Tommy Hofmann
Tommy Hofmann
We will take care of it when overhauling the`singular_ring`, see also https://github.com/oscar-system/Oscar.jl/issues/975 and https://github.com/oscar-system/Oscar.jl/issues/976. It is a broken design, so I would not spend too much time on it.
Here is an example where using `n_unkown` creates a nice segmentation fault without stack trace for the innocent user. ```julia julia> F, = Singular.Nemo.FiniteField(2^17 - 1, 2, "a"); julia> Fx,...
Note that this is crucial if we want to make Singular useful for Oscar. The situation with the segmentation faults is very off-putting for users.
looks good
Ah, no, not quite right. One has to run the `build_tarballs.jl` from within the `P/path_to` directory. Because the local patches have a relative path with respect to the `build_tarballs.jl`. (I...
Whatever the automatic process does, I think it should not be made the official version without the authors giving their "yes". The code blocks without the text have the potential...
Sorry, I reread the proposal and I think I misunderstood something. Ignore my comment.
The example errors with ``` julia> R = parent(h) ERROR: UndefVarError: `h` not defined Stacktrace: [1] top-level scope @ REPL[5]:1 ```
I think we can make the gcd faster, see https://github.com/thofma/Hecke.jl/pull/1505. It still needs a bit of tweaking.
Still under construction, so cannot give a definite answer yet.