scan txt file and change related information

Hi, experts

I have a txt file, in which there are some information I want to adjust with efficiency. The txt file is actually generated by social network software: yEd - Graph Editor

the txt file structure is as below (a part of the txt file):

    <node id="n0">
      <data key="d4"><![CDATA[2.0]]></data>
      <data key="d7">
        
          <y:Geometry height="36.0" width="36.0" x="350.0" y="1550.1944444444443"/>
          <y:Fill color="#FFFFFF" transparent="false"/>
          <y:BorderStyle color="#000000" type="line" width="1.0"/>
          <y:NodeLabel alignment="center" autoSizePolicy="content" fontFamily="Dialog" fontSize="12" fontStyle="plain" hasBackgroundColor="false" hasLineColor="false" height="18.1328125" modelName="internal" modelPosition="c" textColor="#000000" visible="true" width="22.97265625" x="6.513671875" y="8.93359375">2.0</y:NodeLabel>
          <y:Shape type="ellipse"/>
        
      </data>
    </node>
    <node id="n1">
      <data key="d4"><![CDATA[4.0]]></data>
      <data key="d7">
        
          <y:Geometry height="36.0" width="36.0" x="2299.1944444444443" y="127.0"/>
          <y:Fill color="#FFFFFF" transparent="false"/>
          <y:BorderStyle color="#000000" type="line" width="1.0"/>
          <y:NodeLabel alignment="center" autoSizePolicy="content" fontFamily="Dialog" fontSize="12" fontStyle="plain" hasBackgroundColor="false" hasLineColor="false" height="18.1328125" modelName="internal" modelPosition="c" textColor="#000000" visible="true" width="22.97265625" x="6.513671875" y="8.93359375">4.0</y:NodeLabel>
          <y:Shape type="ellipse"/>
        
      </data>
    </node>

It's with common sense social network plot is actually connecting nodes by graphic theory. Codes above describes the properties of two nodes, with each node defined within

<node id="n0">
...
</node>

It's the first node. then the second one is

<node id="n1">
...
</node>

using the first node as an example, the second line

key="d4"<![CDATA[2.0]]></data>

indicates the node ID, here is 2.0, the number after "CDATA". One node property is the fill color

color="#FFFFFF" transparent="false"/

#FFFFFF indicates the colour.

The other property is the node shape:

type="ellipse"/

"ellipse" is the shape.

So #FFFFFF and ellipse can be changed to another color and something like 'rectangle' or 'round'.

Now here is my question: this txt file is actually quite long, containing lots of nodes information, and notice that these nodes ID are not put in sequence. I have a list of IDs of nodes of my particular interests, I want to change "color" and "shape" information for those nodes. So, how can I setup an algorithm to locate "color" and "shape" of a particular node based on its ID, and change the "color" and "shape" to a value.

Many thanks for any help!!