xpath to select nodes excluding by attribute value list

Question

I would like to know if there is shorter approach to selecting a list of nodes ignoring the ones that have an attribute value specified in a list

Working example:

/item[not(@uid='id1') and not(@uid='id2')]

Desired alternative:

/item[not(@uid in('id1','id2'))]

score 26 · Accepted Answer · answered Jul 16 '12 at 07:48

26

You can either use regex - if your xpath implementation supports it - or write

/item[not(@uid='id1' or @uid='id2')]

which might be a bit shorter.

answered Jul 16 '12 at 07:48

Arne

1

You were right, the other method if i had two items like 'id1' and 'id11' it would ignore both by error – Aitorito Jul 16 '12 at 08:56

Dimitre Novatchev · Answer 2 · 2012-07-16T22:00:28.283

7

If the list of attribute names is very long, the following is a good way to handle this situation:

//item[not(contains('|attr1|attr2|attr3|attr4|attr5|attr6|attrN|', 
                    concat('|', @uid, '|')
                    )
           )]

edited Jul 16 '12 at 22:00

answered Jul 16 '12 at 12:42

Dimitre Novatchev

I needed to build very long XPath in .NET. Instead of using stuff like LINQ Aggregate method with seed, you can just simply join strings with '|', add trailing and leading, and use it. Thanks! – Nickmaovich Mar 06 '14 at 11:50

score 4 · Answer 3 · answered Jul 16 '12 at 07:44

4

Maybe something like this??

/item[not(contains('id1 id2', @uid))]

answered Jul 16 '12 at 07:44

VSP

1

works, but risky for changes, as `contains(x, y)` is true for non-existing y. – Arne Jul 16 '12 at 07:50
1

@Arne This one is the shortest but in what cases would it fail? The attribute we use to filter is a unique id without spaces – Aitorito Jul 16 '12 at 07:56

score 0 · Answer 4 · answered Jul 16 '12 at 22:07

0

At least in Xpath 2 general comparisons on sequences are performed pairwise and existentially, so you can write:

/item[not(@uid = ('id1','id2'))]

answered Jul 16 '12 at 22:07

BeniBela

4 Answers4