That hack also works outside cq. I've used it for a number of
applications, and just tested it with an HTTP server. The server is
perfectly happy to serialize a sequence of XML and non-XML items. You
can even serialize binary nodes in the same sequence, if you like.
$ cat test.xqy
(: test.xqy :)
'<?xml version="1.0"?>',
<test/>,
binary { xs:hexBinary('cafebabe') }
$ wget [...]/test.xqy | od -cx
HTTP/1.1 200 OK
Server: MarkLogic
Content-Type: text/plain; charset=UTF-8
Content-Length: 34
Connection: Keep-Alive
Keep-Alive: timeout=1, max=96
0000000 < ? x m l v e r s i o n = " 1
3f3c 6d78 206c 6576 7372 6f69 3d6e 3122
0000020 . 0 " ? > \n < t e s t / > \n 312 376
302e 3f22 0a3e 743c 7365 2f74 0a3e feca
0000040 272 276
beba
0000042
In passing, I'll add that the cq tool doesn't take much notice of which
button you press. Certainly that information isn't used for
serialization options. Primarily, the button pressed is used to set the
Content-Type header for browser display. What happens next is up to the
browser.
-- Mike
On 2010-08-31 08:34, Ron Hitchens wrote:
>
> Yeah, it's an ugly hack and only really works
> in the context of CQ. Forget I mentioned it. :-)
> You're better off adding the not-a-PI XML declaration
> to the serialized XML outside of MarkLogic/XQuery as
> needed.
>
> On Aug 31, 2010, at 4:17 PM, Florent Georges wrote:
>
>> Ron Hitchens wrote:
>>
>> Hi,
>>
>>> The XML declaration is a processing instruction
>>
>> Really, the XML declaration is not a PI. It looks like a PI,
>> it smells like a PI, but the XML recommendation says it is not a
>> PI, it is the XML declaration. That is important because you can
>> not expect to have it in XDM (for instance like a lot of people
>> want to access CDATA sections in an XDM instance).
>>
>>> let $prolog := '<?xml version="1.0" encoding="UTF-8"?>'
>>> let $newXML := element result { ...
>>> return ($prolog, $newXML)
>>
>> Well, if you ask you serializer to use the XML method, that
>> should result to&_lt;...?&_gt;<result> ... If you try it within
>> CQ, you get this result when you ask for the "XML", but strangely
>> enough you get '<?...?><result> ...' when asking for the "Text"
>> (but this is neither a proper text nor xml method, by my reading
>> of the serialization recommendation).
>>
>> Regards,
>>
>> --
>> Florent Georges
>> http://fgeorges.org/
>>
>>
>>
>>
>> _______________________________________________
>> General mailing list
>> [email protected]
>> http://developer.marklogic.com/mailman/listinfo/general
>
> ---
> Ron Hitchens {mailto:[email protected]} Ronsoft Technologies
> +44 7879 358 212 (voice) http://www.ronsoft.com
> +1 707 924 3878 (fax) Bit Twiddling At Its Finest
> "No amount of belief establishes any fact." -Unknown
>
>
>
>
> _______________________________________________
> General mailing list
> [email protected]
> http://developer.marklogic.com/mailman/listinfo/general
_______________________________________________
General mailing list
[email protected]
http://developer.marklogic.com/mailman/listinfo/general
