| From: | Chris Travers <chris(dot)travers(at)gmail(dot)com> | 
|---|---|
| To: | Eden Cardim <eden(at)insoli(dot)de> | 
| Cc: | pgsql-general(at)postgresql(dot)org | 
| Subject: | Re: Column aliases in WHERE clauses | 
| Date: | 2012-09-19 03:57:12 | 
| Message-ID: | CAKt_ZfvoNVVxyZTaHX0Zo0Q9kMQfuY_vWjt3dTraTNckTYj5YA@mail.gmail.com | 
| Views: | Whole Thread | Raw Message | Download mbox | Resend email | 
| Thread: | |
| Lists: | pgsql-general | 
On Tue, Sep 18, 2012 at 2:53 AM, Eden Cardim <eden(at)insoli(dot)de> wrote:
> >>>>> "Craig" == Craig Ringer <ringerc(at)ringerc(dot)id(dot)au> writes:
>
>     Craig> I just wish they hadn't written it backwards!
>
>     Craig> It'd be much less confusing were it formulated as something
>     Craig> like:
>
>     Craig> SELECT FROM thetable WHERE first_letter > 'a' RESULTS
>     Craig> left(value,1) AS first_letter
>
>     Craig> or something, where the order is more obvious. I really
>     Craig> dislike the way SQL is written not-quite-backwards.
>
> It's not "written backwards", it's plain natural language semantics:
> "give me the first letter of all records where the first letter is
> greater than a". Refining a set is better done walking from the more
> general set to a subset, not the other way around, IMO: "give me all
> persons that are females and over the age of 20". Mathematical set
> builder notation does this in a similar fashion, for the same reason.
>
>
Natural language semantics will get you into trouble though.  After all, I
think Lisp follows natural language semantics remarkably closely if your
natural language is Irish Gaelic....
Best Wishes,
Chris Travers
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Edson Richter | 2012-09-19 04:35:43 | Re: Change key primary for key foreign | 
| Previous Message | Eden Cardim | 2012-09-19 01:49:51 | Re: Column aliases in WHERE clauses |