insane.bbclass: Add missing R* variables to pkgvarcheck

* check also RSUGGESTS, RCONFLICTS, RPROVIDES, RREPLACES

---
 meta/classes/insane.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 58fc70c..d285e56 100644
@@ -908,7 +908,7 @@ python () {
 
     issues = []
     if (d.getVar('PACKAGES', True) or "").split():
-        for var in 'RDEPENDS', 'RRECOMMENDS', 'FILES', 'pkg_preinst', 'pkg_postinst', 'pkg_prerm', 'pkg_postrm':
+        for var in 'RDEPENDS', 'RRECOMMENDS', 'RSUGGESTS', 'RCONFLICTS', 'RPROVIDES', 'RREPLACES', 'FILES', 'pkg_preinst', 'pkg_postinst', 'pkg_prerm', 'pkg_postrm':
             if d.getVar(var):
                 issues.append(var)
     for i in issues:

1.8.1.2

These are already checked by recipe_sanity.bbclass. There's probably no need to check them again here.

p.

recipe_sanity.bbclass doesn't get used by default - I guess you are enabling it explicitly? I wonder if we should either (a) move all the recipe checks to recipe_sanity.bbclass and enable it by default, or (b) move any currently there and not covered by insane.bbclass to insane.bbclass and delete recipe_sanity.bbclass.

Cheers,
Paul

Paul Eggleton
Intel Open Source Technology Centre

Yeah, or rather, it's used by default in meta-micro.

Either of those sounds reasonable enough to me. I think the original concept was that recipe_sanity was for errors that you could identify statically at recipe parse time, and insane.bbclass was for checks against the output artifacts (that you could only do after compiling and packaging) but that line does seem to have gotten a bit blurred over time.

p.