>>>>> "TA" == Thomas Anders <[EMAIL PROTECTED]> writes:

TA> It's a start, but I think our current approach of shipping all
TA> autoconf variables in net-snmp-config.h is even more flawed than
TA> that. See some random Debian bug (wrt. some other software) on the
TA> topic:

TA> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=363173

TA> Are we willing to stop bad practice for 5.4?

No.  We're not doing anything awful and I challenge you to tell me we
are.  We distribute the file as net-snmp-config.h so no one else will
have the same name (and it's in a subdirectory too of course).

That file documents all our own stuff that are needed to do proper
source compilation both within our own code and what mib2c produces
and what the general population will likely need to do as well.  There
is *nothing* that says "you can't compile code unless you include it".
Even with the PACKAGE definitions in it, we're not doing anything
wrong by distributing it.  People use it because its about 1000 times
easier to use than to not use it.  In fact, long long ago we didn't
distribute it (install it actually) and people had such a hard time
writing portable code by including the right set of headers that we
started installing it and the number of issues surrounding helping
people use the right headers, for example, went way down.

We shouldn't stop shipping (installing) the results of our configure
tests.  It tells people all about how the package was compiled.  It
drastically simplifies the majority of the coding issues people have
in the real world.

What we *should* do is:

1) to start telling people that have conflicts that they need to not
   rely on the easy to use includes and they'll have to do the hard
   work that all of our users used to have to do (this is .01% of the
   population).

2) fix the incompatibilities when we can (removing the PACKAGE
   variables is an excellent example).

3) do some separation of required defines vs non-required defines.

But not installing the config.h file is right out because it's simply
not harmful to do so.  The people that are complaining about conflicts
are likely just as lazy and don't want to do their own header
analysis.  We can, of course, make it easier and we should but making
it harder is what I'm seeing as being suggested.  And that's worse.
-- 
Wes Hardaker
Sparta, Inc.

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to