Re: information_schema.columns changes needed for OLEDB
От | Konstantin Izmailov |
---|---|
Тема | Re: information_schema.columns changes needed for OLEDB |
Дата | |
Msg-id | 72746b5e0905291514r47d15f62pbb32c9c2cac55cfd@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: information_schema.columns changes needed for OLEDB (Tom Lane <tgl@sss.pgh.pa.us>) |
Список | pgsql-hackers |
Tom,
this is very helpful - thank you so much!
I had to discover those 'missing' functions one by one, usually after users' complaints.
Konstantin
On Fri, May 29, 2009 at 11:35 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
Konstantin Izmailov <pgfizm@gmail.com> writes:The ones that appear not to be there are ones that the SQL standard
> you know that some Postgres functions are listed in pg_proc while others are
> not. For example, all Data Type Formatting function are in pg_proc (to_char,
> to_hex, ...). While several of the Date/Time Functions are not there
> (extract, localtime, ...).
demands special weird syntax for. The grammar translates such calls
to standard function calls to underlying functions, which usually are
named a bit differently to avoid confusion. For instance
extract(field from some_expr) becomes date_part('field', some_expr).
If you want to know what all of these are, see the func_expr production
in parser/gram.y.Complain to the SQL standards committee, especially to those members
> This causes issues to Windows integration as well.
who seem to think COBOL represented the apex of programming language
syntax design :-(
regards, tom lane
В списке pgsql-hackers по дате отправления: