patloeber
✉️ Newsletter - www.python-engineer.com/newsl...
📸 Instagram - / patloeber
🦾 Discord: / discord
▶️ Subscribe: / @patloeber
~~~~~~~~~~~~~~ SUPPORT ME ~~~~~~~~~~~~~~
🅿 Patreon - / patrickloeber
#Python
----------------------------------------------------------------------------------------------------------
* This is an affiliate link. By clicking on it you will not have any additional costs, instead you will support me and my project. Thank you so much for the support! 🙏
Here’s one: deleting an entry from a dictionary, but only if it exists. Because
del dict[key]
will raise KeyError if there is no such entry. So you have to do an explicit check first, the long way:
if key in dict :
del dict[key]
♯end if
but there is a shorter way:
dict.pop(key, None)
Thats a clever one
You can use the dict.get(key)
If the key doesn't exist it will return None so the code will be like this
If dict.get(key) is not None:
del dict[key]
@@ghaythghannouchi3427 Or "del dict[key] if dict.get(key) else None"
I consider myself very proficient in Python, but it had never occurred to me to use #4: print(*data). I will be using that very frequently from now on.
a
A couple of my favorite one-liners:
# Remove whatever punctuation you want from an input string:
outputStr = "".join(ch for ch in inputStr if ch not in ",.?!'")
# Return the Ordinal Suffix for day of the month: 1st, 2nd, 3rd, 4th, etc.
def ordsuf(day): return ["","st","nd","rd"][day%10] if day%10 in [1,2,3] and day not in [11,12,13] else "th"
# alternative that returns both number and suffix together, takes string, gives string:
def orday(day): return day+'th' if int(day) in [11,12,13] else day+{1:'st',2:'nd',3:'rd'}.get(int(day)%10,'th')
Could you please explain the second one? I'm unable to get it
@@jatindua5866 Sure, that one is a bit tricky to explain, but I'll try. It defines a function called ordinal, which when provided a integer, usually a day from 1-31, the function returns a string of either "st", "nd", "rd", or "th". The part after return is a list containing those, but in a specific order, because we then use [day%10] as an index to determine which of those 4 strings is returned. %10 divides the number by 10, but ignores the result and gets the remainder left over instead, so 23 just becomes 3. The next part, if day%10 in [1,2,3] ensures that list gets used for results 1, 2 and 3. Then "and day not in [11,12,13] excludes those specific numbers because they're exceptions to the rule that use "th" instead. Everything else also just uses "th". Hope that all makes sense..
Functionally, this was just the shortest version, I've also used several variations of that function, depending on whether I want to input an int or a str, or whether I want the output string to include the number first, or not. heh
Here's a slightly longer version, more formally written, that takes a string and outputs a string with the number and suffix already together. Works well with time.strftime("%e"):
def ordinalday(day: str) -> str:
return day+['','st','nd','rd'][int(day)%10] if int(day)%10 in [1,2,3] and day not in ['11','12','13'] else day+'th'
And of course these don't have to be used with days, that's just the most likely use, they can also works with other numbers, like 42nd. ;)
@@NikoKun Thank you for your time and such a clear explanation.
Have a great day ahead!
This is why oneliners are bad, it would have been much much cleaner code if you just allowed more vertical space, for example:
```python
def ordsuf(day):
if 0 < day%10 3 or day in [11, 12, 13]:
return "th"
return ("th", "st", "nd", "rd")[day%10]
```
@@Eknoma Or just import datetime & co. 😎
2:16 AM
11/22/2022
0:42 Important to know: the list comprehension is also MUCH faster faster than iteration approach.
Not a specific oneliner, but the fact that despite its reliance on indentation you can turn any python script into a oneliner is very helpful when sharing code snippets for commandline execution.
1:57 If I remember correctly, you can also provide an argument to the print() function called 'sep', this will be the string that separates the elements (it's a space by default)
So, for example:
print(*data, sep='; ')
Will print
0; 1; 2; 3; 4; 5
Instead of sep I'll use end =';'
@@jaylooppworld381that would add a ";" after your last element too. While sep only adds the separation between elements
Always an insightful video. These types of videos consistently inspire me to post coding content myself! Thank you :)
Go for it!
He Pat, here after a long time. Just wanna let you know that I love your videos. Keep on making the good stuff.
