Move matrices to new coercion model

[videlec]

MatrixSpace do not play well with the category framework. One step forward is to rename the attribute _matrix_class to Element so that it will have element_class pointing to the element class generated via Parent.__init__.

Related tickets: #19669

Depends on #25504
Depends on #25505
Depends on #25542
Depends on #25554
Depends on #25555

CC: @tscrim

Component: coercion

Keywords: days94

Author: Jeroen Demeyer

Branch/Commit: b03b8bf

Reviewer: Travis Scrimshaw

Issue created by migration from https://trac.sagemath.org/ticket/23719

[jdemeyer]

Changed dependencies from #23706 to #24742

[jdemeyer]

comment:4

If you want progress on this, it would be good to review #24742.

[jdemeyer]

Changed dependencies from #24742 to none

[jdemeyer]

Author: Jeroen Demeyer

[jdemeyer]

comment:7

Working on this...

[jdemeyer]

comment:8

Vincent, since you created this ticket: are there any concrete issues that would be fixed by this?

[videlec]

comment:9

My aim was to allow matrices on semi-rings (ie structure having + and * but not necessarily - and /). We are far away from there though. So only a wishlist.

[jdemeyer]

comment:10

Replying to @videlec:

My aim was to allow matrices on semi-rings

What do you mean with "allow"? And how would that be related to changing element_class?

[jdemeyer]

Branch: u/jdemeyer/move_matrices_to_new_coercion_model

[sagetrac-git]

Commit: e48d153

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:

ac7cf06	Implement `_an_element_` for matrix spaces
e48d153	Move matrices to new coercion model

[jdemeyer]

Dependencies: #25504

[jdemeyer]

Changed dependencies from #25504 to #25504, #25505

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. This was a forced push. Last 10 new commits:

3c4a06d	Enable _mul_long for matrices
1eaed37	More stuff in the meataxe interface, and a meataxe helper function
8a06c0f	Fix docstring formatting
13da208	Clean up creating Matrix_gfpn_dense matrices
cb52ee3	Mark one doctest optional
c8fa7bb	Clean up `__cinit__` methods of matrices
779dffb	Implement `_an_element_` for matrix spaces
6b8517f	Merge commit '779dffbbb514290400d2b164b80ddb1a6c8ba04e' into t/23719/move_matrices_to_new_coercion_model
55568c8	Misc matrix fixes
c646dc2	Move matrices to new coercion model

[sagetrac-git]

Changed commit from e48d153 to c646dc2

[jdemeyer]

Changed dependencies from #25504, #25505 to #25504, #25505, #25542

[sagetrac-git]

Changed commit from 4b83f2f to ecfb511

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:

ae560ce	Add option to control GC during doctests
ecfb511	Move matrices to new coercion model

[jdemeyer]

comment:42

The error seems unrelated to this ticket. Somebody also got it because of the Singular upgrade, see #25969.

[jdemeyer]

Changed dependencies from #25504, #25505, #25542, #25554, #25555, #25871 to #25504, #25505, #25542, #25554, #25555, #25969

[videlec]

comment:44

update milestone 8.3 -> 8.4

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1 and set ticket back to needs_review. This was a forced push. New commits:

b03b8bf

Move matrices to new coercion model

[sagetrac-git]

Changed commit from ecfb511 to b03b8bf

[tscrim]

comment:46

Is #25969 still a dependency? It seems from the comments like #25969 is not coming from this ticket.

[jdemeyer]

comment:47

Replying to @tscrim:

Is #25969 still a dependency?

I don't know. We can test again whether the problem magically went away.

[jdemeyer]

comment:48

Replying to @tscrim:

It seems from the comments like #25969 is not coming from this ticket.

Well, define "coming from". The problem is certainly unrelated and not directly caused by this ticket, but merging this ticket did cause #25969 to appear.

[jdemeyer]

comment:49

I just tested again and this no longer causes #25969.

[jdemeyer]

Changed dependencies from #25504, #25505, #25542, #25554, #25555, #25969 to #25504, #25505, #25542, #25554, #25555

[vbraun]

Changed branch from u/jdemeyer/move_matrices_to_new_coercion_model to b03b8bf