Adapt to changes in Determinize registrations in the core

Commit automata-library@1a4de925 in the core merged all the no-op Determinize overloads into one.

This brings two issues.

First, the edges between boxes report cast issue

Type
automaton::DFA<Universal type, unsigned>
is not castable to type 
variant<
  automaton::DFA<Universal type, Universal type>, 
  automaton::DFTA<Universal type, Universal type>, 
  automaton::DPDA<Universal type, Universal type, Universal type>, 
  automaton::InputDrivenDPDA<Universal type, Universal type, Universal type>, 
  automaton::OneTapeDTM<Universal type, Universal type>, 
  automaton::RealTimeHeightDeterministicDPDA<Universal type, Universal type, Universal type>, 
  automaton::SinglePopDPDA<Universal type, Universal type, Universal type>,
  automaton::UnorderedDFTA<Universal type, Universal type>, 
  automaton::VisiblyPushdownDPDA<Universal type, Universal type, Universal type>
>

Second, the overload in the algorithm list now reports this and I have no idea how to report the types of the variants.

cc @travnja3

added C-webui P-high T-bug labels

changed the description

changed title from Adapt to changes in Determinize in the core to Adapt to changes in Determinize registrations in the core

You get the whole type and you crop the template params, is that so? Maybe you could nest into the template params of the variant and remember its options. With that, you may check the compatibility against any of the options as usual and display them on demand I guess.

well, it appears that my reply by email sent 2 hours ago didn't go through.

• The cast issue is solvable by parsing the template type and I will implement that someday.
• No idea what to do about the display issue in the algorithm list, perhaps it will be solved by #65 (closed)

assigned to @peckato1

Well, yes, I discard template part. I think I can check whether T is in variant<T,U,...> quite easily.

But I don't know what to do with the visualization in the algorithms list. But that could be solved by another issue that will discard the overloads from the algorithms list. I don't know yet.

I'm not sure about the visualisation at this point either:-(

This is an old comment (see "well, it appears that my reply by email sent 2 hours ago didn't go through.") that actually got through today when the mail replies got fixed and all the emails in the queue were received.

mentioned in commit 206538d3

mentioned in commit d1059458

mentioned in commit 08e51340

mentioned in merge request !53 (merged)

mentioned in commit 679826a8

mentioned in commit 92710e07

This was "unintentionaly" resolved by !53 (merged)

closed

mentioned in commit d34d2525

mentioned in commit 03ca3d2a