Argument regular= makes Partitions(...) silently ignore other arguments

[maxale]

Steps To Reproduce

Consider the code:

sage: Partitions(24,length=8,max_part=5).cardinality()
33
sage: Partitions(24,length=8,max_part=5,regular=5).cardinality()
950
sage: Partitions(24,regular=5).cardinality()
950

Expected Behavior

When in addition to length=8,max_part=5 we specify regular=5 we expect to see the cardinality at most 33, but we get a larger cardinality of 950.

Actual Behavior

Partitions(24,length=8,max_part=5,regular=5) seems to ignore length=8,max_part=5 and to produce the same result as Partitions(24,regular=5).

Additional Information

No response

Environment

• OS: Ubuntu 24.04.1 LTS
• Sage Version: 10.5.beta9

Checklist

• I have searched the existing issues for a bug report that matches the one I want to file, without success.
• I have read the documentation and troubleshoot guide

[mantepse]

Did you check whether this is fixed with the pr concerning #38897 ?

Indeed, with #38904 I get

sage: Partitions(24,length=8,max_part=5,regular=5).cardinality()
...
ValueError: the parameters 'parts_in', 'starting', 'ending', 'regular' and 'restricted' cannot be combined with anything else

[maxale]

@mantepse: That's better and somewhat fixes the issue in the original code example. However, the new ValueError is inconsistent with the documentation for regular=, which says:

regular=ell specifies that the partitions are $\ell$-regular, and can only be combined with the max_length or max_part, but not both, keywords if $n$ is not specified

Note that it always allows combining regular= with one of max_length or max_part, and even with both when argument n is given.

Hence, from reading this I'd expect the following modified example work Partitions(24,max_length=8,max_part=5,regular=5).cardinality() and its result not exceed that without regular=5.

[maxale]

Or am I misreading the docs? The other potential meaning may be as follows:

regular=ℓ specifies that the partitions are ℓ-regular; it cannot be combined with any other keywords unless n is not specified, in which case it can be combined with one, but not both, of the max_length or max_part keywords.

If that's the intended meaning, then the docs wording needs to be improved (eg. like I suggest above).

PS. And even here the "both keywords" case is not actually forbidden to the docs contrary:

sage: Partitions(max_length=8,max_part=8,regular=5)
Partitions satisfying constraints max_length=8, max_part=8, regular=5