Binary relations in Sage implemented

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

I have implemented binary relations (including composition of relations, set theoretic operations, conversion back/forth to graphs and to posets, etc.)

Component: packages: standard

Author: Victor Porton

Branch/Commit: public/relations @ aa1564c

Issue created by migration from https://trac.sagemath.org/ticket/24542

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

Attachment: relations.patch.gz

Patch for Sage "develop" branch

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

Changed branch from develop to none

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

Changed commit from a178204 to none

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

Branch: public/relations

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

Changed branch from public/relations to develop

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

Commit: a178204

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

Changed branch from develop to public/relations

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

Changed commit from a178204 to none

[7ed8c4ca-6d56-4ae9-953a-41e42b4ed313]

Commit: bdf7549

[7ed8c4ca-6d56-4ae9-953a-41e42b4ed313]

Branch pushed to git repo; I updated commit sha1. Last 10 new commits:

0ff39ae	to_graph() -> to_digraph()
fc900dc	methods made properties
73e5c98	transitive_closure
ef76125	removed a @property
72628f5	binary relations to/from (di)graphs
f5a9be2	(pre)order tests
bcd24cd	TODO comment
43d743a	added cartesian product and identity relation
7714732	integration of posets with binary relations
bdf7549	doc comments changes

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

Author: Victor Porton

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

Description changed:

--- 
+++ 
@@ -1,3 +1 @@
 I have implemented binary relations (including composition of relations, set theoretic operations, conversion back/forth to graphs and to posets, etc.)
-
-The patch is attached.

[kcrisman]

comment:8

This could be quite useful. It looks like you need quite a bit of documentation/examples for the many methods involved. Especially the back/forth with posets, if robust, would be great - see e.g. this stack overflow question.

[7ed8c4ca-6d56-4ae9-953a-41e42b4ed313]

Changed commit from bdf7549 to 5fc5429

[7ed8c4ca-6d56-4ae9-953a-41e42b4ed313]

Branch pushed to git repo; I updated commit sha1. New commits:

5fc5429

added FiniteBinaryRelation.{image,reverse_image}

[7ed8c4ca-6d56-4ae9-953a-41e42b4ed313]

Branch pushed to git repo; I updated commit sha1. New commits:

aa1564c

added FiniteBinaryRelation.enumerate_all()

[7ed8c4ca-6d56-4ae9-953a-41e42b4ed313]

Changed commit from 5fc5429 to aa1564c

[embray]

comment:11

I'm surprised something like this isn't already in Sage though that's not something I'm really qualified to comment on. I suspect there's also probably a better place to put this than starting a new top-level package.

I can make some minor nitpicks on the code itself if you think that would be helpful.

[cd38fdd3-72ba-48dd-ac57-d8359c63fec8]

comment:12

@embray I don't know if your nitpicks are good. After all, I know nothing about you. You decide if you do the work.

I am going to add yet one feature: Converting a function into a binary relation. However I will be busy today. I have not yet dived into details of this feature.

[slel]

comment:13

There is interest for this, see

• Ask Sage question 45079: Binary relations

[kwankyu]

comment:14

Replying to @embray:

I suspect there's also probably a better place to put this than starting a new top-level package.

I agree. A top-level package is supposed to contain many modules and/or packages.

I suggest to put this module into sage.combinat.relations.

On the other hand, sage.sets.relations would be a bad choice as sage.sets is concerned with generic sets.