For 4: The `end` argument can be a `
` to print the items in their own lines. I know the tip title says "without new lines" but I use both very often. There's also the `str.join` approach, e.g. `' '.join(data)` but that requires the items to be cast as strings first, if they're not already. Very handy if the joined string is to be used in another area however (i.e. not to be printed)
I am really fond of that one-liner as it's the one that made unpacking click for me. I also use the "equivalent" of it in JS a lot (spread and rest)
For #6 Another alternative is to use reversed(x), which makes an iterator rather that a list, so it's useful when iterating it.
For #8 I'd recommend using a list comprehension instead my_list = [int(x) for x in user_input.split()] , as it's more readable.
For #9 It's a bad practice to use open on a single line rather than using with open(x). Also, strip will remove starting and ending whitespace which is not what you always want.
It's all true.
A small note for #8. Indeed, a PEP recommends the list comprehension over the map call. But the map call is less verbose and more functional oriented ;)
Please do more videos like this. Kind of refresher
Waking up in the morning seeing this as the first video with so much useful information. How can a day be more better than this .
Thank you 😊
Glad you enjoyed it!
Morning sex?
@@patloeber You forgot that you can simultaneously step through two iterables in a single for loop:
for (name,adress) in zip(names, adresses):
print(name, address)
@@RyanClark-gr9yb The correct way for a video about One Liners would be: print([f"{name}, {address}" for name, address in zip(names, addresses)])
Amazing video man, cheers!
Thank you! Cheers!
Very cool video. I've been using python, and am always looking to make life easier. Thanks!
Might fall into the realm of one-liners, but one tip I have is to remember that booleans are a sub class of integer, and thus can be used as the index of a list/tuple, etc.
So for your #3 example, you could also write it as
var = (99, 42)[3 > 2]
I've used that for example when I'm writing a csv, and don't want to used the long "True" and "False" text representations.
boolAsText = "NY"[boolValue]
boolAsText = ("No", "Yes")[boolValue]
etc.
Maybe someone could come up with a better real world use of that.
I often use this when I have to return a specific value in the case the condition it's true or when the condition is false
Thanks so much for sharing! Greetings from Mexico!
Intresting!
Thanks
return a or b
Will return b if a is falsy, else a (this helps if you want a default value b in case a == None or something similar).
Great tips, thanks ☺
amazing work
thank you so much.
Thank you, I love you bc I learn good tricks from you bro. Waves from Bolivia.
Really good video.. thanks!
The dict setdefault method is one that has come in handy for me more times than I can count.
You could also just use list comp for number 8 too [int(i) for i in a.split()]
I think map is faster.
@@mienislav only by a microscopically small amount, e.g. 1usec.
This is reversed when it requires a lambda.
@@DecimasoN yeah, so it is (marginally) faster, AND more readable. That's a win-win
@@Eknoma I prefer the comprehension here bc you have to remember to use both map and then convert back to a list in the other one. The comprehension is easier to read
Unironically a fantastic resource to have.
gr8 video :
btw you can use if statments without using them :
print(["hello","world"][1>2])
The multiple variable assignments ie (a, b = b, a) blew my minds a few weeks ago coming from other languages so handy
Great tips! Tqqqqqq
Great video: A nice one-liner for CSV to JSON: python -c 'import csv, json, sys; print(json.dumps([dict(r) for r in csv.DictReader(sys.stdin)]))'
In list comprehension, you can even use else, but you have to put the if-else before the for, like so:
a = [n if n % 2 else -n for n in some_list]
This will result in a list like [1, -2, 3, -4, 5], from some_list as [1, 2, 3, 4, 5].
And you can do comprehension with other types like dicts too!
A lot of people have pointed out that in 8, you could use a list comp even though map is shorter, more clear, and faster. The real issue it that you gave no safe catch for if the data could not be converted to integers.
I doubt the map is faster. Is it really so? The interpreter should convert a comprehension into a single C call as well as it does with a map call.
If you want to ignore an exception,
instead of try, except use supress exception
#8. For better clarity, we can use list comprehensions like this
a = '1 3 5 6'
b = a.split()
c = [int(i) for i in b]
print(c)
Memory efficiency
Using gerator objects
a = '1 3 5 6'
b = a.split()
c = list(int(i) for i in b)
print(c)
4:08 I would suggest pathlib for reading and writing files these days.
from pathlib import Path; names = Path(“names.txt”).read_text().split()
Why is this better than with open?
