proposed patch for function parameters name refs

Поиск
Список
Период
Сортировка
От Gevik Babakhani
Тема proposed patch for function parameters name refs
Дата
Msg-id 003c01c830f0$7e35f9d0$0a01a8c0@gevmus
обсуждение исходный текст
Ответы Re: proposed patch for function parameters name refs  (Bruce Momjian <bruce@momjian.us>)
Список pgsql-patches
This patch implements the function parameter referencing by name as
discussed before.
Please find the version 1.0 of this patch at http://www.postgresql.nl/gevik/

The zip archive contains a combined and separate patches to make reviewing
easier.

How this patch is implemented:
As discussed before, a generic parser callback functionality has been
implemented
to create hooks from within the parser. The mechanism of handling refnames
acts only
when the parser is in context of parsing a function AND all of the previous
attempts to
recognize an unknown refname (also after implicit RTE) has failed. This way
the match
for a possible function parameter name is executed as a last measure.

notes:

- a new error code has been added: ERRCODE_UNDEFINED_FUNCTION_PARAMETER_NAME
- when a given refname is unrecognized after all matches, one of the two
following
error messages is shown:

a) "there is no parameter %s defined for function %s"
b) "there is no parameter %s.%s defined for function %s or this \n is
    an invalid reference to FROM-clause entry for table \"%s\""

- 2 new files have been added parse_callback.h and .c
- the patch also contains test for regression
- the main logic is implemented in transformColumnRef/ case 1 and case 2
- patch is created using VC++ 2005, tested on Win32 and RH4


------------------------------------------------
Gevik Babakhani

PostgreSQL NL       http://www.postgresql.nl
TrueSoftware BV     http://www.truesoftware.nl
------------------------------------------------


В списке pgsql-patches по дате отправления:

Предыдущее
От: Mathias Hasselmann
Дата:
Сообщение: Avahi support for Postgresql
Следующее
От: Jorgen Austvik - Sun Norway
Дата:
Сообщение: Re: pg_regress: stat correct paths