How do I access the complex text in an xml file?

I have multiple xml files that have this format:

    <Element>
      <Attribute value="2.0">
        <Nextline name="Hello" value="9999">
           <item name="data" value="111">
        </Attribute>
    </Element>

I want to access the name and value of Nextline and be able to write them into an excel document. If anyone has any advice on how to do this or what I could try, all advice is welcome. I've searched online and have yet to find anything helpful.

Also, if there is a good tutorial for using xml in MATLAB I would love to hear about it!

[per isakson]

Isn't this helpful? Importing XML Documents

Answer 1

[per isakson]

A quick and dirty variant:

    str = fileread( 'cssm.txt' )
    cac = regexp( str, '(?<=<Nextline name=")([^"]+)" value="([^"]+)">', 'tokens')
    cac = 
        {1x2 cell}    {1x2 cell}

    >> cac{2}
    ans = 
        'Hello'    '9999'

where cssm.txt contains two sets of your sample text

&nbsp

Easier to read

    str = fileread( 'cssm.txt' )
    abq = '([^"]+)';    % anything but quotation mark
    xpr = ['<Nextline name="',abq,'" value="',abq,'">'];
    cac = regexp( str, xpr, 'tokens');

[Haley Inniger]

Thank you! This helps immensely. I had never heard of the regexp command before haha

Thanks again!

[Ken Atwell]

Haley, regular expressions will change your life. :)

Answer 2

[Patrick Lloyd]

I have some XML files that I parse like so:

function struct_out = my_xmlread(xml_in)

    % Open file in read mode with fopen() and next line information 
    fid = fopen(xml_in,'r');
    tline = fgetl(fid);

    % Empty struct creation
    struct_out = struct('varname', {}, 'datatype', {});

    % count tracks of each parameter
    count = 1;

    % Loops line by line until end of file is reached. It would be more
    % robust w.r.t. string variations (and more importantly cooler) to use
    % regular expressions to search through this. In its current form, the
    % tags are presumed to have fixed lengths and params are parsed using
    % string indexing.
    while ~feof(fid)

        if strcmp(tline,'<Name>VARIABLE NAME</Name>')
            tline = fgetl(fid);
            struct_out(count).varname = tline(6:end-14);
        elseif strcmp(tline,'<Name>TYPE</Name>')
            tline = fgetl(fid);
            struct_out(count).datatype = tline(6:end-6);
            cout = count + 1;
        end % if strcmp(tline,'<Name>...</Name>')

        % Get the next line
        tline = fgetl(fid);

    end % while ~feof(fid)

    % Close file after reading
    fclose(fid);

end % struct_out = xmlread(xml_in)

It's probably not the best way of doing this but the XML files are all very similar so shortcuts like string indexing can be used. The XML I use looks something like:

<String>
  <Name>VARIABLE NAME</Name>
  <Val>I_AM_THE_PARAMETER (COM1)</Val>
</String>
<String>
  <Name>TYPE</Name>
  <Val>REAL</Val>
</String>

My application isn't identical to yours but some of the techniques may be useful for your application. There's also a built-in xmlread() function but I don't really know how to use that effectively.

[Haley Inniger]

Thanks for your response! Part of my project is working with what you are working with above and I am using a similar technique as you to parse that. However I have found that working with the text nodes like < Name> Variable Name < /Name> is different that working with the complex text elements like < Name value="Hello" >.