[Q] sequence and index name limits in 8.4

Started by V S Palmost 17 years ago3 messagesgeneral
Jump to latest
#1V S P
toreason@fastmail.fm

Searched docs and the web but only saw references to the older version
of postgres
where changing source code was recommended ( NAMEDATALEN) if more than
31 characters
in the names are desired.

wanted to ask if this is still the case in 8.4 (I need at least 128)

thank you in advance,
VSP
--
Vlad P
author of C++ ORM http://github.com/vladp/CppOrm/tree/master

--
http://www.fastmail.fm - Send your email first class

#2Scott Marlowe
scott.marlowe@gmail.com
In reply to: V S P (#1)
Re: [Q] sequence and index name limits in 8.4

On Sat, Jun 27, 2009 at 7:09 PM, V S P<toreason@fastmail.fm> wrote:

Searched docs and the web but only saw references to the older version
of postgres
where changing source code was recommended ( NAMEDATALEN) if more than
31 characters
in the names are desired.

wanted to ask if this is still the case in 8.4 (I need at least 128)

Unless you want to compile your own version, I think the curent limit
is 64 or 63 characters. There are real and measurable performance hits
as you go over that number.

#3Tom Lane
tgl@sss.pgh.pa.us
In reply to: Scott Marlowe (#2)
Re: [Q] sequence and index name limits in 8.4

Scott Marlowe <scott.marlowe@gmail.com> writes:

On Sat, Jun 27, 2009 at 7:09 PM, V S P<toreason@fastmail.fm> wrote:

Searched docs and the web but only saw references to the older version
of postgres
where changing source code was recommended ( NAMEDATALEN) if more than
31 characters
in the names are desired.

Unless you want to compile your own version, I think the curent limit
is 64 or 63 characters. There are real and measurable performance hits
as you go over that number.

Also, it hasn't been 32 by default since 7.2, more than seven years ago.
Whatever references you are reading are so out of date as to be
completely untrustworthy ...

regards, tom lane