[HACKERS] XLogDir

2002-08-16 Thread Bruce Momjian
A recent change made XLogDir dynamically sized. There was a question of whether this was a good idea since there are lots of other places that use MAXPGPATH. I have now found that pg_resetxlog doesn't compile anymore because XLogDir is a char* now instead of a char[]. Should I fix pg_resetxlog

Re: [HACKERS] XLogDir

2002-08-16 Thread Tom Lane
Bruce Momjian <[EMAIL PROTECTED]> writes: > A recent change made XLogDir dynamically sized. There was a question of > whether this was a good idea since there are lots of other places that > use MAXPGPATH. > I have now found that pg_resetxlog doesn't compile anymore because > XLogDir is a char*

Re: [HACKERS] XLogDir

2002-08-16 Thread Bruce Momjian
Tom Lane wrote: > Bruce Momjian <[EMAIL PROTECTED]> writes: > > A recent change made XLogDir dynamically sized. There was a question of > > whether this was a good idea since there are lots of other places that > > use MAXPGPATH. > > > I have now found that pg_resetxlog doesn't compile anymore b

Re: [HACKERS] XLogDir

2002-08-19 Thread Thomas Lockhart
> > Revert. The XLogDir change was incomplete and basically useless to > > start with ... > Yea, but it was tied into the PGXLOG commit. Thomas, what are we doing > with that? Why ask me? - Thomas ---(end of broadcast)--- TIP 4

Re: [HACKERS] XLogDir

2002-08-19 Thread Bruce Momjian
Thomas Lockhart wrote: > > > Revert. The XLogDir change was incomplete and basically useless to > > > start with ... > > Yea, but it was tied into the PGXLOG commit. Thomas, what are we doing > > with that? > > Why ask me? Sorry, I mean XLOGDIR. Somehow I kept saying PGXLOG. Anyway, it is al