This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.

Bug 9540 - QueryTokenInfo definition seems wrong
Summary: QueryTokenInfo definition seems wrong
Status: CLOSED FIXED
Alias: None
Product: XPath / XQuery / XSLT
Classification: Unclassified
Component: Full Text 1.0 (show other bugs)
Version: Proposed Recommendation
Hardware: All All
: P2 normal
Target Milestone: ---
Assignee: Jim Melton
QA Contact: Mailing list for public feedback on specs from XSL and XML Query WGs
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-04-16 15:51 UTC by Paul J. Lucas
Modified: 2011-01-06 01:28 UTC (History)
2 users (show)

See Also:


Attachments

Description Paul J. Lucas 2010-04-16 15:51:13 UTC
From section 4.1.2:

[Definition: A QueryTokenInfo is the identity of a token inside a query string. ] Each QueryTokenInfo is associated with a starting and ending position that captures the relative position of the query string in the query.

However, in section 4.2.1.3, the schema for queryTokenInfo has only queryPos, a single integer, and not "a starting and ending position," two integers, per the definition above.
Comment 1 Mary Holstege 2010-05-25 15:53:55 UTC
In investigating this as directed by the WG, it appears that neither the queryPos attribute nor the described the starting and ending positions in the query are used directly in any of the semantic functions.  However, a single queryPos is carried through the semantic functions (and the accompanying diagrams).  I therefore propose to resolve the reported inconsistency by changing the definition of QueryTokenInfo.
Comment 2 Michael Dyck 2010-05-25 16:11:54 UTC
queryPos is used by fts:ApplyFTOrder().
Comment 3 Mary Holstege 2010-05-25 19:37:31 UTC
The queryPos in ApplyFTOrder is the stringInclude queryPos, which comes from
the parameter passed through the call stack.  Regardless, there is no call for
startPos/endPos on a QueryTokenInfo, so the proposal stands.
Comment 4 Michael Dyck 2010-06-08 02:32:17 UTC
(In reply to comment #3)
> The queryPos in ApplyFTOrder is the stringInclude queryPos, which comes from
> the parameter passed through the call stack.

My mistake. In fact, neither of queryTokenInfo's attributes is referenced by any of the functions in section 4.
Comment 5 Mary Holstege 2010-06-15 15:01:29 UTC
Fixed definition to refer to only one position.

Please indicate your satisfaction with this resolution by closing the bug.
Comment 6 Paul J. Lucas 2010-12-29 16:57:00 UTC
I'd be happy to mark this as CLOSED but I can't see the updated spec from the "outside".
Comment 7 Paul J. Lucas 2011-01-06 01:28:07 UTC
Verified as fixed.