Help with multiple deletes

[Michael Kay]

>Or better yet, can anyone think of a  one pass solution to this problem?
 
Change the second delete to
 
delete node $x//branch[count(.//leaf[@ID != 25]) = 0]
Regards,

Michael Kay
http://www.saxonica.com/
http://twitter.com/michaelhkay 


  _____  

From: http://x-query.com/mailman/listinfo/talk [mailto:http://x-query.com/mailman/listinfo/talk] On Behalf
Of Newman, John W
Sent: 28 August 2009 17:31
To: http://x-query.com/mailman/listinfo/talk
Subject:  Help with multiple deletes



Hello,

 

I have a tree structure and I am trying to use XQuery to delete nodes from
it.   Here's a simple example of what I have and what needs to happen:

 

<tree ID="1">

                <branch ID="1">

                                <branch ID="4"

                                                <leaf ID="6" />

                                </branch>

                </branch>

                <branch ID="2">

                                <branch ID="13">

                                                <branch ID="34">

                                                                <leaf
ID="15" />

                                                                <leaf
ID="16" />

                                                </branch>

                                                <branch ID="65">

                                                                <branch
ID="15">

 
<leaf ID="25" />

                                                                </branch>

                                                </branch>

                                </branch>

                </branch>           

</tree>

 

The user is allowed to delete leaf nodes.  Say the user clicks delete on
leaf 25 there, I'm building an XPath expression down to it, like so:

//tree[@ID = 1] /branch[@ID = 2] /branch[@ID = 13] /branch[@ID = 65]
/branch[@ID = 15] /leaf[@ID = 25]

 

 

The problem is, after deleting the leaf node, I am left with a 'dead'
branch.  So branch 15 should also be deleted, and so on up the tree, in this
case branch 65 also.  Branches 2 & 13 are left in place since count(.//leaf)
> 0

 

I think there are probably several ways to do this with XQuery.   My current
approach should work, in theory, although I am too new to XQuery to figure
out how to write it correctly. J  I think there is probably a clean, 1 line
solution somewhere.  Anyway, this is what I am currently trying:

 

 

Step 1, delete the leaf itself

 

copy $x := doc("tree.xml")
     modify (
         delete node $x//tree[@ID = 1] /branch[@ID = 2] /branch[@ID = 13]
/branch[@ID = 65] /branch[@ID = 15] /leaf[@ID = 25]

     )

     return $x

 

This correctly gives back the tree without that leaf, but it leaves dead
branches:

<tree ID="1">

                <branch ID="1">

                                <branch ID="4"

                                                <leaf ID="6" />

                                </branch>

                </branch>

                <branch ID="2">

                                <branch ID="13">

                                                <branch ID="34">

                                                                <leaf
ID="15" />

                                                                <leaf
ID="16" />

                                                </branch>

                                                <branch ID="65">

                                                                <branch
ID="15">

                                                                </branch>

                                                </branch>

                                </branch>

                </branch>           

</tree>

 

So, step 2, delete any branches that don't have leaves underneath, and then
likewise any trees that don't have branches:

 

copy $x := doc("tree.xml")
     modify (
         delete node $x//tree[@ID = 1] /branch[@ID = 2] /branch[@ID = 13]
/branch[@ID = 65] /branch[@ID = 15] /leaf[@ID = 25]

                , 
         delete node $x//tree//branch[count(.//leaf) = 0]

                , 
         delete node $x//tree[count(.//branch) = 0]            

     )

     return $x

 

However the 2nd delete isn't exactly working, the dead branches are still
there.  I think since it is operating on $x which is the original document
that has the leaf in place.  If I save the output of each delete into a
separate doc, and run each step one at a time, they work fine.  So how I do
write this query to operate on the same xml structure throughout?  Or better
yet, can anyone think of a  one pass solution to this problem?

 

Thanks,

John

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://x-query.com/pipermail/talk/attachments/20090828/84dc363d/attachment-0001.htm