Thanks, Jason! I appreciate any solution that does not break old code. 
Deprecation warnings are useful to tell you why your code is going to be 
broken (if informative enough) but they don't take away the necessity to 
change all your code, unless you have have been setting the new standard 
option anyway.

By the way, I don't see the problem with the current solve output of e.g.
[x == 2*sqrt(y)], as this can be used as a replacement rule either by 
.subs() or subs_expr(). Actually, I deleted the solution_dict = True 
option in your example code and got the same result. Were you not aware 
that the arguments accepted by .subs() have changed?

Cheers
Stan

Jason Grout wrote:
> Stan Schymanski wrote:
>   
>> To reduce the problem about sharing code, all the custom settings could 
>> be saved in a dictionary and a simple call to your own dictionary would 
>> restore your own custom settings again.
>>
>> I see your point about reduced compatibility between different people's 
>> code, though. I'd still much more prefer this to a change in the default 
>> behaviour that would break my own code without providing a way of 
>> reverting this change to run my old code.
>>     
>
>
> Mike Hansen's @options decorator that is used in the graphics provides a 
> way to set the default options.  We use it in the graphics code all the 
> time.  However, I agree with Robert here in this case (where the output 
> would be different for different defaults).
>
> One could make the case that solve with solution_dict should actually be 
> a different function.
>
> I think making subs() and the __call__ method accept solve output would 
> take care of most of the needs for me, and would not be disruptive at all.
>
> Jason
>   

--~--~---------~--~----~------------~-------~--~----~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---

Reply via email to