Ask HN: Catching Up with Python – Resources?
12 comments
The Python manual.
There are two kinds of programming languages: (i) where the official manual is clear and complete, and (ii) ones where it is not.
In a type (i) language you should read the manual while spinning at the gym and train yourself to be very fast to look up answers to questions like "how do I generate a random integer in the (3,7) range" in the official manual.
Books, blog posts, etc. tend to inject errors and also become stale over time. For instance there is a lot about Python 3.5 and 3.6.1 that you don't want to know. A Pythoner in 2021 could be working with anything from Py 3.9 to 3.7 or late 3.6.9. A book written to Python 3.7 is going to be a basically good foundation to work on, but I know Py 3.8 and 3.9 feature use is common in the wild and 3.10 is not far away.
If you learn the manual like the back of your hand, you can look up whatever version you need, what the changes are, etc.
There are two kinds of programming languages: (i) where the official manual is clear and complete, and (ii) ones where it is not.
In a type (i) language you should read the manual while spinning at the gym and train yourself to be very fast to look up answers to questions like "how do I generate a random integer in the (3,7) range" in the official manual.
Books, blog posts, etc. tend to inject errors and also become stale over time. For instance there is a lot about Python 3.5 and 3.6.1 that you don't want to know. A Pythoner in 2021 could be working with anything from Py 3.9 to 3.7 or late 3.6.9. A book written to Python 3.7 is going to be a basically good foundation to work on, but I know Py 3.8 and 3.9 feature use is common in the wild and 3.10 is not far away.
If you learn the manual like the back of your hand, you can look up whatever version you need, what the changes are, etc.
What sorts of features from 3.8 and 3.8 have you been seeing regularly used?
Not sure the first version you're referencing...The biggest features that I've seen proliferate quickly in the past few releases are f-strings and dataclasses.
3.8's walrus operator will probably start popping up more once people calm down about how it looks funny. 3.9 has more minor tweaks and tricks. 3.10 will introduce pattern matching (case/switch statement) that python developers have been begging for (I HAVE NO IDEA WHY!!!???) for near a decade.
3.8's walrus operator will probably start popping up more once people calm down about how it looks funny. 3.9 has more minor tweaks and tricks. 3.10 will introduce pattern matching (case/switch statement) that python developers have been begging for (I HAVE NO IDEA WHY!!!???) for near a decade.
I have a collection of resources here: https://learnbyexample.github.io/py_resources/
I'll echo other comments to say that official Python tutorial and documentation is a great place to start. 'Fluent Python' 2nd edition is coming later this year. 'Practices of the Python Pro' and 'Serious Python' are other books that'd fit your request.
I'll echo other comments to say that official Python tutorial and documentation is a great place to start. 'Fluent Python' 2nd edition is coming later this year. 'Practices of the Python Pro' and 'Serious Python' are other books that'd fit your request.
It really depends on what you want to achieve. The documentation is very good but also very verbose and it would take ages to read through all of it.
I’d say pick a project, choose the relevant libraries (for web I’d suggest Django for a full-featured framework, or Flask for a lightweight one) and then learn by trial and error.
A good IDE such as PyCharm will allow you to look into symbols’ underlying implementations which helps a lot in understanding how something works behind the scenes.
I’d say pick a project, choose the relevant libraries (for web I’d suggest Django for a full-featured framework, or Flask for a lightweight one) and then learn by trial and error.
A good IDE such as PyCharm will allow you to look into symbols’ underlying implementations which helps a lot in understanding how something works behind the scenes.
- Python Design Patterns: https://python-patterns.guide/
- Python 3 Patterns, Recipes and Idioms: https://python-3-patterns-idioms-test.readthedocs.io/en/late...
- Effective Python: 59 Specific Ways to Write Better Python https://github.com/SigmaQuan/Better-Python-59-Ways
- Python 3 Patterns, Recipes and Idioms: https://python-3-patterns-idioms-test.readthedocs.io/en/late...
- Effective Python: 59 Specific Ways to Write Better Python https://github.com/SigmaQuan/Better-Python-59-Ways
It depends what you used to do with Python, really. There are a lot of users whose code didn't change from 2.x to 3.x except for the print "statements".
I replied to similar threads[0][1], reproduced here:
"""
The Python documentation is pretty cool. I frequently go to Chapter 3 (Data Model) and always learn something new reading something I had already read but didn't quite get. The mailing list has top notch posters who are always willing to help.
The IRC channel (#python) has very helpful members who've been patient with me on more than one occasion.
Books:
Stacking the deck:
"Python Guide" - Kenneth Reitz: The bits one doesn't necessarily think about but that simplify life: virtual environments, unit test, project structure, good practices, scenarios for different kinds of applications, documentation, etc.
Doing stuff:
"Python Cookbook" (several editions by different authors, Alex Martelli and David Beazley): as the title states, it's a cookbook that shows you how to do a bunch of stuff in Python (files, network, threads, etc). This can get you "operational" in Python pretty fast.
Doing stuff well (meta):
"Writing Idiomatic Python" - Jeff Knupp:
The book comes for Python 2 and Python 3 and contrasts "harmful" and "good". Jeff Knupp also has videos showing you the process of refactoring code. He also has a blog (https://jeffknupp.com/) where he addresses many topics.
"Effective Python: 59 Specific Ways to Write Better Python" - Brett Slatkin: Pretty good book. The subtitle says it all: 59 specific ways to write better Python. (600 + pages).
