SLEPcEigenSolver.solve(n) seems to return n+1 instead of n Eigenpairs

[Sebastian Otten]

Description is in the title.

If I do eigensolver.solve(0), then I can actually plot the 0-th Eigenpair afterwards. Not sure if the docs are wrong or the solver behaviour is wrong.

Issue is not obviously not very big but definitely causes confusion.

I attached a working example.

best regards :)

[Anders Logg (Chalmers)]

Removing milestone: 1.7 (automated comment)

[Jan Blechta]

I think that SLEPcEigenSolver.solve(n) is required to get at least n eigenpairs. So I don't find this as a problem.

[Jan Blechta]

You correctly check in you example number converged. This is what user should do. So I'm closing this. Please propose a pull request fixing documentation if in desperate need.