[JS CS3] Problem with iterating xml tags

Question

Hello,
Below is a script that is supposed to find text in a document and then tag it with an xml tag. I have run into a problem, which if I solve in the way I know how, causes another problem.
When I iterate in a positive direction--i++--the first instance of the text (in this example "abc") is correctly tagged. But the second instance the tag recedes by two characters, that is, tags the three characters prior to the "c." A third instance will recede two more characters, ad nauseam, if my Latin is correct.
The solution is to iterate backwards, i.e., i--. But this causes another problem. If there is more than one story, then the xml structure reverses the order of the stories. The elements within a story are ordered correctly however.
So why is this happening and how can I solve it?
Thanks,
Tom
var myDoc = app.activeDocument;
var findABC = theTextFinder(myDoc,"abc");
xmlTagIt(myDoc,"abc",UIColors.RED,findABC);
//*****functions*******
function xmlTagIt(docRef, tagName,colorOfTag, arrToTag){
    var tagName, colorOfTag,  arrToTag;
    myDoc.xmlViewPreferences.showStructure = true;
    myDoc.xmlViewPreferences.showTaggedFrames =true;
    myDoc.xmlViewPreferences.showTagMarkers = true;
    myDoc.xmlViewPreferences.showTextSnippets =false;
    myDoc.xmlPreferences.defaultStoryTagColor = UIColors.WHITE;
    var rootElement = myDoc.xmlElements.item(0);
    if (myDoc.xmlTags.item(tagName) == null) {
        var tagRef = myDoc.xmlTags.add(tagName, colorOfTag);
  } else {//if tag already exists then stop the script. Avoid double tagging
       myDoc.xmlViewPreferences.showStructure = true;
       alert("You have already run this script.");
       exit();
   }//end else
    //for (var i = arrToTag.length-1; 0 <= i; i--){
    for (var i = 0; i <= arrToTag.length-1;  i++){
        var elementRef = rootElement.xmlElements.add (tagRef,arrToTag[i]);       
     }//end for i
}//end function xmlTagIt
function theTextFinder(docRef,textFindIt){//last argument is the string of text to find
    app.findTextPreferences = NothingEnum.NOTHING;
    app.changeTextPreferences = NothingEnum.NOTHING;
    app.findTextPreferences.findWhat = textFindIt;
    var arrTextFindIt = myDoc.findText();
    return arrTextFindIt;
}//end theTextFinder

Answer 1

Dave,
Thanks for the reply.
I'm not sure what you mean by "convert the references on the fly." You mean find a way to advance two characters forward and tag that? In the script I submitted here I am only searching for "abc," which of course I would not search or at all  in real life. The script uses grep to search for about twenty different typographical errors that typically appear and tags them all.
Reiterating backwards seems, at first sight the way to go. But I haven't a clue as to how to reverse the order of the stories. I can get an array of stories, and array.reverse() them, but am not sure how to then make the script put this reversed order into the xml structure. Any hints?
There, of course is bigger problem with the stories. If the document is written with one story created after another then the order of them is top to botom. But if say a text frame edited into the middle of all this, that story is still the last one in the xml structure. At least in one test, that was so. Maybe the way to solve this is to build an array of stories that have as one xml attribute the page it first appears on. Then order the array according to that attribute.
Any advice you give will be appreciated.
Thanks,
Tom