Re: [isabelle-dev] Default simprules for division in fields

2014-04-06 Thread Thomas Sewell
This makes me realise that I may have been misunderstanding field_simps. 
I've been using it as a replacement after ring_simps disappeared a while 
ago. It always struck me as odd, since I'm nearly always in a word type, 
which is a ring but not a field.


Is it just a coincidence that field_simps usually does what I want?

Cheers,
Thomas.

On 05/04/14 03:09, Tobias Nipkow wrote:

I set up field_simps to yield a decision procedure for field equality, provided
all denominators can be proved to be 0. Hence I am sceptical if adding new rules
to it is a good idea. Florian, can you give an example where previously
field_simps was too weak but with the two additional rules it works?

Tobias

On 04/04/2014 17:37, Lawrence Paulson wrote:

A very good idea, which reduces the impact of the change on existing proofs. I 
am trying it out now.

Seeing no objections, I am quite likely to push this change later today.

Larry

On 4 Apr 2014, at 15:08, Florian Haftmann 
florian.haftm...@informatik.tu-muenchen.de wrote:


lemma divide_minus_left [simp]: (-a) / b = - (a / b)
lemma divide_minus_right [simp]: a / - b = - (a / b)”

It would be a mistake to reorient the simprules, but I am suggesting that they 
should not be declared as simprules with either orientation.

Maybe

lemma divide_minus_left [field_simps]: (-a) / b = - (a / b)
lemma divide_minus_right [field_simps]: a / - b = - (a / b)”

instead?

Florian

--

PGP available:
http://home.informatik.tu-muenchen.de/haftmann/pgp/florian_haftmann_at_informatik_tu_muenchen_de

___
isabelle-dev mailing list
isabelle-...@in.tum.de
https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev

___
isabelle-dev mailing list
isabelle-...@in.tum.de
https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev


___
isabelle-dev mailing list
isabelle-...@in.tum.de
https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev


___
isabelle-dev mailing list
isabelle-...@in.tum.de
https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev


Re: [isabelle-dev] Default simprules for division in fields

2014-04-06 Thread Tobias Nipkow
You can use algebra_simps for everything up to and including rings. Field_simps
is separate because it can rearrange your formula dramatically by multiplying
with all the denominators - you don't always want that.

Tobias

On 07/04/2014 03:11, Thomas Sewell wrote:
 This makes me realise that I may have been misunderstanding field_simps. I've
 been using it as a replacement after ring_simps disappeared a while ago. It
 always struck me as odd, since I'm nearly always in a word type, which is a 
 ring
 but not a field.
 
 Is it just a coincidence that field_simps usually does what I want?
 
 Cheers,
 Thomas.
 
 On 05/04/14 03:09, Tobias Nipkow wrote:
 I set up field_simps to yield a decision procedure for field equality, 
 provided
 all denominators can be proved to be 0. Hence I am sceptical if adding new 
 rules
 to it is a good idea. Florian, can you give an example where previously
 field_simps was too weak but with the two additional rules it works?

 Tobias

 On 04/04/2014 17:37, Lawrence Paulson wrote:
 A very good idea, which reduces the impact of the change on existing proofs.
 I am trying it out now.

 Seeing no objections, I am quite likely to push this change later today.

 Larry

 On 4 Apr 2014, at 15:08, Florian Haftmann
 florian.haftm...@informatik.tu-muenchen.de wrote:

 lemma divide_minus_left [simp]: (-a) / b = - (a / b)
 lemma divide_minus_right [simp]: a / - b = - (a / b)”

 It would be a mistake to reorient the simprules, but I am suggesting that
 they should not be declared as simprules with either orientation.
 Maybe
 lemma divide_minus_left [field_simps]: (-a) / b = - (a / b)
 lemma divide_minus_right [field_simps]: a / - b = - (a / b)”
 instead?

 Florian

 -- 

 PGP available:
 http://home.informatik.tu-muenchen.de/haftmann/pgp/florian_haftmann_at_informatik_tu_muenchen_de


 ___
 isabelle-dev mailing list
 isabelle-...@in.tum.de
 https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev
 ___
 isabelle-dev mailing list
 isabelle-...@in.tum.de
 https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev

 ___
 isabelle-dev mailing list
 isabelle-...@in.tum.de
 https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev
 
 ___
 isabelle-dev mailing list
 isabelle-...@in.tum.de
 https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev
___
isabelle-dev mailing list
isabelle-...@in.tum.de
https://mailmanbroy.informatik.tu-muenchen.de/mailman/listinfo/isabelle-dev