RE: DASL Where statement.

In order to turn a prefix tree into infix, all you have to do is go
recursively down the first operand, then emit the operator, then go
recursively down the second operand, emitting parentheses around the
operands. So it's very easy to generate SQL from the operators as currently
specified. Mixing prefix and infix would make generating SQL more
complicated. That would be a bad idea. 

The goal is to make the programmatic processing easy. Making the protocol
more readable according to someone's sense of style of readability is a non
goal. Attempts to do that has led to a lot of problems in XML protocols, for
example in regards to newlines and whitespace in strings. 

Alan

-----Original Message-----
From: Sung Kim [mailto:hunkim@cse.ucsc.edu]
Sent: Wednesday, January 30, 2002 7:20 PM
To: Babich, Alan
Cc: www-webdav-dasl@w3.org
Subject: RE: DASL Where statement.


Although it is a characteristic of XML, the specified style looks
prefix and postfix. :-)

<d:and>          // prefix
   <d:A/>
   <d:B/>
</d:and>         // postfix

--
Sung Kim <hunkim@cse.ucsc.edu>
http://www.cse.ucsc.edu/~hunkim

 "Dreams become reality!"

On Wed, 30 Jan 2002, Babich, Alan wrote:

> Yes, you are correct in your e-mail below about prefix, postfix, and
infix.
> As currently specified, all the DASL operators are prefix. None are
postfix
> or infix.
>
> Alan
>
> -----Original Message-----
> From: Sung Kim [mailto:hunkim@cse.ucsc.edu]
> Sent: Wednesday, January 30, 2002 6:53 PM
> To: Jim Davis
> Cc: www-webdav-dasl@w3.org
> Subject: RE: DASL Where statement.
>
>
> I thought
>
> prefix : [+ a b]
> infix  : [a + b]
> postfix: [a b +]
>
> :-)
>
> --
> Sung Kim <hunkim@cse.ucsc.edu>
> http://www.cse.ucsc.edu/~hunkim
>
>  "Dreams become reality!"
>
> On Wed, 30 Jan 2002, Jim Davis wrote:
>
> > At 06:03 PM 1/30/2002 -0800, Sung Kim wrote:
> >
> > >In point of implementation view, currently specified prefix and postfix
> > >mixed style needs stack things to parse
> >
> > Sorry, but which operators are postfix?  I thought all the operators
were
> > prefix?
> >
> >
>
>
>

Received on Wednesday, 30 January 2002 23:17:01 UTC