@@moahmadi22 read_text closes your file
Helpful 🙏
Glad you think so!
checking zero on None value with 'or'
if not var:
var = 'something'
you can use:
var = var or 'something'
usefull when you expect a certain type and wanna rule out None:
(d or {}).get('element')
(L or [None])[0]
#9 - you need an additional line to close the file
Could use pathlib.Path - `pathlib.Path(filename).read_text().splitlines()`
Will 'with' work?
@@lokeshkalamalla if you use use “with” you don’t need to close it, because it closes itself.
Thats what i immediately thought too
Not true IMHO. The file will be closed when the file object runs out of scope, which is at the end of the line.
func=lambda x : x+1
One liner for a function
the list(map(int, ...)) is fine but I prefer [int(x) for x in ] for readability.
WARNING: ⚠⚠⚠ This video has BAD ADVICE
To read a file please use a "with" statement as such:
with open(filename,"r") as file:
arr = [line.strip() for line in file]
"with" statements will automatically close the file if there is an exception or when the "with" block is done.
Otherwise the file may not be closed right away and can cause weirdness.
Indeed. But it's still a good one-liner if it's used through a command line :)
@@demitri_t wdym command line?
@@gigachad8810 python3 -c 'print(one-liner)'
@@demitri_t bro why would you use python to open files in a cli. just use cat or echo with a pipe. also, that doesnt fix the issue with that one liner outside of clis
@@gigachad8810 There are things difficult to do in bash. But the reason why I mentioned it is at 2:22 ;)
cool! thanks
You should be careful when using that reversing trick, because it creates a copy of your list (at least for the standard List; I think numpy arrays don't do that). You generally want to use `reversed()` instead, which won't copy.
Also: NEVER use a bare `open()`. Only use it in a with statement.
4:06 The "r" mode is default, so ', "r"' can be omitted :)
I would include dict comprehension right after list comprehension.
#2 is also range(0,10,2), instead of the if statement
Hi, do you have videos with regex module?
wonderful gift.
Which theme are you using for your vscode?
Thank you
You're welcome
dict2|=dict1 to merge two dictionaries in one line(3.9+)
excellent!
Thanks!
We can also use eval fn for str list 🙌🏾
eval is EVIL
Hi how to write to a file with oneliners? Thank you!
Thanks
Doest that 9th one liner requires to manually close the opened file? Cause with "with" you wouldn't need that. So does the list comprehension bypass the closing or not?
This is a real deal!
You are the best ;)
This man is just 😎 😎 !
is it possible to use print(*data) to print each value on a new line?
One way to do that is print(*data, sep='
').
After it unpacks the data, the items essentially become different params in the print() call and then you use sep='
' to insert a new line instead of the default space between each item.
I know the point was the one-liner part of it, but opening a file in text mode without explicitly specifying the encoding just makes me shiver. You should never just assume the encoding will be the same as your system default, and if you're the one making the file you should always be able to know exactly what encoding your file is going to be in. Plus having the same script make different files when run on different systems is bad.
can i use these one liners on a date
Good list. My only gripe is that ternaries are expressions, not statements!
But assigning the ternary to a variable does make it a statement
@@elhombreloco3680 True, but the ternary itself is still an expression, that's *why* you can assign its value to a variable
What version of python were you using here? 3.10 or 3.1?
3.11
I always use this trick to print iterables, item by item in a single line:
_ = [print(a) for a in iterable]
You forgot the conditional assignment it s a great technique to use
I usually like one liners, but sometimes i take it too far... like when i made a 200+ letter long one liner... since then i think again if it's worth
2:07 "add a semicolon to stay in the same line" that's one way to make a one liner 😂
hi, does anyone know which extension for vscode he uses so his output updates as he types?😊
Don't I dont know but you can us
import os; os.system('cls')
Put this code in the top level of you code and each time you run the code, the terminal will delete the Last output before the new output is shown in the terminal
useful
"One liners"
"Let's use the semicolon to make it technically a one liner"
you forgot the.. "subscribe" one liner :P
A few of these remind me of excel formulae
Tmatrix = [list(i) for i in zip(*matrix)]
For #2 why wouldn't you just use ... print([i*i for i in range(10) if i %2 ==0]) and then it truly is a one liner. Not declaring a variable and then printing the variable?
does 9. have an implicit context manager?
I was wondering as welll ..
