Opened 19 months ago
Last modified 19 months ago
#67288 new defect
R: many packages ask to link to Fortran libs, while having no Fortran code at all: keep that or patch out ldflags?
| Reported by: | barracuda156 | Owned by: | |
|---|---|---|---|
| Priority: | Normal | Milestone: | |
| Component: | ports | Version: | 2.8.1 |
| Keywords: | Cc: | kjellpk (Kjell Konis), catap (Kirill A. Korinsky), i0ntempest | |
| Port: | R |
Description
Numerous R packages put FLIBS into PKG_LIBS in attached Makevars files, while having zero Fortran code (well, I did not read every line of code, admittedly, but no .f90 or whatsoever Fortran file formats, nothing is compiled with gfortran etc.). It is not a big deal to satisfy that requirement, of course, all it needs is compilers.setup in the portfile. However I am not sure if that is a desirable, after all. It is also possible to patch out FLIBS, at least when we notice that Fortran is not involved.
Any opinions?
(To be clear, I do not propose to do this as a once-for-all clean-up measure for 1500+ packages, it is unrealistic. But whenever it is relevant and feasible to do.)
Change History (3)
comment:1 Changed 19 months ago by ryandesign (Ryan Carsten Schmidt)
| Cc: | kjellpk added; ryandesign removed |
|---|
comment:2 follow-up: 3 Changed 19 months ago by ryandesign (Ryan Carsten Schmidt)
comment:3 Changed 19 months ago by barracuda156
Replying to ryandesign:
Example: R-hts: https://build.macports.org/builders/ports-10.14_x86_64-builder/builds/163974/steps/install-port/logs/stdio
Oops. Will fix that. Wonder how it passed CI or what happened.
Example: R-hts: https://build.macports.org/builders/ports-10.14_x86_64-builder/builds/163974/steps/install-port/logs/stdio