[GiNaC-list] questions about CLN build system [Was: about CLN versus win32 patch]

[Richard B. Kreckel]

Sheplyakov Alexei wrote:

>>But I don't think I'm ever going to see one.
>>And I'm sure that  pursuing such a patch is not worth the effort since
>>it would only save a couple of minutes of every aspiring maintainer
>>(how many of these are there?).
>>    
>>
>It would save _a lot_ of time for many CLN users.
>  
>

Sorry, I don't buy that. I estimate that about 90% of all CLN users take 
.rpm or .deb package from their distros. Of the remaining users, 90% do 
configure / make / make install, nothing more. Heck, they don't even 
read the INSTALL file! Of the remaining library users, only 10% fiddle 
with their own options or flags. Of these remaining 0.1%, nobody has 
ever complained about distcc not working (well, until two days ago, that 
is).

Don't get me wrong: I very much appreciate your keen interest in making 
CLN more, well, easier to use for the casual autoconf hacker. But in my 
experience, hacking these scripts is an extremely time-consuming task 
and in my opinion not at all worth it. Case in point: your last patch 
that disabled shared libraries for MinGW turned out to be buggy -- not 
in theory, but in practice (check out the fixed version and believe me, 
it's needed, as silly as it looks). I was going to release CLN 1.1.12 
and this issue has ruined my day until I realized how it interfered with 
a bug in certain autoconf versions.  :-(

For me, these scripts are simply a time-sink that keeps me away from 
more important tasks. If you wish to submit an elegant patch, you are 
welcome. But if you want me to apply it, you must also take the 
responsibility and assume maintainership of the build-system until the 
initial problems have been sorted out and fixed! Call me a bullhead, but 
otherwise a change in the build-system is not going to happen anytime soon.

Best wishes
  -richy.

-- 
Richard B. Kreckel
<http://www.ginac.de/~kreckel/>