Include metadata/ match input and output file

[HeddaVik]

Hi, 
I am building a process where I need to match the decipher output (document data) to the original file that was pushed into decipher. 

I would like to have more than one file in each batch, but need to be able to match the output to the original file (or more precisely: to the case number the file belongs to). 

I have looked into the "add batch metadata" action, but not sure it it does the trick. 

Is there a way to achieve my goal? 🙂 

All the best, 

Hedda

------------------------------
Hedda Vik
------------------------------

[KrishnaA]

Hello Hedda,

In the Decipher response collection, there are two columns - "Document ID" (GUID generated by the Decipher) and "File Name" which corresponds to the individual documents in a batch.   I think these columns should help you match the output to the original file.

Best regards,

------------------------------
Krishna A
Blue Prism
------------------------------

[HeddaVik]

Hi,
Thank you for helping me. That sounds like what I am looking for, but the collcetion "DocumentData" retruned by the action "Get Document Data" contains only "Document ID". 

What am I missing? 🙂 
All the best, 
Hedda

------------------------------
Hedda Vik
------------------------------

[KrishnaA · Accepted solution]

Am sorry, missed stating that the meta-information you are looking for is available in the Decipher Database.   One option is to write SQL statement to retrieve these details using SQL VBO by querying on "Batch ID"

Best regards,

------------------------------
Krishna A
Blue Prism
------------------------------

[SandeepSatish]

Hi Krishna ,
Is this fine that we hit the decipher database directly to get this information? 
Just trying to understand if any such parallel sessions can cause any locks or degrade performance of application.

------------------------------
Sandeep Satish
Asia/Kolkata
------------------------------

[BenBobo]

Hi Krishna, I see the file names are stored in the SQL DB in the 'dbo.InputFile' table but the names of the files seem to be encrypted. Am I looking in the correct place?

------------------------------
Ben Bobo
RPA Developer
Constellation Brands
America/New_York
------------------------------

[BenLyons]

Hi All,

The file names are obfuscated (a random name is created) during the upload for security purposes, so you won't be able to track your document through the database using the document name.

The only way to track a unique document through Decipher easily, is to have a single document per batch and set the metadata. You could try with multiple documents, but you would need to disable the functions that separate the documents with high confidence and exceptions. If the batch gets split, the metadata will be copied to both batches.

You can get each document ID as you upload it by getting the latest document ID uploaded, but that wouldn't be a recommended method due to the potential load it could place on the server. Please check this with your IT team before implementing.

We're working on a way to make this easier, but that's what's available at the moment.

Thanks

------------------------------
Ben Lyons
Product Consultant
Blue Prism
UK
------------------------------

[IlincaUrse]

Hi, Ben!
Is there an update to this issue?

------------------------------
Ilinca Urse
------------------------------

[BenLyons]

Hi Ilinca,

This has been added to the next release due out next month.

Thanks

Ben

------------------------------
Ben Lyons
Product Consultant
Blue Prism
UK
------------------------------

[MagnusBrangstru]

Hi Ben, 

Where do I find this? I have tried to look through the templates on the jumpstart page but am unable to find the template that some ppl refer to. 

I'm looking for the same as Hedda, preferably without doing the 1-file-1-batch method. Any suggestions?

kr, Magnus

------------------------------
Magnus Brangstrup
------------------------------