New datastructure for iet

[vdelecroix]

It was decided to develop flatsurf as an independent package

---

Old description

In #8386 we decided to only move iet to sage/dynamics. This ticket stands for the creation of the new datatype and the correction of the following bug

sage: p = iet.Permutation('a b','b a')
sage: q = iet.Permutation('b a','a b')
sage: p == q
True

[vdelecroix]

New commits:

​530fe85

Initial commit: new datastructure for iet.

[chapoton]

rebased. Is this needs review ?

---

New commits:

​4d69202

Merge branch 'u/vdelecroix/14683-iet_new_datastructure' of ssh://trac.sagemath.org:22/sage into 14683

[vdelecroix]

Hi Frederic,

Thanks for the update. It would be nice if you provide clearer message with your commit. The following is completely meaningless

Merge branch 'u/vdelecroix/14683-iet_new_datastructure' of
ssh://trac.sagemath.org:22/sage into 14683

Vincent

[chapoton]

This was just a rebase, as explained in the comment. Once again, is this ticket in "needs review" ?

[vdelecroix]

Hi Frederic,

Your comment on trac will not appear in the git history. So please, set your message to "rebase #14683 on sage-X.betaY" or something meaningful. The fact that it is "just a rebase" is not an argument to have an ugly commit message.

And yes it is in needs review.

Vincent

[git]

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

​7fe8734

trac #14683 pour raise statements into py3 syntax

[git]

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

​c4255b3	Merge with 6.3.beta6
​9464f5c	trac #14683 cleanup of constructor.py (raise in py3, use of ....:, etc)

[git]

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

​ae5d4b5	trac #14683 cleanup of iet.py
​dc8e33f	trac #14683 cleanup of labelled.py
​cf153d8	trac #14683 cleanup of reduced.py

[chapoton]

Hello Vincent,

I am in the middle of the process of reading quickly the code, making small cosmetic changes, like putting doctest continuation and raise statements to new standards, etc. Once done, I will run the tests.

If the tests pass, I could give a positive review, but without checking the new data structure. would it be ok to you ? Have you used this branch heavily, so that you know that it works (and is really better than the current state) ?

[git]

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

​880af14	trac #14683 cleanup of template.py
​db9430e	trac #14683 start of pyflakes cleanup

[jdemeyer]

What's your justification for removing doctests like

TESTS::

    sage: t(-3/2)
    Traceback (most recent call last):
     ...
    ValueError: value must positive and smaller than length

and replacing the raise ValueError with an assert?

(note that I don't really care about this ticket, I am just looking for abuses of assert)

[vdelecroix]

Replying to jdemeyer:

What's your justification for removing doctests like

TESTS::

    sage: t(-3/2)
    Traceback (most recent call last):
     ...
    ValueError: value must positive and smaller than length

and replacing the raise ValueError with an assert?

Bad design choice ;-P

This ticket is pending since it needs some more intensive testing/proof of concepts.

Vincent

[git]

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

​fc249b6	Merge branch 'public/ticket/14683' into 6.8
​62797e9	trac #14683 rebased and all tests pass

[git]

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

​2b2d683

Merge branch 'public/ticket/14683' into 7.2.rc0

[chapoton]

ok, I agree this could be closed. But maybe there would be a short way to solve the bug ?

[vdelecroix]

I am focusing on #20695 and will not fix this bug in the mean time.

[chapoton]

ok, then