Empty hypergraph and __str__ function

[ncohen]

Right now an exception is raised when building the empty hypergraph. This branch fixes it:

sage: sage: IncidenceStructure([])
Incidence structure with 0 points and 0 blocks

Also, right now the class has both a __repr__ and a __str__ function, which means that typing H in a console does not produce the same output as typing print H in a script. The two functions are made equal: I work often with this class and pretty often the result of print H takes a whole screen, as it displays *all blocks*.

Nathann

[git]

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

​871738f

trac #17574: Empty hypergraph and __str__ function

[vdelecroix]

Hey,

It can be useful to keep a function printing all blocks as it is the case with matrices right now with the method .str():

sage: m = random_matrix(ZZ,100)
sage: m
100 x 100 dense matrix over Integer Ring (use the '.str()' method to see the entries)
sage: print m
100 x 100 dense matrix over Integer Ring
sage: m.str()
... # crazy string

And it is true that __repr__ and __str__ coincide.

sage: m.__repr__()
'100 x 100 dense matrix over Integer Ring'
sage: m.__str__()
'100 x 100 dense matrix over Integer Ring'

To get a hint when only m is typed, you need to hack repl/display/fancy_repr.py.

Vincent

[ncohen]

Yo !

It can be useful to keep a function printing all blocks as it is the case with matrices right now with the method .str():

But we have H.blocks(), don't we ? Why would we need something else ?

On the other hand, I often need to control variables, to make sure that the input is a PairwiseBalancedDesign and not just an incidence structure, and all that I could do so far was print H.__repr__().

Nathann

[ncohen]

Beep ?...

[git]

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

​9c0b7d4	trac #17574: Merged with 6.8.beta6
​514dd64	trac #17574: broken optional doctests

[dimpase]

LGTM

[ncohen]

Thaaaaaaaaaaaaaaaaaaaanks!!!

Nathann