Re: [HACKERS] recovery.conf parsing problems

2007-02-03 Thread Bruce Momjian
Added to TODO: o Allow recovery.conf to allow the same syntax as postgresql.conf, including quoting http://archives.postgresql.org/pgsql-hackers/2006-12/msg00497.php --- Simon Riggs wrot

Re: [HACKERS] recovery.conf parsing problems

2006-12-14 Thread Tom Lane
"Simon Riggs" <[EMAIL PROTECTED]> writes: > On Thu, 2006-12-14 at 13:52 +0100, Peter Eisentraut wrote: >> I'm actually not so sure that this is a good idea. The lexical >> structure should be exactly the same, and some things like include >> files might become useful as well, so why should all

Re: [HACKERS] recovery.conf parsing problems

2006-12-14 Thread Simon Riggs
On Thu, 2006-12-14 at 13:52 +0100, Peter Eisentraut wrote: > Simon Riggs wrote: > > > It would probably be far easier for long-term maintenance if you > > > just built an independent lexer, instead of trying to make > > > guc-file.l serve multiple masters. > > > > Will do. > > I'm actually not so

Re: [HACKERS] recovery.conf parsing problems

2006-12-14 Thread Peter Eisentraut
Simon Riggs wrote: > > It would probably be far easier for long-term maintenance if you > > just built an independent lexer, instead of trying to make > > guc-file.l serve multiple masters. > > Will do. I'm actually not so sure that this is a good idea. The lexical structure should be exactly th

Re: [HACKERS] recovery.conf parsing problems

2006-12-14 Thread Simon Riggs
On Wed, 2006-12-13 at 17:02 -0500, Tom Lane wrote: > "Simon Riggs" <[EMAIL PROTECTED]> writes: > > OK, I would propose to extend the guc-file.l to include sufficient code > > to allow the parsing of the conf files to be identical between the > > postgresql.conf and the recovery.conf (it isn't the s

Re: [HACKERS] recovery.conf parsing problems

2006-12-13 Thread Tom Lane
Andrew - Supernews <[EMAIL PROTECTED]> writes: > That only helps if you can trust %p not to contain whitespace or $. If it > is always relative to somewhere in the data dir then this is probably ok, > but if it's an absolute path then you can't assume that. It is relative, so I think this is actua

Re: [HACKERS] recovery.conf parsing problems

2006-12-13 Thread Andrew - Supernews
On 2006-12-13, "Simon Riggs" <[EMAIL PROTECTED]> wrote: > On Wed, 2006-12-13 at 19:28 +, Simon Riggs wrote: >> On Wed, 2006-12-13 at 04:23 +, Andrew - Supernews wrote: >> > While testing a PITR recovery, I discovered that recovery.conf doesn't >> > seem to allow specifying ' in the command

Re: [HACKERS] recovery.conf parsing problems

2006-12-13 Thread Tom Lane
"Simon Riggs" <[EMAIL PROTECTED]> writes: > OK, I would propose to extend the guc-file.l to include sufficient code > to allow the parsing of the conf files to be identical between the > postgresql.conf and the recovery.conf (it isn't the same yet). It would probably be far easier for long-term ma

Re: [HACKERS] recovery.conf parsing problems

2006-12-13 Thread Simon Riggs
On Wed, 2006-12-13 at 19:28 +, Simon Riggs wrote: > On Wed, 2006-12-13 at 04:23 +, Andrew - Supernews wrote: > > While testing a PITR recovery, I discovered that recovery.conf doesn't > > seem to allow specifying ' in the command string, making it hard to > > protect the restore_command aga

Re: [HACKERS] recovery.conf parsing problems

2006-12-13 Thread Simon Riggs
On Wed, 2006-12-13 at 04:23 +, Andrew - Supernews wrote: > While testing a PITR recovery, I discovered that recovery.conf doesn't > seem to allow specifying ' in the command string, making it hard to > protect the restore_command against problematic filenames (whitespace > etc.). This doesn't s

[HACKERS] recovery.conf parsing problems

2006-12-12 Thread Andrew - Supernews
While testing a PITR recovery, I discovered that recovery.conf doesn't seem to allow specifying ' in the command string, making it hard to protect the restore_command against problematic filenames (whitespace etc.). This doesn't seem to be a problem for archive_command, which allows \' (e.g. archiv