Revised patches to add table function support to PL/Tcl (TODO item)
От | Karl Lehenbauer |
---|---|
Тема | Revised patches to add table function support to PL/Tcl (TODO item) |
Дата | |
Msg-id | 2A558223-09B0-402E-9304-E1EF77152D02@gmail.com обсуждение исходный текст |
Ответ на | Re: Patch to add table function support to PL/Tcl (Todo item) (Alvaro Herrera <alvherre@commandprompt.com>) |
Ответы |
Re: Revised patches to add table function support to PL/Tcl (TODO item)
|
Список | pgsql-hackers |
In response to Alvaro Herrera's message from today I've split the PL/Tcl table function patch into three separate, easier-to-digestpatches. (Thanks for the quick response, Alvaro.) The first patch, pltcl-karl-try2-1-of-3-pgindent.patch, does nothing but conform HEAD's pltcl.c with pgindent. Applyingthis patch should have exactly the same effect as running src/tools/pgindent/pgindent src/tools/pgindent/typedefs.list src/pl/tcl/pltcl.c The second patch, pltcl-karl-try2-2-of-3-objects.patch, should be applied after the first, and updates PL/Tcl to use theTcl "Tcl object" C API, the preferred way of interacting with Tcl from C since Tcl 8.0 was released in 1997. The third patch, pltcl-karl-try2-3-of-3-setof.patch, builds on the above to add both the "return_next" command for returningmultiple rows in a SETOF-returning function and to add using "return" with a list of key-value pairs for functionsreturning a non-SETOF record. On Dec 28, 2010, at 12:12 PM, Alvaro Herrera wrote: > Excerpts from Karl Lehenbauer's message of mar dic 28 12:33:42 -0300 2010: >> Project name: Add table function support to PL/Tcl (Todo item) >> >> What the patch does: >> >> This patch adds table function support (returning record and SETOF record) >> to PL/Tcl. This patch also updates PL/Tcl to use the Tcl object-style >> interface instead of the older string-style one, increasing performance. > > While I don't use PL/Tcl myself, this seems a reasonable idea. However, > I think this patch does too many things in one step. It also contains > numerous superfluous whitespace changes that make it hard to assess its > real size. > > I'd recommend splitting it up and dropping the whitespace changes (which > would be reverted by pgindent anyway). > > -- > Álvaro Herrera <alvherre@commandprompt.com> > The PostgreSQL Company - Command Prompt, Inc. > PostgreSQL Replication, Consulting, Custom Development, 24x7 support
Вложения
В списке pgsql-hackers по дате отправления: