Identify to return all hits from a census, not just the first

[saulsch]

The veering census is a census of veering structures, not manifolds. And a manifold can have several veering structures (sometimes several on the same triangulation!). So, if M is a manifold, then M.identify() should return all of the known veering structures on M, not just the first.

So, in the block starting line 381 in database.py

SnapPy/src/snappy/database.py

Line 381 in f23aeb3

for mfld in mflds:

add a list

hits = []

instead of returning N as soon as we get a hit, instead

hits.append(N)

and do the same for the following blocks. At the end of the function replace "return None" by "return hits".

Then the calling function merges the lists of hits.

[unhyperbolic]

Interesting suggestion.

Along similar lines, for some manifolds in the tetrahedral census, there are several triangulations by regular ideal tetrahedra of the same manifold. Since all SnapPy censuses are manifold and not triangulation censuses (and SnapPy focused on presenting a hyperbolic manifold without giving the user too many methods to access and modify the underlying triangulation), I picked one such triangulation for SnapPy and created a separate regina file that lists for each manifold the different triangulations, see https://www.unhyperbolic.org/tetrahedralCensus/tetrahedralCensus/regina/

So what you are suggesting is a bit of a departure.

Also: if you can have several veering structures for the same triangulation, how do we surface that to the user?

On another related note: if several links have the same complement, SnapPy would only pick one of links - even though they are different hyperbolic manifolds when taking the peripheral curves into account.