[sage-devel] Re: Pointers needed

[Simon King]

Dear Nils, dear John,

On 2018-04-02, John H Palmieri  wrote:
> On Sunday, April 1, 2018 at 4:08:14 PM UTC-7, Nils Bruin wrote:
>> On Sunday, April 1, 2018 at 3:43:30 PM UTC-7, Simon King wrote:
>>>
>>> Also: Have there been changes in how to run optional tests? I tried sage 
>>> -t --optional=meataxe, but then Sage was skipping all compulsory tests 
>>> (i.e., those that haven't been marked "optional"), so that underlying 
>>> variable definitions were missing in the actual optional tests. 
>>>
>>> I think this has been for a long time: if you test without specifying 
>> --optional, the tester prints:
>>
>> Using --optional=mpir,python2,sage
>>
>> so probably when you invoke with --optional=mpir,python2,sage,meataxe your 
>> get what you want.
>>
>
> Right, use the "sage" tag along with "meataxe". This is documented here: 
> http://doc.sagemath.org/html/en/developer/doctesting.html#run-optional-doctests

Thank you! So, I also have to change this in the test suit of my spkg.

Best regards,
Simon

-- 
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.

[sage-devel] Re: Pointers needed

[John H Palmieri]

On Sunday, April 1, 2018 at 4:08:14 PM UTC-7, Nils Bruin wrote:
>
>
> On Sunday, April 1, 2018 at 3:43:30 PM UTC-7, Simon King wrote:
>>
>> Also: Have there been changes in how to run optional tests? I tried sage 
>> -t --optional=meataxe, but then Sage was skipping all compulsory tests 
>> (i.e., those that haven't been marked "optional"), so that underlying 
>> variable definitions were missing in the actual optional tests. 
>>
>> I think this has been for a long time: if you test without specifying 
> --optional, the tester prints:
>
> Using --optional=mpir,python2,sage
>
> so probably when you invoke with --optional=mpir,python2,sage,meataxe your 
> get what you want.
>

Right, use the "sage" tag along with "meataxe". This is documented here: 
http://doc.sagemath.org/html/en/developer/doctesting.html#run-optional-doctests

-- 
John
 

-- 
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.

[sage-devel] Re: Pointers needed

[Nils Bruin]

On Sunday, April 1, 2018 at 3:43:30 PM UTC-7, Simon King wrote:
>
> Also: Have there been changes in how to run optional tests? I tried sage 
> -t --optional=meataxe, but then Sage was skipping all compulsory tests 
> (i.e., those that haven't been marked "optional"), so that underlying 
> variable definitions were missing in the actual optional tests. 
>
> I think this has been for a long time: if you test without specifying 
--optional, the tester prints:

Using --optional=mpir,python2,sage

so probably when you invoke with --optional=mpir,python2,sage,meataxe your 
get what you want.

-- 
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.

[sage-devel] Re: Pointers needed

[Simon King]

Hi!

On 2018-04-01, Simon King  wrote:
> This *used* to work in the past and it still works when using a backend
> different from MeatAxe. Since there have been some recent changes in
> the matrix framework in SageMath that I didn't closely follow: What
> could be the reason that the coercion model got broken only on
> sage.matrix.matrix_gfpn_dense?

At least I can partially answer that part of my questions (but I can't
answer the question about "sage -t --optional" skipping the compulsory
tests): There is the "magical Sage method" _mul_long, that is still used
in sage.structure.element and is also used in matrix_gfpn_dense, but
apparently it stopped to work (at least on matrices), and it isn't used
for any other matrix type.

So, to narrow down that part of my questions: Is it really the case that
support for _mul_long was dropped for matrices? Why? Where? What to use
instead? Why haven't optional tests be done?

_mul_long is a short-cut that is useful for speed, as it avoids
a conversion into the matrix' base ring or even into the matrix' parent
(when it is a square matrix).

Best regards,
Simon

-- 
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.