On Thursday, May 25, 2017 at 6:11:02 PM UTC-7, Nils Bruin wrote:
>
> On Thursday, May 25, 2017 at 5:07:22 PM UTC-7, John H Palmieri wrote:
>>
>>
>>
>> On Thursday, May 25, 2017 at 4:40:01 PM UTC-7, Nils Bruin wrote:
>>>
>>> incidentally, it would be nice if we could equiv local/bin/ipython with 
>>> the proper sage-python23 as well. Having a functional shell with an 
>>> interactive debugger is a much nicer way to debug "import sage.all".
>>>
>>
>> This is a design question for which we do not have an answer, or even any 
>> real suggestions. Right now, SAGE_PYTHON3=yes is purely for building Sage, 
>> not for running it, and maybe long-term it should get replaced with a 
>> configure option anyway. Python conventions say that "python" should always 
>> mean "Python 2", so if we want to follow those  conventions, then 
>> local/bin/python should not be sage-python23.
>>
>
> I don't really care at this point. 
>

Then I would say, if SAGE_PYTHON3=yes, make a symlink python -> python3 (in 
the ugly_py3 style).
 

> In the ugly_py3 style, it's just a good idea to equip local/bin/ipython 
> with something that allows it to run python3 (so that sage -ipython works 
> on a py3 build). Otherwise, there should be a local/bin/ipython3, and 
> possibly sage -ipython can then select which exec to use. This is not about 
> final design, it's about having the right tools to make an inventory of how 
> badly broken sage/py3 is (and what fixes we might make).
>

If you build Sage's IPython package with SAGE_PYTHON3=yes, then it will 
install a script local/bin/ipython3. (Note then that if the sage-location 
script runs, its function make_scripts_relative will break 
local/bin/ipython3 because it will replace the #! line, which initially 
uses Sage's Python3, to a generic /usr/bin/env python. Another broken part 
to fix.)

  John

 

>
> I changed local/bin/ipython on my install, but it didn't show up in my git 
> diff. Perhaps local/bin/ipython isn't tracked? does one make the change 
> somewhere else? I was just looking for a way to include the change in 
> Frederic's branch. With that branch, we're finally getting to a state where 
> people can easily see what's wrong with Sage/Py3.
>
>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Reply via email to