"Learn Python the Hard Way" - Zed Shaw: I've read that starting out in Python. Maybe not for a 15 year Perl veteran, but this is for memo.
There's certainly some overlap in these books, but it helps to repeat a message to understand the importance of something, or to drive a point home. I didn't use virtual environments until the message was repeated enough in different ways.
Pornography:
PyCon talks. A lot of these are great.
David Beazley (a bunch of talks that are pretty cool. Still haven't finished the ones on generators but they're very good).
Brandon Rhodes (on aesthetics and design patterns (or lack thereof), and testing web applications).
Kenneth Reitz on design.
Raymond Hettinger (you'll find "Writing Beautiful Idiomatic Python" and "Beyond PEP8" most helpful).
"""
"""
Not a channel, but talks by David Beazley on Python topics are gems. High insight density. The presentation are often 60+ pages, and you can spend hours playing with each concept in every 'slide'. His talks are rabbit holes. Raymond Hettinger tweets snippets of code that are delicious. His talks are really useful, too. Specifically, the one entitled 'Beyond PEP8' which focuses you on the impact you can have by zooming out of details and concentrate on good design.
This is something I've been using. I write code that uses a yet to be API, and ask myself and colleagues if it makes sense for them to use it.
For example, I'm writing a Python wrapper for MinIO's admin command line interfaces 'mc' and 'minio' which the Python client lacks. Put up the docs first at https://big-mama-tech.gitlab.io/bmc and asked colleagues and the MinIO folks if it's acceptable/useful before really going for parity.
"""
- [0]: https://news.ycombinator.com/item?id=15576496
- [1]: https://news.ycombinator.com/item?id=24671742
I replied to similar threads[0][1], reproduced here:
"""
The Python documentation is pretty cool. I frequently go to Chapter 3 (Data Model) and always learn something new reading something I had already read but didn't quite get. The mailing list has top notch posters who are always willing to help.
The IRC channel (#python) has very helpful members who've been patient with me on more than one occasion.
Books:
Stacking the deck:
"Python Guide" - Kenneth Reitz: The bits one doesn't necessarily think about but that simplify life: virtual environments, unit test, project structure, good practices, scenarios for different kinds of applications, documentation, etc.
Doing stuff:
"Python Cookbook" (several editions by different authors, Alex Martelli and David Beazley): as the title states, it's a cookbook that shows you how to do a bunch of stuff in Python (files, network, threads, etc). This can get you "operational" in Python pretty fast.
Doing stuff well (meta):
"Writing Idiomatic Python" - Jeff Knupp:
The book comes for Python 2 and Python 3 and contrasts "harmful" and "good". Jeff Knupp also has videos showing you the process of refactoring code. He also has a blog (https://jeffknupp.com/) where he addresses many topics.
"Effective Python: 59 Specific Ways to Write Better Python" - Brett Slatkin: Pretty good book. The subtitle says it all: 59 specific ways to write better Python. (600 + pages).
"Learn Python the Hard Way" - Zed Shaw: I've read that starting out in Python. Maybe not for a 15 year Perl veteran, but this is for memo.
There's certainly some overlap in these books, but it helps to repeat a message to understand the importance of something, or to drive a point home. I didn't use virtual environments until the message was repeated enough in different ways.
Pornography:
PyCon talks. A lot of these are great.
David Beazley (a bunch of talks that are pretty cool. Still haven't finished the ones on generators but they're very good).
Brandon Rhodes (on aesthetics and design patterns (or lack thereof), and testing web applications).
Kenneth Reitz on design.
Raymond Hettinger (you'll find "Writing Beautiful Idiomatic Python" and "Beyond PEP8" most helpful).
"""
"""
Not a channel, but talks by David Beazley on Python topics are gems. High insight density. The presentation are often 60+ pages, and you can spend hours playing with each concept in every 'slide'. His talks are rabbit holes. Raymond Hettinger tweets snippets of code that are delicious. His talks are really useful, too. Specifically, the one entitled 'Beyond PEP8' which focuses you on the impact you can have by zooming out of details and concentrate on good design.
This is something I've been using. I write code that uses a yet to be API, and ask myself and colleagues if it makes sense for them to use it.
For example, I'm writing a Python wrapper for MinIO's admin command line interfaces 'mc' and 'minio' which the Python client lacks. Put up the docs first at https://big-mama-tech.gitlab.io/bmc and asked colleagues and the MinIO folks if it's acceptable/useful before really going for parity.
"""
- [0]: https://news.ycombinator.com/item?id=15576496
- [1]: https://news.ycombinator.com/item?id=24671742
This.
The best references I know of are listed in that comment.
I just wanted to emphasize David Beazley's videos. The video [0] about python concurrency I consider a masterpiece.
[0] https://www.youtube.com/watch?v=MCs5OvhV9S4
The best references I know of are listed in that comment.
I just wanted to emphasize David Beazley's videos. The video [0] about python concurrency I consider a masterpiece.
[0] https://www.youtube.com/watch?v=MCs5OvhV9S4
I’m not sure if there is any definite resource that covers it all.
Besides the official documentation perhaps have a look at these resources:
https://openbookproject.net/thinkcs/python/english3e/
https://realpython.com/
Besides the official documentation perhaps have a look at these resources:
https://openbookproject.net/thinkcs/python/english3e/
https://realpython.com/
Fluent Python is a great book. 2nd edition is accessible as well https://learning.oreilly.com/library/view/fluent-python-2nd/...
I've last done python seriously in 2005, so early 2.x versions.