From: smcv@debian.org 
  
 On Sat, 27 Sep 2025 at 10:48:44 +0200, Samuel Thibault wrote: 
 >Martin-€ric Racine, le sam. 27 sept. 2025 09:41:58 +0300, a ecrit: 
 >> IMHO, in order for Lintian's severity levels to be meaningful in 
 >> determining a package's fitness for inclusion in the Debian 
 >> repository, an Error ought to refer to a MUST[NOT] Policy item, 
 > 
 >Tag: apache2-configuration-files-need-conf-suffix 
 >Severity: error 
 >Check: apache2 
 >Explanation: The package is installing an Apache2 configuration but that 
 file 
 does not 
 > end with a '.conf' suffix. Starting with Apache2 2.4 all 
 configuration 
 > files except module '.load' files need that suffix or are 
 ignored otherwise. 
 > 
 >This is clearly an error that we want to highlight, while not actually 
 >being a problem for inclusion in debian, so won't be covered by the 
 >policy. 
  
 I think this is an excellent example, and I agree with Samuel that it 
 would not make sense for Policy to forbid installing 
 /etc/apache2/conf.d/foobar: the relevant more general meme is "Policy 
 does not aim to forbid every broken situation". 
  
 It sort of is a problem for inclusion in Debian, because it's a bug of 
 unknown severity: if you install /etc/apache2/conf.d/foobar into a 
 package, expecting Apache to read it, then your package will not work as 
 designed, which could be anywhere between Severity: grave and Severity: 
 minor (depending on the package, and the importance of its Apache 
 configuration being read). But not every bug is a Policy violation. 
  
 Similarly, I don't think it makes sense for every small-p policy to be 
 part of Policy. We have things like a Python policy and a 
 GObject-Introspection mini-policy, which not every DD needs to know 
 about, only the ones who actually maintain relevant packages. 
  
 Historically, Lintian internally classified tags along two axes - impact 
 and confidence - and then multiplied them to get the severity. For 
 example, assessing /etc/apache2/conf.d/foobar as a mistake is 
 high-confidence (I don't see any legitimate reason why a package 
 maintainer would install a file to that directory unless they were 
 expecting Apache to read it!), and that makes it a Lintian error, even 
 though we cannot be sure that the resulting bug report from "Apache 
 doesn't actually read this file" would be RC. 
  
 We also have Lintian tags for things like "Depends on a deprecated 
 package that we are trying to phase out" or "relies on a deprecated 
 behaviour that we are trying to phase out", as an alternative to 
 mass-bug-filing. I hope you're not suggesting that we should be going 
 through the intentionally formalized and heavyweight process of 
 proposing Policy changes in order to say "Packages SHOULD NOT depend on 
 python3-pdm-pep517" and "Packages SHOULD NOT depend on 
 libgirepository1.0-dev", among others... 
  
      smcv 
  
 --- SoupGate-Win32 v1.05 
  * Origin: you cannot sedate... all the things you hate (1:229/2)