| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Susanne Ebrecht <susanne(at)2ndQuadrant(dot)com> |
| Cc: | Andreas Pflug <pgadmin(at)pse-consulting(dot)de>, Thomas Kellerer <spam_eater(at)gmx(dot)net>, pgsql-bugs(at)postgresql(dot)org |
| Subject: | Re: char(0) |
| Date: | 2011-10-18 17:17:15 |
| Message-ID: | 7412.1318958235@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
Susanne Ebrecht <susanne(at)2ndQuadrant(dot)com> writes:
> On 17.10.2011 16:41, Andreas Pflug wrote:
>> This is a little bit annoying on migration topics.
>> While not move on to a cleaner approach during the migration and use a
>> "boolean not null"?
>> Sounds much too straight forward, not mysql-ish artistic enough...
> Depends if you want / are able to touch the application source code or not.
If you're expecting to move a mysql application to postgres with zero
source code changes, you're living in a fantasy world anyway ... but
this difference is hardly likely to be your worst problem.
AFAICT the SQL standard is perfectly clear on this. *Values* of type
varchar can be of zero length, but that does not mean that you can
*declare* a column to be varchar(0), and that NOTE says specifically
that you can't.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2011-10-18 18:22:20 | Re: incompatible pointer type |
| Previous Message | Kevin Grittner | 2011-10-18 16:44:41 | Re: Tablespace files deleted in continuous run |