Re: Compiling 2.4-STABLE2 under MacOS X 10.1

From: Henrik Nordstrom <hno@dont-contact.us>
Date: Sun, 21 Oct 2001 15:17:08 +0200

This is basically the same thing as the "shadow" issue you have been
fixing up recently.

As Squid has it's own radix implementation it needs to be named uniquely
to not collide with various system provided implementations, or coded to
use the system provided implementations if available AND compatible with
what Squid expects.

The rn_lookup function is not covered by any standard, and may thus
differ slightly from implementation to implementation.

What can be said for sure is that having the prototype in asn.c is plain
bad. It should be in a header file "squid/include/radix.h". This is
fixed in HEAD since some time back by Duane, but my concern about
shadowing still applies to HEAD as it uses the system definitions src,
but provides a custom implementation in lib/radix.c using it's own
definitions.

Regards
Henrik

Adrian Chadd wrote:

> > > what was the "different prototype" ?
> >
> > struct radix_node *rn_lookup(void *v_arg, void *m_arg,struct
> > radix_node_head *head)
> >
> > same as freebsd it seems...not sure why the same error doesnt occur under
> > Free BSD. I should also have said the configure script doesn't pick up
> > MacOS X automatically (although some do) but I guess its just a matter of
> > creating configure with a newer version of autoconf (I configured for
> > powerpc-apple-rhapsody).
>
> hrm. Let me email cc squid-dev about this.
>
> squid-dev, any ideas? :)
>
> adrian
>
> --
> Adrian Chadd "Auntie Em, Hate you. Hate Kansas.
> <adrian@creative.net.au> Taking the dog."
> -- Dorothy
Received on Sun Oct 21 2001 - 07:17:24 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:14:34 MST