Functorial constructions in categories #8881

hivert · 2010-05-05T02:20:49Z

This patch refactors completely the framework for (covariant)
functorial constructions, improves the existing ones, and adds new ones.

Improved constructions:

Cartesian products: generalization to non modules_with_basis,
monoid operations, ...

New constructions:

Subquotients, Quotients, Subobjects, IsomorphicObjects
DualObjects, Algebras

Miscellaneous category fixes and improvements:

is_subcategory now handles properly join categories
(courtesy of Florent Hivert)
For a field K, Modules(K) returns VectorSpaces(K)
As a temporary workaround, Parent.an_element calls the
_an_element_ method provided by the categories, if available
Adds a method Category.or_subcategory
Cleans up _repr_ by extracting a _repr_object_names method
Cleans up _latex_ by extracting a _short_name method

Depend on #9104

CC: @sagetrac-sage-combinat

Component: categories

Keywords: Functorial constructions

Author: Nicolas M. Thiéry

Reviewer: Florent Hivert

Merged: sage-4.4.4.alpha0

Issue created by migration from https://trac.sagemath.org/ticket/8881

The text was updated successfully, but these errors were encountered:

hivert · 2010-06-01T17:04:49Z

comment:1

The patch is on review on sage-combinat-queue and should be ready soon. Please Nicolas, can you fold my last review patch if you are ok with it and upload the bunch here. I'll probably put a positive review after a last re-reading (and at least 5 folded review patches) :-)

Florent

hivert · 2010-06-01T17:04:49Z

Reviewer: Florent Hivert

hivert · 2010-06-01T17:04:49Z

Author: Nicolas M. Thiéry

nthiery · 2010-06-01T20:19:06Z

Changed keywords from Functorial construction to Functorial constructions

hivert · 2010-06-01T22:09:55Z

comment:3

Replying to @nthiery:

One last remark, otherwise I'm ready to give positive review: there is a slight naming inconsistency:

CartesianProductFunctor, CartesianProducts, CartesianProductsCategory, cartesian_product
TensorFunctor, TensorProducts, TensorProductsCategory, tensor

I'm just not 100% sure that we don't want TensorProductFunctor and tensor_product...

Any arguments ?

By the way I put Priority=critical because they are at least a dozen of patch on sage-combinat which depend on this one...

nthiery · 2010-06-02T08:45:32Z

Renames TensorFunctor to TensorProductFunctor

nthiery · 2010-06-02T08:46:07Z

Attachment: trac_8881-functorial_constructions-nt.patch.gz

Attachment: diff.gz

Diff to previous version of the patch

nthiery · 2010-06-02T08:52:07Z

comment:4

Replying to @hivert:

One last remark, otherwise I'm ready to give positive review: there is a slight naming inconsistency:

CartesianProductFunctor, CartesianProducts, CartesianProductsCategory, cartesian_product

TensorFunctor, TensorProducts, TensorProductsCategory, tensor

I'm just not 100% sure that we don't want TensorProductFunctor and tensor_product...

Any arguments ?

Yup, we had discussed this with David Roe during the review of the
category patches. And we had agreed that tensor([A,B,C]) was
shorter and more practical than tensor_product([A,B,C]), yet
clear and unambiguous. On the other hand cartesian([A,B,C])
wasn't that clear. So we decided to favor here practicality over
consistency.

On the other hand, I fixed TensorFunctor to TensorProductFunctor in
the updated patch.

hivert · 2010-06-02T09:09:37Z

comment:5

Excellent ! Positive review unless Massena find a new bug (which is VeryUnlikely(TM)).

nthiery · 2010-06-02T13:36:21Z

comment:6

Replying to @hivert:

Excellent ! Positive review unless Massena find a new bug (which is VeryUnlikely(TM)).

