I'm trying to figure out how to incorporate pattern matching on two fields into a workflow so that multiple folders are not created on the same individual. Users want the pattern matching to first look at last name, first name and then an identification number(JID). We currently have it set up to match on last name, first name and it is working but if they add a middle name or initial a new folder is created. We have tried adding the other pattern matching field several different ways but it creates new folders or subfolders. What is the best way to set this up?
Question
Question
Replies
Is the identification number unique? Could you do a match on just that? Can you give a brief overview of how you have the workflow set up? Is it pulling metadata from a document and using that to search for existing folders?
Yes, identification number is unique. Metadata is entered by staff when they scan the documents. All filing is done by last name, first name - identification number and we would like to keep the same structure. I've attached a screenshot of how we have our workflow set up currently.
I guess I am a little confused with what you are pattern matching. I don't see anything in the workflow that looks for existing folders. Can you show a screen shot of your pattern matching.
From what I see on your workflow if someone scans a file today for Doe, John 123456 it would create a folder. If someone scans a file tomorrow for Doe, Jon 123456 it would create a new folder because nothing is looking in the repository to see if there is already a John Doe with ID 12346.
I believe we have the pattern matching set up under Route to Folder with this expression:
\Classifications\%(PatternMatchingLastname_Last Name Initials)\%(RetrieveFieldValues_Name) - %(RetrieveFieldValues_JID Number)\%(RetrieveFieldValues_Incident Type - Classifications)\
I'm not sure if our pattern matching is even set up right as it was done prior to my team taking over administration of Laserfiche - we are all kind of new to this.
No problem - we were all new to it at one point or another. I don't think this workflow is set up to do what you think it is going to do.
I have the same type of situation where multiple people are entering metadata and they are not consistent. Things like Doe, John 123456; Doe, Jon 123456 and Doe, John L 123456 would get entered for the same client. My have metadata on the client folder then the workflow does a triple search of existing folders to try to identify if the person already had a folder created.
Let me see if I am reading your stuff right. Name field is both first and last name in one field? ID is it's own field? There are no restrictions on the field so in Name someone could possibly enter first name last name instead of last name first name? Is there a comma or anything entered in the name field like Doe, John? Or are some of them Doe John?
Yes, name is entered into one field- we've trained them enter the data in that field as Last Name, First Name(Doe, John) but there are no restrictions set(and human error!)... sometimes we have a middle name or middle initial and sometimes we do not(Doe, John Lee or Doe, John L). ID has its own field and is unique to each individual. It sounds like the situation you have is very similar to ours. We will have an individual come to us multiple times and would like to avoid multiple folders for the same person.
Here is a quick rough sketch of something you could try. As long as the ID number is unique (and there are already multiple folders for the same person), you could search for an existing document with that ID number, and move the new document into the same folder as the document found in the search.
It also looks like your path in your current workflow is all tokens. You should be able to eliminate the multiple branches for each type and just through the tokens in one activity.
You can also eliminate the Pattern Matching if all you are using it for is to pull the first letter of the last name. You can do that in the token editor. Use the Name field and just use the token editor to apply the regular expression of (.) to pull the first character in the field.
The Search Repository- is there a way to narrow this down to only search a specific parent folder rather than searching all folders? We have several departments using the same repository for different things and other departments use the ID number as well. We would want a new folder created if there is no match within the Classifications parent folder.
Yep. You can narrow your search by using the "Within Folder" criteria. And a hint to use is if you do your search in Client and use the Search Syntax, you can copy the search and paste it in your workflow.
As to the new folder, that is why I put 2 branches in the workflow. One if matches are found to put it in the existing folder. One if no matches found, it will make the new folder.
Thank you so much for all your help! I think this will work for us