You are viewing limited content. For full access, please sign in.

Question

Question

Weblink - Piping the output of one Laserfiche Search into another.

asked on January 6, 2023

I understand that in Laserfiche Search syntax "|" means 'or' but that's not what I'm talking about. I need to try to get Weblink to do a search on a large number of documents for two fields as efficiently as possible. The user wants to enter the values for these two fields and get an EXACT hit.

The search for the first field, Batch Transaction Key, for a specific value looks like this:

{LF:LOOKIN="\Repository Location"} & {[]:[Batch Transaction Key]="202002010251-*"}

This search will yield 13 results and while not the fastest search in the world it's quick enough. The search for the second field is what complicates things.

Each of those 13 results has an an Envelope Number field numbered correspondingly 1-13. A search for the [Envelope Number] looks like this:

{LF:LOOKIN="\Repository Location"} & {[]:[Envelope Number]="1"}
 

And if I combine the two into one search it looks like this:

{LF:LOOKIN="\Repository Location"} & {[]:[Batch Transaction Key]="202002010251-*"} & {[]:[Envelope Number]="1"}

This search takes much longer to run. Why? Because there are literally millions of documents in the same location that have [Envelope Number]=1. It appears to be plowing through them all as part of the search.

But since the first search is relatively quick I wish I could the take 13 results from the 1st search and somehow pipe them into the second search so that it only searches those first 13 results for a particular envelope. That would be rather quick. Does anybody have any ideas on how to do this in Weblink?

I think I could probably workout how to do this with a Workflow but using Workflow as a backend isn't an option. It has to go through Weblink.

Would indexing the fields help make this search faster in any way? I know the Full Text Search service isn't really intended to handle this type of data but give the large number of documents maybe it would help?

 

 

0 0

Replies

replied on January 11, 2023

Yes, indexing fields that are frequently used in this way would absolutely speed up your search results. Indexed Searches and Indexing (laserfiche.com)

0 0
replied on January 11, 2023

Hey Michael,

I would try grouping your search syntax using parenthesis (see below).

 

({LF:LOOKIN="\Repository Location"} & {[]:[Batch Transaction Key]="202002010251-*"}) & {[]:[Envelope Number]="1"}

I am referencing the portion below from the search syntax documentation.

0 0
You are not allowed to follow up in this post.

Sign in to reply to this post.