Generic not_equals methods for objects in coding theory library

[sagetrac-dlucas]

In this ticket, we propose a default implementation of __ne__ methods for AbstractLinearCode, Encoder and Decoder classes, which means one will only need to implement __eq__ in the subclasses.

This new __ne__ simply returns the inverse of the return value of __eq__.

Component: coding theory

Author: David Lucas

Branch/Commit: f55e892

Reviewer: Vincent Delecroix

Issue created by migration from https://trac.sagemath.org/ticket/19743

[sagetrac-dlucas]

Branch: u/dlucas/default_not_equals_for_linear_codes

[sagetrac-dlucas]

comment:3

I removed the old implementation of __ne__ in grs.py, created generic implementation for the top classes Decoder, Encoder and AbstractLinearCode and implemented __eq__ for encoders in linear_code.py.

This is now open for review.

---

New commits:

490f739

Removed old `__ne__` methods, implemented generic `__ne__` in Decoder, Encoder and AbstractLinearCode

[sagetrac-dlucas]

Commit: 490f739

[videlec]

comment:4

Better use not self == other as it is faster and more flexible. The only reason to not do so is if your equality might return NotImplemented.

[jdemeyer]

comment:5

I agree with [comment:4]: avoid a.__eq__(b) unless you have a good reason for it.

[sagetrac-git]

Changed commit from 490f739 to f55e892

[sagetrac-git]

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

f55e892

Changed not a.__eq__(b) to not a == b in generic `__ne__` methods

[sagetrac-dlucas]

comment:7

Thanks for the advice!

I changed my code accordingly.

David

[videlec]

Reviewer: Vincent Delecroix

[vbraun]

Changed branch from u/dlucas/default_not_equals_for_linear_codes to f55e892