Re: [xml-dev] XQueryX - re: was : re : was .... too many threads

[Jim Melton <jim.melton@oracle.com>]

David,

At 4/15/2010 05:31 AM, David wrote:
>Rewriting an XQueryX converer  in XQueryX doesnt seem particularly 
>"helpful" until there are native XQueryX parsers.
>But an XQuery implementation would have the right spirit ...
>Then an XQueryX implementation would be a good exercise.

Actually, I'm aware of at least one (and, I think, two) XQuery 
implementations whose *native syntax* is XQueryX.  It (or they) parse 
XQuery into XQueryX, then evaluate the XQueryX.  The implementation 
of which I am certain is a server-based implementation for which high 
throughput and high performance are prime goals; the other 
implementation of which I'm less certain is a middleware-based implementation.

Neither implementation pretends that application programmers write 
very much XQueryX directly, but both are based on the premise that 
machine-generated queries and, possibly more relevantly, 
machine-modified (e.g., edited, enhanced) queries are much more 
easily done in the XML syntax than in the "english-like" syntax.

As the editor of the XQueryX spec, and thus the person responsible 
for modifying the XQueryX XML Schema and XSLT stylesheet when new 
features are added to XQuery, I have to write XQueryX queries by 
hand.  It's not terribly difficult if one has a schema-driven 
authoring tool (I happen to use Stylus Studio, but I have no business 
relationship with them other than their kind grant of a full license 
to me for use in developing standards).  But it's also not as 
intuitive as writing the "english-like" syntax of XQuery itself.  If 
I were writing a lot of ad hoc queries, or a lot of queries embedded 
in applications, I'm sure I'd use the more human-friendly syntax than 
the XML syntax most of the time.  (On the other hand, how many people 
write XSLT by hand?  Quite a few, I believe.  So, clearly, writing 
programs or queries in XML syntax is not all that unreasonable.)

Hope this helps,
    Jim



>Any reverse parsers ?  XQuery -> XQueryX ?
>
>I'm thinking of doing the same for xmlsh (making an XML 
>representation) but havent put in the work.
>Several people have suggested it would be "a good idea" but noone's 
>suggested the'd actually use it.
>
>Any ideas on the actual user base for XQueryX ? I'm curious if 
>anyone given the choice would prefer writing in it over XQuery.
>Presumably (?) its mainly for machine generation/consumption.
>This is my main reason for considering an xmlshX version.  My early 
>attempts at writing XProc to xmlsh I stopped, not because xmlsh couldnt
>do everything xproc can (it can and much more) but because getting 
>the syntax right in a machine generated transformation was tricky.
>That and the static and dynamic errors from XProc are tricky to 
>emulate exactly.  I did have about a 90% coverage of the "success" 
>test suite at one point.
>The error test suite was tough though (and covers about 80% of the tests).
>
>So I consider an xmlshX syntax might make the code generation 
>easier... but I shudder to think of anyone humanly writing in the 
>XML format ...
>
>Is that the experience of XQueryX ? Or do some people prefer the XML 
>syntax for human authorship ?
>
>
>-------------------------
>David A. Lee
>dlee@calldei.com
>http://www.calldei.com
>http://www.xmlsh.org
>
>
>On 4/15/2010 7:19 AM, Michael Kay wrote:
>>There is a published XSLT stylesheet* that converts XQueryX to XQuery. So
>>you just do the conversion, and then run the query using a product such as
>>Saxon.
>
>========================================================================
>Jim Melton --- Editor of ISO/IEC 9075-* (SQL)     Phone: +1.801.942.0144
>   Chair, W3C XML Query WG; XQX (etc.) editor       Fax : +1.801.942.3345
>Oracle Corporation        Oracle Email: jim dot melton at oracle dot com
>1930 Viscounti Drive      Standards email: jim dot melton at acm dot org
>Sandy, UT 84093-1063 USA          Personal email: jim at melton dot name
>========================================================================
>=  Facts are facts.   But any opinions expressed are the opinions      =
>=  only of myself and may or may not reflect the opinions of anybody   =
>=  else with whom I may or may not have discussed the issues at hand.  =
>========================================================================