[sage-devel] Re: TypeError: unsupported operand parent(s) for *: <(1,3)(2,4), (1,4,3,2)> and S4

2018-01-30 Thread Simon King 
PS:

On 2018-01-30, Simon King  wrote:
> The coercion model chooses coercion to the left factor's parent, if
> there are coercions available in *both* directions. However, if there
> only is a coercion from the left factor's parent to the right factor's
> parent but not vice versa, that coercion is supposed to be used.

Jeroen just pointed out on trac, that indeed Sage believes that there is
not only a coercion from the subgroup to the group, but also from the
group to the subgroup, which is totally wrong (and the "coercion map"
from group to subgroup indeed raises an error when being applied to an
element of the group).

That would explain it.

-- 
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: TypeError: unsupported operand parent(s) for *: <(1,3)(2,4), (1,4,3,2)> and S4

2018-01-30 Thread Simon King 
Dear Hsin-Po Wang,

On 2018-01-30, Hsin-Po Wang  wrote:
> But it is still interesting to see if this is an unintentional bug in, say, 
> the coercion model and can be fixed.

We have
  sage: S4.has_coerce_map_from(S4.subgroups()[19])
  True

Hence, the coercion model is supposed to *automatically* convert the
element S4.subgroups()[19][1] into an element of S4, and then do the
multiplication.

Oddly enough,
  sage: S4[1]*S4.subgroups()[19][1]
  (1,4,2,3)
works (i.e., the right factor is automatically converted into the parent
of the left factor), but
  sage: S4.subgroups()[19][1]*S4[1]
doesn't work (as you reported).

The coercion model chooses coercion to the left factor's parent, if
there are coercions available in *both* directions. However, if there
only is a coercion from the left factor's parent to the right factor's
parent but not vice versa, that coercion is supposed to be used.

So, shortly, it is a clear bug. I created
https://trac.sagemath.org/ticket/24612 for it.

Thank you for reporting it!
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: TypeError: unsupported operand parent(s) for *: <(1,3)(2,4), (1,4,3,2)> and S4

2018-01-29 Thread Hsin-Po Wang 


> use
> S4(S4.subgroups()[19][1]) * S4[1]
>

Thank you. That makes perfect sense.

This does not bother me anymore since I found that pre-casting all 
subgroup-elements to S4 accelerates my work a lot.
(i.e. S4subgroups=[ [S4(h) for h in H] for H in S4.subgroups()] )

But it is still interesting to see if this is an unintentional bug in, say, 
the coercion model and can be fixed.

-- 
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: TypeError: unsupported operand parent(s) for *: <(1,3)(2,4), (1,4,3,2)> and S4

2018-01-29 Thread John H Palmieri 
I'm not sure if Sage can always tell, if you multiply an element of a 
subgroup by an element of the larger group, that the result should make 
sense and lie in the larger group. You should be able to fix this by 
forcing Sage to consider the subgroup element as lying in the larger group: 
rather than

S4.subgroups()[19][1] * S4[1]

use

S4(S4.subgroups()[19][1]) * S4[1]


On Monday, January 29, 2018 at 3:20:16 PM UTC-8, Hsin-Po Wang wrote:
>
> The following code
>
> sage: S4=SymmetricGroup(*4*)
>
> sage: S4.subgroups()[*19*][*1*]*S4[*1*]
>
> raises an error
>
> TypeError: unsupported operand parent(s) for *: 'Subgroup of (Symmetric 
> group of order 4! as a permutation group) generated by [(1,3)(2,4), 
> (1,4,3,2)]' and 'Symmetric group of order 4! as a permutation group'
>
> I am not sure what exactly happened but this is quite peculiar. Among al 
> 30 subgroups of S4 only the 19th and 27th raise errors.
>
> sage: *for* Sid,S in enumerate(S4.subgroups()):
>
> : *for* s in S:
>
> : *for* g in S4:
>
> : *try*:
>
> : sg=s*g
>
> : *except* *TypeError*:
>
> : #print'For {}-th subgroup: {}*{} is 
> bad'.format(Sid,s,g)
>
> : *print* Sid,
>
> : 
>
>
> 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 
> 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 
> 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 
> 19 19 19 19 19 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 
> 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 
> 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 
> 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 
> 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 
> 27 27 27 27 27 27 27 27
>
> I am using Sage 8.0.0 on macOS 10.13.2
>
>

-- 
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.