For an update 2.1, I would very much like the ability to assign (unused) keyboard shortcuts (e.g. command+Q on an external keyboard) to script shortcuts (under the wrench, as one of the options for edit shortcut, similar to icon).
That way, you could make the editor fit your own taste :-)

You can use \1 to refer to matched text grouped by () in the regex. So:

import re
my_text = 'Word word'
src_str  = re.compile('(word)', re.IGNORECASE) 
my_text = src_str.sub(r'\1xxx', my_text)
print(my_text)

dict is a built-in type, not a module that needs importing
What are you trying to accomplish, @Sway?

@jaredbidlow, Pythonista 2.0 provides appex, so you don't need Workflow and the clipboard anymore to Open in Pythonista. See e.g. https://forum.omz-software.com/topic/2494/appex-getfromurl-to-share-from-other-apps-to-pythonista for an alternative

Today 18 Jan 2023, a beta of a new Pythonista was made available to testers. This contains a more recent numpy ☺ and pandas ☺☺☺.

When the beta is promoted to the App Store, all users will benefit from this (and other evolutions like Python 3.10 instead of 3.6).

I'd say, list of dict with ('class', 'View') as one of its items and 'frame' as another key would be pretty telling. But it wouldn't provide 100% absolute dead-sure unmistakable certainty, I'm afraid, so don't use this if trying to control a nuclear reactor or pacemaker

Both in Pythonista2 and Pythonista3 (latest betas) a long press to run (i.e. with args in 2; either with args or doctest in 3) does not clear the console if clear output before running is set in settings. I don't know whether this is by design, but (especially with verbose option; great option BTW) the doctest output tends to get longish, so this behaviour may lead to confusing output

@dgelessus , I agree objc_util needs an __all__
However, to exclude the imported modules, you don't necessarily need to enumerate; instead you can use inspect:

{name for name in dir(objc_util)
      if not name.startswith("_") and
         not inspect.ismodule(getattr(objc_util, name))
}

Interesting!
Off topic:
As Nichtmuttersprachler who learned German in school way before the 1998 Rechtschreibereform, I was totally unaware of this German discussion. Do people actually propose writing "zwai"?

iPad Air 2: (1024.00, 768.00)

@ihf try automatic Python 2 to 3 conversion

Of course, that will not use all newer Python-3 niceties but it should allow most Python-2 scripts to run under Python 3

Today 18 Jan 2023, a beta of a new Pythonista was made available to testers. This contains a more recent numpy ☺ and pandas ☺☺☺.

When the beta is promoted to the App Store, all users will benefit from this (and other evolutions like Python 3.10 instead of 3.6).

@ihf I cannot reproduce, import pandas works for me.

But the pandas.__file__ I then get: …/pylib/site-packages/pandas/__init__.py seems to differ from the error messages you get …/Documents/site-packages-3/pandas/__init__.py.

Could it be you’re not using the built-in pandas?

Great news! Thanks @omz!

And seems already submitted 7 Dec, hence 48 days remaining of 3-months TestFlight period. Whatever, it’s downloading and I look forward to more modern Python features

You can use \1 to refer to matched text grouped by () in the regex. So:

import re
my_text = 'Word word'
src_str  = re.compile('(word)', re.IGNORECASE) 
my_text = src_str.sub(r'\1xxx', my_text)
print(my_text)

dict is a built-in type, not a module that needs importing
What are you trying to accomplish, @Sway?

@edenb thanks!

Happy to see you @omz and the update!

@JITASIDA
No, you can't
That would require adding a method to a built-in type and that's not possible (without going to the C source for Python):

>>> str.clean=lambda self:'clean'+self
Traceback (most recent call last):
  File "<string>", line 1, in <module>
TypeError: can't set attributes of built-in/extension type 'str'

The error results from your lines not being well-aligned
Unindenting the x=... and y=... lines should fix it