Selection edges not constrained to intersection

[ghost]

It seems to me that the highlighted region edges do not match the selected intersection. For example, in the screen shot below, the highlighted area seems to include A∩B∩C and B∩C, i.e. (A∩B∩C) ∪ (B∩C) . However, the mouse is only hovering over B∩C, and the count is correctly given for just B∩C.

Or am I not interpreting things correctly here?

This particular example was produced with this data:

[
{"sets":["A","B","C"],"size":1},
{"sets":["A","B"],"size":1},
{"sets":["A","C"],"size":1},
{"sets":["B","C"],"size":1},
{"sets":["A"],"size":3},
{"sets":["B"],"size":3},
{"sets":["C"],"size":3}
]

Thank you for this great package. I have found creating venn diagrams surprisingly tricky!

[benfred]

This package assumes that the set areas overlap - so the size of A includes the size of A∩B and A∩C and A∩B∩C in it. In other words, the size of the inputs aren't assumed to be disjoint from all the other areas.

When it selects B∩C its trying to show all the items that are in common between B and C, which will include those that are also shared by A.

There is some talk about this in this issue #23

Hope this helps!

[ghost]

Yes, fair enough. Indeed, I now realize that (A∩B∩C) ∪ (B∩C) is the same as just (B∩C).

But in our use case it is important to know how many are, for example, in B and C but not A, i.e. (B∩C) - (A∩B∩C). I.e., how many are just in the region where the mouse pointer is in the screen shot above. Sounds like that is not possible in the present version?