W3C home > Mailing lists > Public > www-xpath-comments@w3.org > January to March 2002

Re: Comparisons in XPath 2.0

From: Jeni Tennison <jeni@jenitennison.com>
Date: Tue, 8 Jan 2002 08:23:57 +0000
Message-ID: <2385597062.20020108082357@jenitennison.com>
To: www-xpath-comments@w3.org
CC: xsl-list@lists.mulberrytech.com
Mike wrote:
>> For all that there are lots of ways of comparing values in XPath 2.0,
>> there doesn't seem to be one that compares sequences.
>> 
>> If you have something like:
>> 
>>   <line start="0 300" end="300 0" />
>> 
>> where start and end both have typed values - @start is the sequence
>> of two integers (0, 300) and @end is the sequence of two integers
>> (300, 0). In this particular language, @start and @end are x,y
>> coordinates. How can you work out whether start and end have the
>> same value?
>
> I think you're right, this is an omission that we need to remedy.
> I'd suggest a function compare-sequences(sequence, sequence),
> perhaps with an optional collation argument, that returns -1, 0, or
> +1, in the same way as compare(). It should be based on pairwise
> comparison of items in the two sequences, each pair being compared
> using the eq and lt operators. Phil Wadler has referred to this as
> "lexicographic comparison", but I'm not sure most people would
> understand that term correctly.

There are two functions that do compare sequences:

  xf:sequence-deep-equal()
  xf:sequence-node-equal()

Following the same scheme, the one I'm after here would have the
signature:

  xf:sequence-value-equal(item* $parameter1, item* $parameter2)
    => boolean?

  xf:sequence-value-equal(item* $parameter1, item* $parameter2,
                          anyURI $collationLiteral)
    => boolean?

Cheers,

Jeni

---
Jeni Tennison
http://www.jenitennison.com/
Received on Tuesday, 8 January 2002 03:23:59 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 3 October 2007 16:05:54 GMT