From: | Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com> |
---|---|
To: | Robert Haas <robertmhaas(at)gmail(dot)com> |
Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: info about patch: using parametrised query in psql |
Date: | 2009-12-25 19:09:36 |
Message-ID: | 162867790912251109i700a952ei9618fc26909e86d2@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
2009/12/25 Robert Haas <robertmhaas(at)gmail(dot)com>:
> On Fri, Dec 25, 2009 at 1:41 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com> writes:
>>> there are two quoting styles, so we need two syntax. I proposed
>>
>>> :[var] and :{var} - for ident quoting and literal quoting.
>>> Theoretically we could to use :(var) for bytea escaping.
>>
>> And if you need a fourth style, you're at a dead end. I don't think
>> this is really an improvement over the single-flag-character approach.
>> Neither one has got any mnemonic value whatever, unfortunately, but
>> at least the flag character method is fairly extensible.
>
> The lack of mnemonic value kind of sucks, but I don't see that Pavel's
> style is any more or less extensible than your proposed flag
> character. Basically, he's saying that the flag characters will be [
> and { and adding a closing delimeter to match. If we do want to go
> with a single flag character, maybe it should just be a single or
> double quote:
>
> :'foo - quote as a literal
> :"foo - quote as an ident
I could to live with
:'foo' and :"foo" although ' and " characters are not the best for
readability. But the mnemonic is clear.
Pavel
>
> I dunno what to do about bytea-escaping under this framework, though.
>
> ...Robert
>
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2009-12-25 19:12:43 | Re: info about patch: using parametrised query in psql |
Previous Message | Tom Lane | 2009-12-25 19:08:26 | Re: ORDER BY clause in aggregate doesn't work well with multi argument aggregates |