Received: from sss.sss.pgh.pa.us (sss.pgh.pa.us [206.210.65.6]) by hub.org (8.9.3/8.9.3) with ESMTP id SAA60574; Fri, 14 May 1999 18:01:08 -0400 (EDT) (envelope-from tgl@sss.pgh.pa.us) Received: from sss.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.sss.pgh.pa.us (8.9.1/8.9.1) with ESMTP id RAA16290; Fri, 14 May 1999 17:59:52 -0400 (EDT) To: Oleg Bartunov cc: pgsql-hackers@postgreSQL.org, pgsql-interfaces@postgreSQL.org Subject: Re: [HACKERS] 6.5 cvs: problem with includes in src/interfaces/libpq++/ In-reply-to: Your message of Sat, 15 May 1999 00:37:26 +0400 (MSD) Date: Fri, 14 May 1999 17:59:51 -0400 Message-ID: <16288.926719191@sss.pgh.pa.us> From: Tom Lane Oleg Bartunov writes: > To compile postgres using gcc 2.7.2.1 I had to modify 2 files > src/interfaces/libpq++/pgconnection.cc > src/interfaces/libpq++/pgenv.h > Particularly, > #include to #include > #include to #include I am seeing the same thing here with gcc 2.7.2.2. We need to adopt a considered policy about whether libpq++ will still support gcc 2.7.*, not just break it without thinking. I'd vote for still supporting 2.7.*, but I know that the C++ library shipped with this gcc release is not real up-to-date. It may not be practical to support both latest-C++-spec compilers and the older generation; I'm not sure what the issues are. If the conclusion is "no", then the configure script ought to be changed to not try to build libpq++ unless up-to-date libraries are available. regards, tom lane