Re: ACTION A-613-26: Norm Walsh to replace the text in XDM 3.1 section...

Norm,

I think the following changes are needed:

(1) Where it currently says "Each key in the map is unique (there is no other key to which it is equal) and has associated with it a value that is a single item or sequence of items.”  We need to say somehow that keys are compared using op:same-key:
https://www.w3.org/XML/Group/qtspecs/specifications/xpath-functions-31/html/Overview-diff.html#func-same-key

(2) Remove this paragraph:  "If any date/time values (xs:date, xs:dateTime, xs:time, xs:gYear, etc.) are used as keys, all such values must either have a timezone or must not have a timezone. An error is raised if an attempt is made to mix date/time key values that have a timezone with values that do not. This constraint applies only to map keys and applies independently to each map (it is not an error to have one map that uses date/time keys with timezones and another which uses values without timezones).”

Thanks,
Josh



> On Oct 6, 2015, at 7:10 AM, Norman Walsh <norman.walsh@marklogic.com> wrote:
> 
> Andrew Coleman <andrew_coleman@uk.ibm.com> writes:
>> ACTION A-613-26: Norm Walsh to replace the text in XDM 3.1 section 
>> 17.1 by implementing the changes from comment 6 of bug 28795. 
>> https://www.w3.org/Bugs/Public/show_bug.cgi?id=28795#c6 
>> 
>> Status: Pending; ETA 2015-09-29 
> 
> Sorry. Perhaps I'm being especially thick. Several weeks ago, I
> proposed that this was an action on F&O and not on the data model.
> Someone else proposed that I was mistaken, that based on comment 6,
> there were data model changes required.
> 
> I agreed to take another look.
> 
> I've taken another look and, as I said, perhaps because I'm a moron,
> I'm not seeing how the remarks in comment 6 apply to the data model.
> The data model says there are maps with unique keys and provides a
> single accessor that returns all of them. It says nothing about how
> keys are compared (though perhaps it should link to the relevant part
> of F&O). The comment about keys with dateTimes with and without
> timezones is still present, but that doesn't immediately seem to be
> problematic.
> 
> Someone feel free to club me in the back of the head with a
> clue-by-four.
> 
>                                        Be seeing you,
>                                          norm
> 
> -- 
> Norman Walsh
> Lead Engineer
> MarkLogic Corporation
> Phone: +1 512 761 6676
> www.marklogic.com

Received on Tuesday, 6 October 2015 15:43:15 UTC