Weird multiplication by identity in set_partition_ordered.py

[darijgr]

See the attached patch which, I hope, speaks for itself. I don't see any reason why that permutation had to be multiplied by an identity permutation. I thought it might have been meant to coerce the permutation into a larger symmetric group or to avoid some corner-case bugs, but this doesn't seem to be an issue (cf. the doctests I added). Also, notice that OrderedSetPartitions class does already check for the set and the composition to be of the same size, whereas the un-exported OrderedSetPartitions_scomp class throws various errors if they aren't (both before and after my patch). The patch has a minor but nontrivial effect on speed:

sage: timeit("list(OrderedSetPartitions([1,2,3,4,5],[3,2]))")

now takes 6.75 rather than 7.67 ms.

Depends on #14772
Depends on #8386
Depends on #14519
Depends on #14808
Depends on #14143
Depends on #14015
Depends on #14516

Component: combinatorics

Keywords: combinat, ordered set partitions, permutations

Author: Darij Grinberg

Reviewer: Travis Scrimshaw

Merged: sage-5.12.beta4

Issue created by migration from https://trac.sagemath.org/ticket/14883

[darijgr]

comment:1

Attachment: trac_14883-remove_times_id-dg.patch.gz

[fchapoton]

comment:2

patch does not apply, maybe because the dependency is not written correctly, let us see.

[fchapoton]

Changed dependencies from 14772 to #14772

[darijgr]

comment:3

Oops, the dependency definitely needed a hash. But how does that affect the failure of the patch? Do you have a bot which pulls dependencies automatically?

[fchapoton]

comment:4

yes, the patchbot is smart enough to do that, indeed. Click on the red circle and you will see !

[darijgr]

comment:5

Oh, the patchbot. I'm not (yet) smart enough to understand its output...

[tscrim]

Reviewer: Travis Scrimshaw

[tscrim]

comment:6

Looks good to me.

[jdemeyer]

Changed dependencies from #14772 to #14772, #8386, #14519, #14808, #14143, #14015, #14516

[jdemeyer]

Merged: sage-5.12.beta4