W3C home > Mailing lists > Public > public-qt-comments@w3.org > April 2008

[Bug 5459] [FS] Static type analysis for the fn:abs, fn:ceiling, fn:floor, fn:round, and fn:round-half-to-even functions

From: <bugzilla@farnsworth.w3.org>
Date: Tue, 01 Apr 2008 10:50:12 +0000
To: public-qt-comments@w3.org
Message-Id: <E1Jge4C-0004Ms-Ar@farnsworth.w3.org>

------- Comment #9 from oliver@cbcl.co.uk  2008-04-01 10:50 -------
> The only suggestion I could find that there even *are* multiple "versions" of
> fn:abs is in F+O section 1.4, where it says that a signature involving
> 'numeric' represents four signatures. But the XQuery and FS specs are fairly
> clear that fn:abs has only one signature and one implementation.

I had presumed from F&O 1.4 that there are four separate fn:abs functions (or
at least four separate signatures that are chosen from at compile time).  Which
parts of the specification imply otherwise?

If it is the case that fn:abs has four signatures then the rules given in FS
4.1.5 under "Dynamic Evaluation" deduces four different results for an argument
of type xs:integer (since it can be promoted to match any signature).

Should this be marked as a bug against Functions & Operators?
Received on Friday, 4 April 2008 10:00:15 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:57:22 UTC