[SCRIPT] Any way to check a hyperlink using specific pageItem as sourcePageItem?

[lfcorullon]

Is there a way to point directly to the hyperlink is using a specific pageItem as sourcePageItem?

For example, when we try to add a new hyperlink to the document, and the sourcePageItem is already in use... I want to remove the hyperlink that is using it and then create the new one.

[brianp311]

Can't really think of a better way than this, but something like: 

var removeSource = function(doc, id) {
    var hlpis = doc.hyperlinkPageItemSources;
    for (var i = 0; i < hlpis.length; i++) {
        if (hlpis[i].sourcePageItem.id == id) {
            hlpis[i].remove();
            break;
        }
    }
}

var main = function() {
    var doc = app.activeDocument;
    var pi = app.selection[0];
    
    try{  
        var myHyperlinkSource = doc.hyperlinkPageItemSources.add(pi);
        myHyperlinkSource.name = "hi" + pi.id;
    }
    catch(e) {
        if (/already in use/gi.test(e)) {
            removeSource(doc, pi.id);
            var myHyperlinkSource = doc.hyperlinkPageItemSources.add(pi);
            myHyperlinkSource.name = "hi" + pi.id;
        } else {
            alert("Something is wrong with the source " + e); 
            return; 
        }
    }
}
main();

[brianp311]

I guess that only gets you to the source. You could point to all hyperlinks instead, then if its source is an instanceof PageItemSource, run that ID check on the page item? 

[lfcorullon]

Very similar to the approach I used.
But, in a document really FULL of hyperlinks, it takes a while in the loop.

[TᴀW]

myText.findHyperlinks() will give you an array of all hyperlinks that intersect with the specified text range.

Ariel

[lfcorullon]

It's a rectangle, not text.
I can't find any similar method for pageItems.

Thanks, Ariel.

[Laubender]

Hi lf.corullon,

could you provide a simple test document?

Some options are already at the table…

You could loop the sources and retrieve the id of a given source if it is a pageItem and compare it to the one you like to use.

Regards,
Uwe Laubender

( ACP )

[lfcorullon]

It's a super simple hyperlink schema.

But the document is FULL of hyperlinks.
As you can see in page 1 of this sample document, I have links to some pages.
I need to update some links (when I run the script again, for example). I have text frames and, when run for the first time, the main script creates rectangles in top of them (client request). If the client runs the script again, I used an approach very similar to Brian's suggestion to check, then loop, then remove to create again. But, since the document is really FULL of hyperlinks, each loop is taking more time than I like.

Thanks, Uwe!

[brianp311]

Perhaps then the script first assigns names to the sources by their corresponding page item IDs, and then uses getByName to recall? Perhaps the bottleneck is somewhere else, ie in the rectangle creation step. 

[lfcorullon]

But the point is I don't want to recall the sourcePageItem. I want to remove the hyperlink where this sourcePageItem is being used in.

[brianp311]

Then hash the hyperlinks and name them by corresponding source id. When you add new ones, name them after the corresponding source. 

[Marc Autret]

Hi @lfcorullon 

For best performance you should probably create a pageItemId-to-hyperSourceId map (once) and cache it in your remove function. Now the way to maintain this map highly depends on the client script. If hyperlinks and/or page items are randomly created, either manually or from your main processing routine, you obviously need to update that map while addressing such and such events.

Anyway the basic mechanism might be implemented as follows.

var removeHyperSource = function(/*Document*/doc,/*PageItem*/item,  id,K,q,re,t,a,b,i,r)
//----------------------------------
// Conditionally removes the hyperlinkPageItemSource object that points out to item.
// If the link no longer exists or cannot be removed, returns false.
// => true [OK]  |  false [KO]
{
   // Checkpoint.
   // ---
   if( (!doc) || 'Document'!=doc.constructor.name ) throw "Invalid document";
   if( !(item||0).isValid || !(id=item.id) ) throw "Invalid PageItem";
   
   // HyperlinkPageItemsSources (collection.)
   // ---
   K = doc.hyperlinkPageItemSources;

   // Id-to-id map, made up once (cache.)
   // q :: { PageItemId => HyperlinkPageItemSourceId }
   // ---
   if( !(q=callee.MAP) )
   {
      re = /\d+(?=\]"\))/g;
      t = K.everyItem();
      a = t.sourcePageItem.toSource().match(re)||[]; // `<pageItemId>`[]
      b = t.id;                                      // <hypiSrcId>[]
      if( a.length != b.length ) throw "Invalid 1-to-1 correspondance.";
      q = callee.MAP = {};
      for( i=-1 ; ++i < a.length ; q[a[i]]=b[i] );
   }

   // Identify and remove the hypersource.
   // ---
   r = (t=q[id]) && (t=K.itemByID(t)).isValid;
   if( r )
   {
      try{ t.remove(); }catch(_){ r=false; }
      delete q[id];
   }

   return r;
};

// TEST
//----------------------------------
var doc = app.properties.activeDocument;
var item = app.selection[0];
var r = removeHyperSource(doc,item);
alert( r ? "OK" : "KO" );

Best,

Marc