| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | "Brett W(dot) McCoy" <bmccoy(at)chapelperilous(dot)net> |
| Cc: | Singer Wang <swang(at)cs(dot)dal(dot)ca>, pgsql-novice(at)postgresql(dot)org |
| Subject: | Re: varchar vs char vs text |
| Date: | 2002-02-12 21:45:13 |
| Message-ID: | 24880.1013550313@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-novice |
"Brett W. McCoy" <bmccoy(at)chapelperilous(dot)net> writes:
> I'd go with text. It's not SQL92, though.
Check.
> varchar is technically supposed to have a limit of 255,
Certainly not; the spec says
The maximum value of <length> is implementation-defined.
There may be implementations that are lame enough to limit it to 255,
but Postgres isn't one of them. IIRC, we set a rather arbitrary upper
limit of 10000000 on the length (mainly on the theory that anything
larger is either a typo, or you really don't want a limit at all, in
which case you oughta be using text).
At least in 7.2, it also works to say just "varchar" with no length
limit; this is functionally equivalent to "text" except perhaps for
some corner cases involving ambiguous-data-type resolution. But
this is not SQL-spec-compliant either.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Brett W. McCoy | 2002-02-12 22:44:41 | Re: varchar vs char vs text |
| Previous Message | Josh Berkus | 2002-02-12 21:43:10 | Re: "Hot Backups" |