Matthew Hambley
Matthew Hambley
We may need to support old fashioned fixed format Fotran 77 code with the `.f` or `.F` suffix. Some testing to ensure this.
It seems that failure of some steps doesn't always lead to the build stopping. In particular, when analysis failed due to issue #221 Fab carried on to compile which obviously...
An experiment script to build the `io_dev` LFRic application (attached) is having trouble compiling `iterative_solver_mod` due to not being able to find `abstract_linear_operator_type` and `abstract_preconditioner_type` despite both module files having...
Cray compiler can produce `.lst` files for each file compiled. Maybe keep them next to the `.o` files. Maybe also keep compiler stdout/stderr output there too. How would this affect...
Logging is a little slapdash. Some thought needs to be given to what should be logged and at what level. Other sub-issues: - [ ] #355 - [ ] #446...
It looks like the analysis stage is being tripped up by compiler line control directives. These are used by the preprocessor to indicate how lines in the resulting file match...
Access is needed to compiler `stdout` so that warnings may be inspected. A hook allowing post compile processes to crawl this output would allow existing warning checking scripts to be...
It seems like `program` program units are being incorrectly flagged as "unused". Is this because they are used by the linker, not by Fortran `use` statements?
Fab should be able to identify operations it performs which do not contribute to the build. For example source which is extracted but never used. Source which is processed but...
Build scripts involve a lot of boiler plate. If common usage patters are identified there might be scope for helper functions to reduce verbiage. There is a balance to be...