Thanks for all your work on this!

williamstein · 2010-06-03T04:26:12Z

comment:7

This causes doctest failures, including the following.


sage -t  sage/categories/sets_cat.py
**********************************************************************
File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/devel/sage-main/sage/categories/sets_cat.py", line 624:
    sage: A = Monoids().example().algebra(QQ); A
Exception raised:
    Traceback (most recent call last):
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/bin/ncadoctest.py", line 1231, in run_one_test
        self.run_one_example(test, example, filename, compileflags)
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/bin/sagedoctest.py", line 38, in run_one_example
        OrigDocTestRunner.run_one_example(self, test, example, filename, compileflags)
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/bin/ncadoctest.py", line 1172, in run_one_example
        compileflags, 1) in test.globs
      File "<doctest __main__.example_16[4]>", line 1, in <module>
        A = Monoids().example().algebra(QQ); A###line 624:
    sage: A = Monoids().example().algebra(QQ); A
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/lib/python/site-packages/sage/categories/sets_cat.py", line 637, in algebra
        return CombinatorialFreeModule(base_ring, self, category = category.Algebras(base_ring))
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/lib/python/site-packages/sage/misc/classcall_metaclass.py", line 256, in __call__
        return cls.__classcall_private__(cls, *args, **options)
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/lib/python/site-packages/sage/combinat/free_module.py", line 850, in __classcall_private__
        return super(CombinatorialFreeModule, cls).__classcall__(cls, *args, **keywords)
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/lib/python/site-packages/sage/misc/cachefunc.py", line 115, in __call__
        w = self.f(*args, **kwds)
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/lib/python/site-packages/sage/structure/unique_representation.py", line 447, in __classcall__
        instance = type.__call__(cls, *args, **options)
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/lib/python/site-packages/sage/combinat/free_module.py", line 892, in __init__
        element_constructor = self._element_constructor_)
      File "parent.pyx", line 457, in sage.structure.parent.Parent.__init__ (sage/structure/parent.c:3851)
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/lib/python/site-packages/sage/categories/magmas.py", line 115, in __init_extra__
        if (self.product != self.product_from_element_class_mul) and hasattr(self, "element_class") and hasattr(self.element_class, "_mul_parent"):
      File "element.pyx", line 860, in sage.structure.element.Element.__richcmp__ (sage/structure/element.c:7061)
      File "element.pyx", line 801, in sage.structure.element.Element._richcmp (sage/structure/element.c:6441)
      File "coerce.pyx", line 907, in sage.structure.coerce.CoercionModel_cache_maps.canonical_coercion (sage/structure/coerce.c:8537)
      File "sage_object.pyx", line 101, in sage.structure.sage_object.SageObject.__repr__ (sage/structure/sage_object.c:1370)
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/lib/python/site-packages/sage/categories/homset.py", line 302, in _repr_
        self._domain, self._codomain, self.__category)
      File "sage_object.pyx", line 101, in sage.structure.sage_object.SageObject.__repr__ (sage/structure/sage_object.c:1370)
      File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/local/lib/python/site-packages/sage/combinat/free_module.py", line 1163, in _repr_
        return self._name + " over %s"%self.base_ring()
      File "parent.pyx", line 676, in sage.structure.parent.Parent.__getattr__ (sage/structure/parent.c:5239)
      File "parent.pyx", line 263, in sage.structure.parent.getattr_from_other_class (sage/structure/parent.c:2753)
      File "parent.pyx", line 171, in sage.structure.parent.raise_attribute_error (sage/structure/parent.c:2625)
    AttributeError: 'CombinatorialFreeModule_with_category' object has no attribute '_name'
**********************************************************************
File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/devel/sage-main/sage/categories/sets_cat.py", line 626:
    sage: A.category()
Expected:
    Category of monoid algebras over Rational Field
Got:
    Category of set algebras over Rational Field
**********************************************************************
1 items had failures:
   2 of   8 in __main__.example_16
***Test Failed*** 2 failures.
For whitespace errors, see the file /scratch/wstein/sage//tmp/.doctest_sets_cat.py

hivert · 2010-06-03T06:09:33Z

comment:8

Replying to @williamstein:

This causes doctest failures, including the following.

