From: Andreas Scherer (as_at_[hidden])
Date: 2000-10-03 13:30:12
--- In boost_at_[hidden], jsiek_at_l... wrote:
> Would you recommend that we add a #ifdef and an include of
> gb_graph.h for the case when the installation of SGI has the
> PROTOTYPES patches installed?
Only if my assumption is correct that the PROTOTYPES really are
sufficient for <gb_graph.h> to be #included safely in
<stanford_graph.hpp> in an "extern "C"" group. Moreover, not every
user of SGB is aware of the possibility to "ln -s PROTOTYPES/* ."
before running "make tests" and "make install" (nor is he/she
necessarily in a position to run these steps himself/herself).
Unfortunately, all I have here is a Pentium-I PC at 133 MHz and 32 MB
of RAM, and it takes _a_long_time_ to compile the "graph" stuff, so I
am not (yet) in a position to test a modified <stanford_graph.hpp>.
If possible, I'll report on this tomorrow, when I have access to a
P-II at 350 MHz with sufficent RAM.
However, I could imagine that a patched <gb_graph.h> headerfile could
well be used in <stanford_graph.hpp>, thus eliminating the SGB stuff
for "updated" setups of SGB.
Today I submitted a report to Don Knuth regarding the Boost Graph
Library and the compatibility issue raised by <gb_graph.h>. However,
I doubt that DEK would apply extensive changes to sGB per se.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk