Hi Frank,
Or how about if you do a language-sensitive search. For example:
let $doc :=
<a>
<b id="car">
<c xml:lang="en">car</c>
<c xml:lang="nl">auto</c>
<c xml:lang="fr">voiture</c>
</b>
<b id="house">
<c xml:lang="en">house</c>
<c xml:lang="nl">huis</c>
<c xml:lang="fr">maison</c>
</b>
<b id= "child">
<c xml:lang="en">child</c>
<c xml:lang="nl">kind</c>
<c xml:lang="fr">enfant</c>
</b>
</a>
return $doc//c[cts:contains(., cts:or-query((
cts:word-query("auto", "lang=nl"),
cts:word-query("kind", "lang=nl"))) )]
=>
<c xml:lang="nl">auto</c>
<c xml:lang="nl">kind</c>
-Danny
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Christopher Hamlin
Sent: Monday, December 08, 2014 10:14 AM
To: MarkLogic Developer Discussion
Subject: Re: [MarkLogic Dev General] Marklogic node replace of same node
results in inconsistent output
Hi Frank,
Not exactly sure if it's what you want, but maybe helps you get closer:
for $x in $doc/b/c[@xml:lang eq"nl"]
where $x/fn:string() = fn:tokenize('auto, kind', ',')
return
$x/../@id/fn:string()
returns 'car'.
I changed it so it compares the string value of $x with a list of strings from
tokenize. Note that I'm using = instead of eq. That lets you use a sequence
compare. Also the return now returns the @id attribute node.
- Chris
_______________________________________________
General mailing list
[email protected]
http://developer.marklogic.com/mailman/listinfo/general
_______________________________________________
General mailing list
[email protected]
http://developer.marklogic.com/mailman/listinfo/general
