Steve Linton
Steve Linton
The problem is how to tokenise something like ``` myrecord.17 ``` without a clue from the parser level, this could be either a name followed by the float `.17` or...
I'd support Frank's proposal. All representations should imply exactly one of the four base representations.
I just ran into the same problem. Also the methods run into a divide by zero over finite fields.
What we want is, I think, impossible. We want to find every reference to the "ToBeDefined" placeholder anywhere in the workspace and replace it by a reference to the exact...
Working to clean up the test failures. Mostly straightforward, but there is a problem with tests in `hpcgap/tst/testinstall/comprvec.tst` like ``` gap> F:=GF(41^3);; gap> v:=Filtered(F,x -> x in GF(41));; gap> IS_VECFFE(v);...
If this is merged, we might want to add a virtual enumerator for Galois Fields instead of making a plain list.
The remaining test failures arise from interactions between `factint` and HPC-GAP and from the problems building `NormalizInterface` already mentioned.
Now just down to the `normaliz-interface` issues. Thinking about that, though, I wondered about where `ffdata.c` and `ffdata.h` should really go. At the moment they are in `gen` but they...
@fingolfin Thanks a lot. Take your time. Although, before this PR gets too old, we need to decide whether to include it in 4.11, forcing the change on `NormalizInterface` or...
@fingolfin Given that, as you say, it is not actually a problem on modern machines, I'm not inclined to worry about this. The root problem is that we need to...