[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#359059: lintian: Allow vendor/user specific checks



On 2012-10-12 11:45, Jakub Wilk wrote:
> * Niels Thykier <niels@thykier.net>, 2012-10-11, 23:22:
>>> How will overrides play with contrib checks? [I had a user who wanted
>>> to add an override for lintian4python, though there's currently no
>>> way to do that.]
>> I think adding them to the standard overrides file will "just work".
>> Do you have a sample lintian4python case where the override fails to
>> work?
> 
> Hmm. I was convinced that lintian would emit unused-override if it sees
> an override for a non-existent tag[0], but apparently this is not the
> case. (Bug or feature?)
> 

It is intentional (i.e. a feature).  I see this part is not covered by
our Manual, but it was in the "Vendor Profile" proposal(s)[0].

> 
> [0] I have a distinct memory of seeing spurious unused-override tags in
> early stage of lintian4python development. But maybe my memory is
> playing tricks on me.
> 

Depending on when you did this testing last, your memory may be true.
At the very least, Vendor Profiles were not implemented until 2.5.2
(from Aug 2011).  Plus, it may not have worked properly until 2.5.5[1].

Up to version 2.5.2, Lintian would indeed emit unused-override for
unknown tags[1].  But with the introduction of Vendor profiles, Lintian
will only emit unused-overrides for "active tags" (i.e. tags from the
select profile that has not been suppressed).



~Niels

[0] http://wiki.debian.org/Lintian/Spec/VendorCustomization
https://lists.debian.org/debian-lint-maint/2011/04/msg00277.html

"""
 * A tag not included in the current profile shall not be emitted.
   * An override for such a tag is not considered "unneeded" and must be
     ignored.
"""

[1] I haven't tested, but the following change sounds like it would
affect it (as L::Tags determines if an override is unused or not).

  * lib/Lintian/Tags.pm:
    [...]
    + [NT] Use a Profile to determine if a tag is suppressed or not.


Reply to: