To claim/decide whether something is a bug and/or documentation failure, it 
> would be nice to have some documented behavior... A loooong time ago when I 
> was writing interacts myself, it was indeed quite frustrating to see 
> discrepancies between SageNB and SageMathCell (and then CoCalc too). 
> Documentation for all features also was not available.
>


Yes, you are sure right about that. 
 See https://groups.google.com/d/topic/sage-devel/y5v-BGJ5MjY/discussion - 
from SIX years ago.

In this case, it's documented in the sense that those interacts must have 
worked at some time - in fact, they must have worked with py2 Sage cell 
even!  I think it really is something about iterators/generators that 
changed.
 

>
> With the current situation, I think it is desirable
> 1) not to break things that are working now
>

Absolutely.
 

> 2) have consistency with Jupyter/CoCalc
>

Unfortunately those two are not the same, though hopefully not terribly far 
apart.

3) follow some documentation (if things are consistent, no need for 
> separate documentation from other systems)
>
>
I updated the https://wiki.sagemath.org/interact page with my best search 
for places where this is "documented".   The original sagenb stuff probably 
was supposed to be the default implementation at one time, but of course it 
doesn't have nested interacts and some other stuff, which I believe may not 
be properly documented anywhere - again, only in examples that work/worked.
 

> Help in incremental achieving of these goals is surely appreciated - in 
> the form of examples that behave differently as well as pull requests!
>

I will do my best!  That is what this thread was intended to (re)start. 

-- 
You received this message because you are subscribed to the Google Groups 
"sage-cell" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-cell/64b93389-b4fd-4204-8b09-8893d3c45e82%40googlegroups.com.

Reply via email to