home bbs files messages ]

Forums before death by AOL, social media and spammers... "We can't have nice things"

   linux.debian.bugs.dist      Ohh some weird Debian bug report thing      28,835 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 27,703 of 28,835   
   Nilesh Patra to Helmut Grohne   
   Bug#1126934: lintian: raise an error on    
   15 Feb 26 09:40:01   
   
   From: nilesh@debian.org   
      
   On 12/02/26 2:10 pm, Helmut Grohne wrote:   
   >> Do you mind rewording this and sending across a MR? I'll be happy to quickly   
   >> merge that.   
   >   
   > There are several cases here and we may want to consider the most common   
   > patterns. I think by far the most common pattern is not having installed   
   > tests. At that point, on ignores noinsttest and just goes with nocheck.   
   >   
   > The second most common pattern probably is not having any tests.   
   >   
   > Neither of these are relevant to the tag as it only comes up once you   
   > skip installing anything based on nocheck. So when the developer reads   
   > the tag they are quite definitely dealing with installed tests. Most   
   > commonly, installed tests are also run. So my expectation is that the   
   > common case here is needing both. That's why I specifically added it.   
      
   I do understand your reasoning here. The problem occurs in cases where there   
   are B-D that need to be marked with just !nocheck and B-D that need to be   
   marked   
   with !nocheck and !noinsttest both in the same source package; dbus for   
   example.   
      
   I find it problematic if the maintainer sees the explanation and ends up   
   marking   
   more than required B-D with both the tags. It'd be another problem if   
   maintainer   
   carries down this (wrong) understanding to other packages.   
      
   > Would the following rephrasing improve the situation?   
   >   
   > | If a build dependency is required both for running build-time tests   
   > | and for building installed tests, ...   
   There are packages that install tests also in the same binary package along   
   with   
   the resulting program. It is not ideal, and should be split (which is another   
   issue).   
   However, I also think that !noinsttest would not apply in such a situation?   
      
   What do you think about this?   
      
   | If a build dependency is needed both to run build-time tests and to build   
   | installed tests that reside in a separate binary package with !noinsttest   
   profile, ...   
      
   Best,   
   Nilesh   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca