Обсуждение: pgsql-server/src/backend/nodes read.c
CVSROOT: /cvsroot Module name: pgsql-server Changes by: tgl@svr1.postgresql.org 04/01/07 17:12:56 Modified files: src/backend/nodes: read.c Log message: Add missing inclusion of nodes/value.h.
tgl@svr1.postgresql.org (Tom Lane) writes: > Add missing inclusion of nodes/value.h. Strange, on my machine this wasn't necessary (at least, 'make clean && make -s' completed without warnings or errors). Sorry about that. -Neil
Neil Conway <neilc@samurai.com> writes: > tgl@svr1.postgresql.org (Tom Lane) writes: >> Add missing inclusion of nodes/value.h. > Strange, on my machine this wasn't necessary (at least, 'make clean && > make -s' completed without warnings or errors). [ scratches head... ] I don't see why we'd get different results. I see four warnings about undeclared functions in read.c if the inclusion isn't there. Is there perhaps some change in your sources that you missed committing? regards, tom lane
Tom Lane wrote: > Neil Conway <neilc@samurai.com> writes: > > tgl@svr1.postgresql.org (Tom Lane) writes: > >> Add missing inclusion of nodes/value.h. > > > Strange, on my machine this wasn't necessary (at least, 'make clean && > > make -s' completed without warnings or errors). > > [ scratches head... ] I don't see why we'd get different results. > I see four warnings about undeclared functions in read.c if the inclusion > isn't there. Is there perhaps some change in your sources that you > missed committing? I saw the same. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania 19073
Tom Lane <tgl@sss.pgh.pa.us> writes: > [ scratches head... ] I don't see why we'd get different results. > I see four warnings about undeclared functions in read.c if the > inclusion isn't there. Is there perhaps some change in your sources > that you missed committing? Bizarre -- upon reversing your change to read.c and recompiling, I now get the warnings. I could have sworn that I'd double-checked that I'd fixed the necessary dependencies for the value node change, but apparently not. Mea culpa, and sorry for the sloppiness. -Neil