Remember that CPython does reference counting, so objects are automatically disposed when their refcount goes to zero. For file objects, this means the file is automatically closed.
I wouldn’t recommend relying on that for files open for writing, since then you might miss I/O errors because the volume is full etc. For files open readonly, I do this sort of thing all the time.
@@lawrencedoliveiro9104 thank you
Good question! Yes I agree with Lawrence. Python docs say "Do not depend on immediate finalization of objects when they become unreachable (so you should always close files explicitly)."
So my one-liner is not best practice, but probably also doesn't do harm in this case...
See docs: docs.python.org/3/reference/datamodel.html#index-2
@@patloeber thank you and @Lawrence ... Yes, i'd also argue that it isn't best practice, but interesting to know that cpython takes care of this, so from a technical perspective it is indeed "clean" code.
I guess I'm an advanced Python developer because I've used all of those before. I consider myself to be a beginner to Python, so maybe I'm selling myself too short and should start applying for jobs.
Is it all in the... timing?
Importing doesn't count. By that logic, I could make my own module with any function I want, and say that it is 1 line.
Also, I could separate my entire program with semicolons and say it is one line, so that definitely doesn't count.
3:47 this should have been a list comprehension:
my_list = [int(x) for x in user_input.split()]
Using map is much more readable, and almost always faster
@@Eknoma I doubt it's faster. The comprehension should basically be interpreted into a map call (in the lower C level). Otherwise PEP wouldn't recommend comprehensions over map calls.
@@demitri_t It literally takes 2 seconds to check for yourself... why write a comment disregarding a result without even checking?
On my system, list(map(int, s.split())) takes roughly 18% less time than [int(i) for i in s.split()]
There would be literally no point in map if it was slower than a normal for loop... so obviously it has to be faster at some things
@@Eknoma Yep, ~20% is expected for int(). But this is just because you tested the simplest case. Try e.g. lambda i: i * i instead. And I didn't tell anything about a normal loop. Did you test it as well?
P.S. Indeed I didn't profile anything myself, I just found results in the stackoverflow :)
@@demitri_t What are you talking about? Of course using a lambda adds performance overhead...
So you first say you doubt using map for it's intended purpose is fast, and then you later say using it in a bad way is slow, what arr you getting at?
2:13 Don’t you mean, the first day of the next year? So that on 31st December, it will return 1 day left, not 0 days left.
nice catch, yeah I should probably not exclude New Year's Eve
5th is not exactly a oneliner. There are two instructions of code. If we treat this as a one-liner, then we can wrap all Python codes in just one line adding everywhere semicolons.
#5 is actually cheating since you can write the whole core in one line using semicolon instead of a line break. But this video is full of good tips, thank you..
Nice. I'm beginner and I didn't know any of them. Thanks. But what I don't like in your videos 📸....you are too fast for me 😂 I need to watch your videos many times to catch it 😉
Thanks for the video, I just wanted to add this :)
# traditional function with def
```
def sum(a, b):
return a + b
```
# lambda function
``` sum = lambda x, y: x + y
```
Or just use operator.add.
The function one is 99% of the times more readable than using a lambda.
Only use a lambda when it's inline when e.g. calling a function. For example sorted(a, key=lambda x: x[0]) .
8/10 👌
Pls anybody ans why if i write this code will return false
A = 20
A = int(A)
Print(a is int)
you are comparing value to a class, it will only return true in this case:
print(type(a) is int)
this is what you need to check if value is int:
print(instance(a, int))
@@bilaldz9304 Thanks sir
I know them all but anyway I enjoy when I watching it 😂😂
Thanks ♥♥
so in 2 you show how to use list comps but in 8 you miss out on the most obvious use case of them? a list comp would look much nicer than a map here imo
Reading a file in one line isn't great practice. you do need to close the file when you are done
very fast paced. Wouldn't hurt to give the people a decent amount of time to read without rewinding the video.
And since 1 and 7 are the same, its kinda cheated :D
Otherwise, nice vid
ternary == danger
You're really good but a bit fast. Your presentation always moves quite fast. If you could adjust the speed a but that would allow me to follow you more easily.
There are 2 types of people that clicked this video...
I am both
pass
👏👏👏
👍
value = not value
Using a semi colon in number five is both cheating and disgusting. Assuming datetime was already imported on a different line however, it is fine
You just commited a crime by using semicolons in python
one-liners are not cool and against PEP8