[csw-maintainers] /usr/{local,share} references

[Peter FELECAN]

Philip Brown <phil at bolthole.com> writes:

> On Wed, Feb 9, 2011 at 1:05 AM, Dagobert Michelsen <dam at opencsw.org> wrote:
>> Hi Phil,
>
> *wave*
>
>
>>>> I inspected a couple of my package
>>>> and every occurence needs to be verified manually. There are even
>>>> scripts looking for configuration stuff which should keep the include
>>>> to /usr/local but must have /opt/csw manually added and the check
>>>> overridden.
>>>
>>> I would be interested to hear more details, about specific cases where
>>> the above script did something "bad". I think that all maintainers
>>> would benefit from this information.
>>
>> Think of "The default location for package installation is /usr/local if
>> you don't specify --prefix". Changing this to /opt/csw would be plain
>> wrong. There are lots of other cases.
>
> Errr.. those are build instructions. The general case for that type of
> "error", is
> "anything that relates to people *building* the package".
> Yes, its technically "wrong" to change that, but on the other hand, it
> will in no way hurt users, since they're not the ones building the
> package.
>
> Any other class of error for this, that you can think of?

In a previous message of this thread I gave the example of gcc3 where my
README.CSW mentions the lack of /usr/local and the usage of /opt/csw
which is very valid.

> btw, I'm all for a more intelligent auto-changer, that does the
> changes, but also reports back to the maintainer, "this is what I
> changed".
> For the gar afficionados, this should be easy, if you leverage the git
> stuff before and after.
>
>>> But if they dont even look, how are they going to come to the point of
>>> examining the defaults, knowing they are wrong, and "looking again"?
>>
>> Because there is a bug report.
>
> You are making the very VERY wrong assumption, that bug reports,
> magically, accurately, and always appear for problems.
> Unfortunately, the opposite is more often true.

This is as strong as the opposite assumption.

> Just a month ago, I discovered a bug, of this very nature(unfixed
> /usr/local reference), in a package, because of the autochecker. It
> had been there for YEARS. No one had noticed it, or at least no one
> reported it. So the package was missing proper functionality.
> Not critial "it doesnt work!" functionality, but still missing.

Can you give the exact circumstances? Which package and what issue?
-- 
Peter