How to use Linq2Xml to remove nodes with identical value?

Question

I've got a xml file which is like below

<?xml version="1.0" encoding="utf-8"?>
<front>
    <mtext>a+l</mtext>
    <keyword-group type="Fis">
        <kwd>Tove</kwd>
        <kwd>Jani</kwd>
        <kwd>Reminder</kwd>
    </keyword-group>
    <keyword-group type="Lio">
        <kwd>Saio</kwd>
        <kwd>Tove</kwd>
        <kwd>EMP</kwd>
        <kwd>CSTC</kwd>
    </keyword-group>
    <keyword-group type="LioFree">
        <kwd>Iola</kwd>
        <kwd>Jani</kwd>
        <kwd>Priest</kwd>
    </keyword-group>
    <title>Iatola of rocknrolla</title>
</front>

I want to remove all <kwd> nodes having the same value while keeping the first one. So the output xml file should look like this

<?xml version="1.0" encoding="utf-8"?>
<front>
    <mtext>a+l</mtext>
    <keyword-group type="Fis">
        <kwd>Tove</kwd>
        <kwd>Jani</kwd>
        <kwd>Reminder</kwd>
    </keyword-group>
    <keyword-group type="Lio">
        <kwd>Saio</kwd>
        <kwd>EMP</kwd>
        <kwd>CSTC</kwd>
    </keyword-group>
    <keyword-group type="LioFree">
        <kwd>Iola</kwd>
        <kwd>Priest</kwd>
    </keyword-group>
    <title>Iatola of rocknrolla</title>
</front>

I did

XDocument xdoc=XDocument.Load(@"C:\docm\12345.xml");

            xdoc.Descendants("kwd")
                .GroupBy(g => (string)g.Value)
                .Where(g => g.Count() > 1)
                .SelectMany(g => g.Take(1))
                .Remove();

But the xml file remains unchanged...what did I miss?

Answer 1

Try this

XDocument xdoc=XDocument.Load(@"D:\test\12345.XML",LoadOptions.PreserveWhitespace);

xdoc.Descendants("kwd")
 .GroupBy(g => (string)g.Value.ToLower())
 .Where(g => g.Count() > 1)
 .SelectMany(g => g.Skip(1))
 .Remove();

xdoc.Save(@"D:\test\12345.XML",SaveOptions.DisableFormatting);

This will work even when the values of the node kwd have different cases. If you don't want to preserve whitespaces ignore LoadOptions and SaveOptions part.