Re: Is it time to require Python 3?

On Mon, Feb 18, 2019 at 4:40 AM Edward K. Ream wrote: > My plan is to make this transition in Leo as soon as Leo 5.8.1 goes out the door. This should happen in a few weeks. Today's work in the py3 branch removes all Python 2 dependencies, that is, all significant tests for g.isPython3. So in

Re: Is it time to require Python 3?

> > Matt, what effect will requiring Python 3 have on your work? > All good effects. I've been testing with both so I expect it to be seamless. I'll be able to remove some testing and if/then parsing. All in all it will be simpler. If it won't cause trouble with downstream distributions I'd

Reasons, if any, to use @asis over @clean?

Hi All, So I recently ran into a situation where checking out different branches and commits of @asis files would regularly have me unwittingly overwriting the external files with old data from within the .leo file, details in https://github.com/leo-editor/leo-editor/issues/1081. Thanks to

Re: I'm back from vacation

On Mon, Feb 18, 2019 at 2:56 AM vitalije wrote: > Perhaps you might want to tackle again outline drawing code. Imo, #1068 is likely the best way to do this. I would welcome your help with this. I promise not to touch anything you do until

Re: Is it time to require Python 3?

On Sunday, February 17, 2019 at 9:28:04 PM UTC-6, Matt Wilkie wrote: Given that Python.org has set the end of life for python v2 to be only 8 > months from now, 2020, I think it's fine to talk about not supporting it > longer in Leo. I suggest doing that in v6 rather than 5.9 though as it >

Re: Is it time to require Python 3?

On Sunday, February 17, 2019 at 6:21:03 PM UTC-6, Zoom.Quiet wrote: > > IMHO: > > if Python 2.7 can support all Leo running env. > not big reason had to jump into Py3; > dont panic, slow slowly jump it. > Leo has been "slowly jumping into Py3" ever since Python 3.0. I agree with Vitalije.

Re: Is it time to require Python 3?

This is an excellent idea. Python3 is really much faster than Python2 when its new features are used. Right now, many of its features can't be used because of Python2 support. Leo still performs lots of unicode/str checking/converting just because it needs to be compatible with both versions.

Re: I'm back from vacation

Something went wrong with my previous message. Google has added some attachment to my message and all the code examples were garbled. So, I had to delete that message and post it again. Vitalije -- You received this message because you are subscribed to the Google Groups "leo-editor" group.

Re: Is it time to require Python 3?

On Mon, Feb 18, 2019 at 2:34 AM vitalije wrote: This is an excellent idea. Python3 is really much faster than Python2 when > its new features are used. Right now, many of its features can't be used > because of Python2 support. Leo still performs lots of unicode/str > checking/converting just

Re: I'm back from vacation

So I am looking for what's next in my life. > > Edward > Perhaps you might want to tackle again outline drawing code. Here is a simple script that shows how much faster can be tree traversal. import timeit def f(): def it(root, lev): yield root, lev for v in

Re: Is it time to require Python 3?

On Monday, February 18, 2019 at 4:22:18 AM UTC-6, Edward K. Ream wrote: Leo has been "slowly jumping into Py3" ever since Python 3.0. I agree with > Vitalije. There are many reasons to prefer Py3. > The Python 2.7 Countdown page indicates that Python 2 will retire

Re: I'm back from vacation

> > > Imo, #1068 is > likely the best way to do this. > I would not be so sure about #1068 being the best way. But it certainly deserves further exploration. The idea of finding the difference and applying only small set of operations to