I'd like all PETSc functions (those with manual pages) to either 1) have a Fortran stub (either generated automatically or handwritten) OR 2) have documented in the manual page No Fortran Equivalent.
We look pretty sorry ass when we are always missing Fortran stubs that could have been generated automatically or have not been implemented for no technical reason. The (slightly) HARD part. This means we need a python tool that generates a list of all manual page functions that DO NOT have fortran stub nor No Fortran Equivalent written in the page. With this we can easily keep up to date on stubs; without we'll never get our stubs right :-( Barry
