XmlPullParser handling multidimensional repeated elements

Question

Im trying to get a list of the top level elements from my XML (that contains duplicated sub elements)

example XML

<feed>
   <folder name="subfolder1">
       <file name="subfile1" />
       <file name="subfile2" />
       <folder name="subsubfolder1">
            <file name="subsubfile1" />
            <file name="subsubfile2" />
        </folder>
   </folder>
   <folder name="subfolder2">
       <file name="subfile1" />
       <file name="subfile2" />
       <folder name="subsubfolder1">
            <file name="subsubfile1" />
            <file name="subsubfile2" />
        </folder>
   </folder>
   <file name="file1"/>
</feed>

I'm trying to get a list of all names of the top level elements e.g.

.subfolder1
.subfolder2

Here is my FeedReader....

private List<Entry> readFeed(XmlPullParser parser) throws XmlPullParserException, IOException {
        List<Entry> entries = new ArrayList<Entry>();

        Log.v("ab", "reed feed started");

        parser.require(XmlPullParser.START_TAG, ns, "feed");
        while (parser.next() != XmlPullParser.END_TAG) {

            if (parser.getEventType() != XmlPullParser.START_TAG) {
                continue;
            }
            String sectionName = parser.getName();

            if(sectionName.equals("folder")) {
                readFolder(parser);
            } 
        }

        return entries;
    }

private void readFolder (XmlPullParser parser) throws XmlPullParserException, IOException  {
        parser.require(XmlPullParser.START_TAG, ns, "folder");
        Log.v("ab", parser.getAttributeValue(null, "name"));
        parser.require(XmlPullParser.END_TAG, ns, "folder");
    }

And here is my LogCat....

09-02 13:40:22.537  31736-31753/? V/ab reed feed started
09-02 13:40:22.537  31736-31753/? V/ab﹕ subfolder1

Can anyone help with why this is stopping after finding the first instance of an folder element?

Lewis McGeary · Accepted Answer · 2015-09-03T13:29:47.483

It looks like a problem with your last parser.require line:

parser.require(XmlPullParser.END_TAG, ns, "folder");

From the documentation, what you're doing here is checking if these conditions are met, and if not, throwing an exception. So you're currently at the 'folder' start tag that you've just read, and you're checking if you're at the 'folder' end tag. Since you're not at the 'folder' end tag, then parser.require will throw an exception.

If you remove that line it should just let your while loop keep going until the next folder start tag.

Edit: here's a full solution

We need to keep going until the end of the document not just until any END_TAG so I amended your while loop to while (parser.next() != XmlPullParser.END_DOCUMENT), then added some extra code after the readFolder method call. If I understood correctly, you are only after the folders named 'subfolder' and skipping the 'subsubfolder's. So I've included a loop which should skip those. I also removed the parser.require lines as I didn't see the need personally, but this is just one way to do it.

private List<Entry> readFeed(XmlPullParser parser) throws XmlPullParserException, IOException {
    List<Entry> entries = new ArrayList<Entry>();

    Log.v("ab", "reed feed started");

    while (parser.next() != XmlPullParser.END_DOCUMENT) {

        if (parser.getEventType() != XmlPullParser.START_TAG) {
            continue;
        }
        String sectionName = parser.getName();

        if(sectionName.equals("folder")) {
            readFolder(parser);

            //these booleans will be used to help us skip the subfolders
            boolean finishedTopLevelElement = false;
            boolean unwantedSubFolderFound = false;

            //this will loop until we are at a "folder" tag and have 
            //confirmed we have finished with the top level folder
            while (!(("folder".equals(parser.getName())) && finishedTopLevelElement)){
                parser.next();

                //we only care about 'folder' tags, for anything else
                //we keep looping
                if ("folder".equals(parser.getName())){

                    if (parser.getEventType() == XmlPullParser.START_TAG){

                        //if we hit a folder start tag, we're at a sub-folder
                        unwantedSubFolderFound = true;

                    } else if (parser.getEventType() == XmlPullParser.END_TAG && !unwantedSubFolderFound){

                        //if we hit a 'folder' end tag and we've not got an unwanted subfolder then
                        //we're done, it's the end tag of the top-level folder
                        finishedTopLevelElement = true;
                        unwantedSubFolderFound = false;

                    } else {

                        //if it's a folder end tag and we HAVE previously found an unwanted sub folder start tag
                        //then we've successfully skipped that sub-folder and can keep looking
                        unwantedSubFolderFound = false;
                    }
                }


            }
        } 
    }

    return entries;
}

private void readFolder (XmlPullParser parser) throws XmlPullParserException, IOException  {
    Log.v("ab", parser.getAttributeValue(null, "name"));
}

hi Lewis, i have tried removing that line but it still only gives the the one first folder name :/ any other ideas? — AndrewBramwell, Sep 03 '15 at 10:14
@AndrewBramwell I've edited the answer above, hopefully it's clear. I've added comments in the code, but if you have any questions let me know. — Lewis McGeary, Sep 03 '15 at 13:32
oh my days! thank you so much for putting the time into this for me! im unable to test at the minute as im about to travel, but i should be able to test later on tonight, ill report back / accept answers and what not a bit later on. again thank you! — AndrewBramwell, Sep 03 '15 at 13:43
No problem, I recently had to do some XML parsing myself so it's useful to consolidate and share the knowledge. — Lewis McGeary, Sep 03 '15 at 14:35

XmlPullParser handling multidimensional repeated elements

1 Answers1