I just figured out that PyCharm supports a docstring with code prefixed by >>> like this:
def foo(x):
"""
Use it like:
>>> foo(42)
"""
print(x)
PyCharm then even applies the usual code highlighting on this.
I tried to find any documentation on this syntax. It doesn't seem to be part of reStructuredText. Also Sphinx does not mention this. Was this introduced in some PEP? Or is this PyCharm specific? But I did not find any PyCharm documentation on this.
(It's hard to use Google search for this. I don't know how to call this. "3-times greater sign"?)
The comments have answered the question. I'm just putting it here for completeness:
It's three fleshes (>>>).
It's used for doc test.
In PyCharm, when you rightclick it, it allows you to Run Doctest.
The syntax probably comes from the Python interactive shell.
Related
I'm currently developing a dynamic software system in python. It includes the use of dependency injection at many points, which is implemented with dynamic class attributes. My aim is to add a function dynamically to a class (with setattr) an this works quite fine. The only problem I have is that the IDE (in my case it's PyCharm) has no idea about those functions and marks it as "unresolved reference". Although the script runs without errors, it looks not very nice and I want the IDE to support other programmers, that don't know about those functions.
Here is an example of what I mean:
class A:
def __init__(self, func_obj):
setattr(self, 'custom_function', func_obj)
a = A(print) # Unresolved attribute reference 'custom_function' for class 'A' ...
a.custom_function('Hello World!')
As expected, this example prints "Hello World!". But I find it's ugly, that the IDE shows a warning for the last line, where I called "custom_function". And also there is no auto completion if you try to explore the script with duck typing.
So let me come to my question: Is there any way to tell the IDE that there are some dynamically added functions in this class? Maybe there is some way with meta classes or something like that... But I didn't find anything on Google and I have no idea where else to try.
I hope you can help me with that :)
I had a similar request and asked directly to PyCharm bug tracker: PY-28326. I think that their answer applies to your question too:
PyCharm at the moment doesn't support methods dynamically added to classes.
A library I'd like to use seems to get confused by triple quote comments when introspecting python code. It appears that Python's inspect doesn't give access to de-commented code though?
This code in Hyperas breaks and looks un-pythonic but I can't figure out how it could be done better. Is AST the way forward?
I've just started looking at the Vim jedi plugin, and it seems pretty impressive. One feature of some of the Java IDEs I've used is the ability to automatically add required imports. Can Jedi do that? For example, if I enter a line such as
arg1 = sys.argv[1]
and then invoke some Jedi command, is it possible for the plugin to automatically insert an import sys line at the top of the source file (if sys is not already being imported)?
I've looked through the Jedi help, and can't see anything like this - but it's possible I missed something. Alternatively, is there another Vim plugin that would do this? (It needs a certain level of understanding of Python syntax to get it right, which is why I looked to Jedi to be able to do it).
Currently Jedi doesn't do refactoringing. This includes import additions. There's an issue for the whole subject: https://github.com/davidhalter/jedi/issues/667.
It's not that easy to implement this command with good performance. However any help is appreciated. :)
FIY, I've defined a generic import feature that can be used on demand in lh-dev. I use it from my C&C++ suite, and from my snippet engine (mu-template).
So far I don't parse anything to add the missing import/include statements. This part would be complex as Dave said. Instead my snippets know which files need to be imported/included and import/include them if not already imported/included.
It's far from being perfect, but it's a start. mu-template provides a hook to do stuff at the start of the file after the snippet has been expanded, this is where I call lh-dev function. If other snippet engines provide similar hooks, you should be able to call lh#dev#import#add() from your snippets.
Here a proof of concept snippet for Python (I seldom program in Python, and don't have many snippets for it): https://github.com/LucHermitte/mu-template/blob/master/after/template/python/path-exists.template
I'm looking for an extension which will help in Python's auto complete feature in Python.
If I type the following code:
a = [4,5,6]
a.p
Then I expect it would give me a suggestion for pop as it is one of the method of Python list. Is this thing achievable in Emacs?
I tried installing Rope, Rope mode, Ropemacs and Pymacs. The auto-complete suggestion, I get from that aren't the methods of list. But it suggests me something like print etc. Am I doing something wrong ?
Try Jedi.el. AFAIK, it should support your example.
Because Python variables don't have types, it is difficult to know what class of object a variable would contain at a given point in your code. Imagine if your code later did a = 1. Emacs would have to know when the variable a refers to a list and when it refers to a number in order to offer the correct completion. Generally this is not possible without actually running the code.
Works for me with vimrope from https://github.com/python-rope/ropevim/ (now, the official home of all rope projects)
I'm noticing that even for system modules, code completion doesn't work too well.
For example, if I have a simple file that does:
import re
p = re.compile(pattern)
m = p.search(line)
If I type p., I don't get completion for methods I'd expect to see (I don't see search() for example, but I do see others, such as func_closure(), func_code()).
If I type m., I don't get any completion what so ever (I'd expect .groups(), in this case).
This doesn't seem to affect all modules.. Has any one seen this behaviour and knows how to correct it?
I'm running Vim 7.2 on WinXP, with the latest pythoncomplete.vim from vim.org (0.9), running python 2.6.2.
Completion for this kind of things is tricky, because it would need to execute the actual code to work.
For example p.search() could return None or a MatchObject, depending on the data that is passed to it.
This is why omni-completion does not work here, and probably never will. It works for things that can be statically determined, for example a module's contents.
I never got the builtin omnicomplete to work for any languages. I had the most success with pysmell (which seems to have been updated slightly more recently on github than in the official repo). I still didn't find it to be reliable enough to use consistently but I can't remember exactly why.
I've resorted to building an extensive set of snipMate snippets for my primary libraries and using the default tab completion to supplement.