> 
> sage -t  sage/categories/sets_cat.py
> **********************************************************************
> File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/devel/sage-main/sage/categories/sets_cat.py", line 624:
>     sage: A = Monoids().example().algebra(QQ); A
> Exception raised:
. [...]
>     AttributeError: 'CombinatorialFreeModule_with_category' object has no attribute '_name'
> **********************************************************************
> File "/mnt/usb1/scratch/wstein/build/release/4.4.3/sage-4.4.3.alpha2/devel/sage-main/sage/categories/sets_cat.py", line 626:
>     sage: A.category()
> Expected:
>     Category of monoid algebras over Rational Field
> Got:
>     Category of set algebras over Rational Field
> **********************************************************************

This bug is fixed in #9104 but it does appear on 4.4.2 so I didn't think about a dependency. Sorry for this.

hivert · 2010-06-03T06:11:25Z

comment:9

Note: I put back positive review though I had no chance to test on sage-4.4.3.alpha2...

hivert · 2010-06-03T06:14:53Z

comment:10

Replying to @hivert:

This bug is fixed in #9104 but it does appear on 4.4.2 so I didn't think about a dependency. Sorry for this.

Actually it did (see the note there), but I completely forgot about it. My mistake.

mwhansen · 2010-06-05T21:44:23Z

Merged: sage-4.4.4.alpha0

hivert added this to the sage-4.4.4 milestone May 5, 2010

hivert added c: categories labels May 5, 2010

hivert assigned nthiery May 5, 2010

hivert added the s: needs review label Jun 1, 2010

This comment has been minimized.

Sign in to view

hivert added p: critical / 2 and removed p: major / 3 labels Jun 1, 2010

hivert added s: positive review and removed s: needs review labels Jun 2, 2010

williamstein added s: needs work and removed s: positive review labels Jun 3, 2010

This comment has been minimized.

Sign in to view

hivert added s: needs review and removed s: needs work labels Jun 3, 2010

hivert removed the s: needs review label Jun 3, 2010

hivert added the s: positive review label Jun 3, 2010

mwhansen removed the s: positive review label Jun 5, 2010

mwhansen closed this as completed Jun 5, 2010

hivert mentioned this issue Jun 5, 2010

Allows triangular morphisms to have domain and codomain with different index sets #8876

Closed

jbandlow mentioned this issue Jun 6, 2010

map_support modifies the zero element of a free module, and related fixes and cleanup #8890

Closed

hivert mentioned this issue Jun 5, 2010

Have CombinatorialClass inherits from Parent #8910

Closed

anneschilling mentioned this issue Jun 9, 2010

Categorification of Crystals #8911

Closed

hivert mentioned this issue Jun 5, 2010

_name is set too late when creating a combinatorial free module #9104

Closed

nthiery mentioned this issue Jan 14, 2023

Improve the category primer and tutorial #9105

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Functorial constructions in categories #8881

Functorial constructions in categories #8881

hivert commented May 5, 2010

hivert commented Jun 1, 2010

hivert commented Jun 1, 2010

hivert commented Jun 1, 2010

nthiery commented Jun 1, 2010

This comment has been minimized.

hivert commented Jun 1, 2010

nthiery commented Jun 2, 2010

nthiery commented Jun 2, 2010

nthiery commented Jun 2, 2010

hivert commented Jun 2, 2010

nthiery commented Jun 2, 2010

williamstein commented Jun 3, 2010

hivert commented Jun 3, 2010

This comment has been minimized.

hivert commented Jun 3, 2010

hivert commented Jun 3, 2010

mwhansen commented Jun 5, 2010

Functorial constructions in categories #8881

Functorial constructions in categories #8881

Comments

hivert commented May 5, 2010

hivert commented Jun 1, 2010

hivert commented Jun 1, 2010

hivert commented Jun 1, 2010

nthiery commented Jun 1, 2010

This comment has been minimized.

hivert commented Jun 1, 2010

nthiery commented Jun 2, 2010

nthiery commented Jun 2, 2010

nthiery commented Jun 2, 2010

hivert commented Jun 2, 2010

nthiery commented Jun 2, 2010

williamstein commented Jun 3, 2010

hivert commented Jun 3, 2010

This comment has been minimized.

hivert commented Jun 3, 2010

hivert commented Jun 3, 2010

mwhansen commented Jun 5, 2010