tags 930679 + moreinfo thanks Felix Lechner wrote:
> > > It would be even better to detect some of the adequate justifications > > > automatically like Haskell packages. > > I think Sam meant that some circumstances should always prevent the > issuance of the tag, so developers are not burdened with the override. > Apparently, Haskell packages do not use dh but cabal. [1] Let's 100% nail this bit down so we don't confuse each other and annoy other maintainers. So, Haskell packages use cdbs calling into a Haskell-specific hlibrary.mk. 1 #!/usr/bin/make -f 2 3 DEB_SETUP_BIN_NAME = debian/hlibrary.setup 4 DEB_CABAL_PACKAGE = yi-misc-modes 5 DEB_DEFAULT_COMPILER = ghc 6 7 include /usr/share/cdbs/1/rules/debhelper.mk 8 include /usr/share/cdbs/1/class/hlibrary.mk I mean we could certainly just whitelist all of src:haskell-*, but isn't the entire point of this tag to ask people to move to the dh sequencer? Or is it "actually fine" for them to use hlibrary.mk and this will just save the Haskell team whole bunch of boilerplate overrides with an identical explanation? > > c) The severity (ie. E/W/I/P) > > I'd recommend starting out with warning. Some day it might move to > error, but I think starting out there would be overly aggressive. My experience of Lintian suggests that W: would be too strong to start with. Sam, would you be okay with "I:" to begin with? Best wishes, -- ,''`. : :' : Chris Lamb `. `'` la...@debian.org 🍥 chris-lamb.co.uk `-