On Tue, Jul 10, 2018 at 08:10:58AM +0100, Chris Lamb wrote: > Bill, > > > However it seems to flags files named *.rda even when they are completly > > unrelated to R > > For example? We could fix that false-positive, rather than change the > "certainty" level.. :)
The package gap-small-groups includes a file /usr/share/gap/small/id5/id1728.rda.gz and is unrelated to R, yet trigger this warning when present in a source package. By the way this seems to be the corresponding source code: # Ensure we have a README.source for R data files if ( $basename =~ m,\.(?:rda|Rda|rdata|Rdata|RData)$, && $entry->is_file && !$info->index_resolved_path('debian/README.source')) { tag 'r-data-without-readme-source', $name; } Clearly it will flag any files named '*.rda' indepently of its content or purpose. > Either way, is anyone really putting any relevance into the certain/ > possible/etc classifications? Yes: it gives an hint whether the lintian maintainers expect the test to have false positives (and one need to add a lintian exception). Cheers, -- Bill. <ballo...@debian.org> Imagine a large red swirl here.