| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> | 
|---|---|
| To: | Edmund Mergl <E(dot)Mergl(at)bawue(dot)de> | 
| Cc: | PostgreSQL Developers Mailinglist <pgsql-hackers(at)postgreSQL(dot)org> | 
| Subject: | Re: [HACKERS] new return value of PQfsize | 
| Date: | 1998-09-09 14:05:42 | 
| Message-ID: | 28265.905349942@sss.pgh.pa.us | 
| Views: | Whole Thread | Raw Message | Download mbox | Resend email | 
| Thread: | |
| Lists: | pgsql-hackers | 
Edmund Mergl <E(dot)Mergl(at)bawue(dot)de> writes:
> the libpq-function PQfsize() now returns int instead of
> short. It looks like, that this needs some adaption in the
> backend, because now the size of a variable length field is
> returned as 65535 instead of -1.
Oops. That's libpq's bug, not the backend's.
I thought that the rearrangement of libpq's interface
would be a good time to hide the fact that field sizes are
currently restricted to short --- which was visible in the
return type of PQfsize, and nowhere else in libpq's API.
But I guess I forgot to ensure that sign extension would happen
correctly :-(.  Will fix.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 1998-09-09 14:14:08 | Re: [HACKERS] Indixing problems... | 
| Previous Message | Edmund Mergl | 1998-09-09 13:07:05 | new return value of PQfsize |