W3C home > Mailing lists > Public > www-ql@w3.org > January to March 2003

RE: XQuery and relational databases...

From: Paul Cotton <pcotton@microsoft.com>
Date: Fri, 28 Feb 2003 14:32:35 -0500
Message-ID: <E7AC4500EAB7A442ABA7521D1881439704CCBDA3@tor-msg-01.northamerica.corp.microsoft.com>
To: "Michael Burbidge" <mburbidg@adobe.com>, <jonathan.robie@datadirect-technologies.com>, "Jerome Simeon" <simeon@research.bell-labs.com>
Cc: <www-ql@w3.org>

An interesting presentation on this topic was made by Jonathan Robie and
Jerome Simeon at the Extreme Markup 2002 conference in August 2002 [1].

" XML queries on relational sources: A syntactic approach 
Jonathan Robie, DataDirect Technologies; and Jerome Simeon, Bell Labs,
Lucent Technologies 

Many applications need to access both XML and relational data. This is
often done by representing relational data as XML or by using an XML
view of relational data. The ISO SQL/XML specification provides a
mapping from relational data to XML. Given that, XQuery can be useful
both for XML and XML views of relational sources, but mapping XQuery to
SQL is complicated by the lack of nested structures in the relational
model and the difference in expressive power between XQuery and SQL. We
present a subset of XQuery that is equivalent in power to SQL/XML,
preliminary mappings that reduce XQuery to this subset, and preliminary
mappings from the subset to SQL/XML. This syntax-centric approach is
simple, effective, and independent of any underlying physical algebra." 

Maybe Jonathan and/or Jerome could let us know if their presentation is
publically available?


[1] http://www.extrememarkup.com/extreme/2002/friday.asp 

Paul Cotton, Microsoft Canada 
17 Eleanor Drive, Nepean, Ontario K2E 6A3 
Tel: (613) 225-5445 Fax: (425) 936-7329 


> -----Original Message-----
> From: Michael Burbidge [mailto:mburbidg@adobe.com]
> Sent: February 28, 2003 10:56 AM
> To: www-ql@w3.org
> Subject: XQuery and relational databases...
> The XQuery specifications alludes to the fact that XQuery was designed
> in such a way as to be able express queries across various kinds of
> data, including relational databases.
> Does this mean that the FLWOR expression was designed to support
> queries that can be performed on relational databases? Was it designed
> to be roughly equivalent to SQL? Given that a relational database has
> some mechanism for exposing XML views on relational tables, is it
> relatively straight forward to convert an FLWOR expressions into SQL?
> Do we expect that relational database vendors will implement XQuery
> processors that directly integrate with their databases in an
> manner?
> Thanks,
> Michael-
Received on Friday, 28 February 2003 14:33:08 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:17:16 UTC