Re: Revised patches to add table function support to PL/Tcl (TODO item)
Robert Haas <robertmhaas@gmail.com>
From: Robert Haas <robertmhaas@gmail.com>
To: Karl Lehenbauer <karllehenbauer@gmail.com>
Cc: pgsql-hackers <pgsql-hackers@postgresql.org>, Alvaro Herrera <alvherre@commandprompt.com>, Tom Lane <tgl@sss.pgh.pa.us>
Date: 2011-02-08T04:30:49Z
Lists: pgsql-hackers
On Tue, Dec 28, 2010 at 9:23 PM, Karl Lehenbauer <karllehenbauer@gmail.com> wrote: > On Dec 28, 2010, at 7:29 PM, Tom Lane wrote: >> This patch appears to be changing a whole lot of stuff that in fact >> pg_indent has never changed, so there's something wrong with the way you >> are doing it. It looks like a bad typedef list from here. > > You were right, Tom. The problem was that typedefs "pltcl_interp_desc", "pltcl_proc_key", and "pltcl_proc_ptr" weren't in src/tools/pgindent/typedefs.list. After adding them (and building and installing the netbsd-based, patched indent), pgindent only changes a handful of lines. > > pltcl-karl-try3-1-of-3-pgindent.patch patches typedefs.list with the three missing typedefs and pltcl.c with the small changes made by pgindent (it shifted some embedded comments left within their lines, mainly). > > As before, but "try3" now, pltcl-karl-try3-2-of-3-objects.patch converts pltcl.c to use the "Tcl objects" C API. > > And as before, but "try3" now, pltcl-karl-try3-3-of-3-setof.patch adds returning record and SETOF record. This patch did not get reviewed, because the person who originally planned to review it had a hardware failure that prevented him from doing so. Can anyone pick this up? -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company