| From: | Peter Eisentraut <peter_e(at)gmx(dot)net> |
|---|---|
| To: | pgsql-hackers(at)postgresql(dot)org |
| Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, "- -" <crossroads0000(at)googlemail(dot)com> |
| Subject: | Re: Unicode support |
| Date: | 2009-04-20 06:29:45 |
| Message-ID: | 200904200929.45214.peter_e@gmx.net |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Sunday 19 April 2009 18:54:45 Tom Lane wrote:
> Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> > On Monday 13 April 2009 20:18:31 - - wrote:
> >> 1) Functions like char_length() or length() do NOT return the number
> >> of characters (the manual says they do), instead they return the
> >> number of code points.
> >
> > I have added a Todo item about possibly fixing this.
>
> I thought the conclusion of the thread was that this wasn't wrong?
The only consensus I saw was that the normal form of an existing Unicode
string shouldn't be altered by PostgreSQL. That's pretty clear.
However, no one was entirely clear on the matter of how combining characters
are supposed to be processed. And even if we think that the current
interfaces give the right answer, there should possibly be other interfaces
that give the other right answer. It needs more research first of all.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | KaiGai Kohei | 2009-04-20 06:43:27 | Re: [PATCH] unalias of ACL_SELECT_FOR_UPDATE |
| Previous Message | Peter Eisentraut | 2009-04-20 06:24:34 | Re: Patch for 8.5, transformationHook |