Hello again

I have issue with removing extra link.   I though I had this solved but I haven't.   Now I can link the file to each other based on the common Field.  For 3 documents creates 4 connect that is 2 for each attachments.      I like to remove the extra link (or the one that is reverse connection).   With my delete function it was hit an miss.  Some files worked and some didn't.    

The question is can I remove the duplicate links?  and , how I do it. 

Thanks for you help

I'm not quite sure where to start...

1. Your search is inefficient. Why are you searching for all documents rather than searching for the ones with the field value you just read from the starting entry?

2. What is the point of reading the field value on each found document and comparing it to itself (first condition in the left branch)?

3. You're not showing the summary for Create Entry (right-click on the activity and select Description Options\Show summaries by default), so I can't tell what you're linking.

4. What is the starting rule?

Guy we need some support from Laserfiche !!!!,

The question is can we create Group Link or one to one connection Links between documents with unique Field Valu?    Thanks for Miruna for sending me the design, above, but it only connects one file to many attachment.   We already have this relationship in the documents, we need equal one to one relationship between files that have the same unique field value.

I ran my own workflow (which creates one to one link relationship but with duplicate links).  In addition, I ran it this morning around 12:00 am for 2500 files.  At 6:30 in the Morning we had around 631 file that had the links.   The process was very slow and CPU utilization was high.

We are open for implementation  via SDK or Workflow and need someone from Laserfiche to help. 

I think I have a solution. Consider the following workflow. It goes through all of the entries in the operating folder, finds entries with the "LINK" tag with a matching UNID, links them, and then gives the current entry the "LINK" tag.

It's the tagging that prevents duplicates, so make sure the 'Search Repository' activity is returning tagged entries with a matching UNID. Your advanced search syntax will look something like this:

{[]:[UNID]="%(UNID)"} & ({LF:Tags="LINK"}) & {LF:LOOKIN="Repository\path\to\files"}

A couple of notes:

• This workflow is designed to be run on a folder of documents that needs to be sorted out. It's not made to link incoming documents--you'll need to make some modifications to make that work (though the design would be very similar).
• When you run it, it'll probably tax your system. I can't think of a way around having to run a search for every entry in the operating folder, but I tried to minimize other expensive calls. 
• This workflow does not take into consideration relationship direction. You didn't seem to care about it in your posts above, so I didn't worry about it either.
• You're going to have to remove the "LINK" tag from all of your documents at the end, though you could tack that on to the end of the workflow pretty easily if that's a concern.
• There may be a more efficient way to do this, but my way should work.