Automating FullAppPDF import to TUP from WebAdMIT

Detailed description of desired deliverable:

Looking to determine if it is possible to automate a process which is currently manually done by admissions offices in TUSMGP and Vet, and if it is possible, what is the best tool to use to support it.

PDF documents of application materials are generated by WebAdMIT and made available for download. The file generated is a zip file containing PDF file(s), and the filename of each PDF contains the identifying information used for record matching in Slate (there is no delivered ability to generate a different file type than .zip or a .zip which contains an index file (DIP format)).

These PDF documents are part of the application record for the student and need to be imported to Slate. Slate offers two ways to import and match these PDF documents to student/application records without manual intervention:

1. Fully supported method: If the .zip file contains an index file which "must contain a row for each document in the zip archive. It should contain a column that specifies the file names of the documents within the zip archive. If the Source Format is used to import multiple document types, there should also be a column that specifies the type of document for each row in the file." https://knowledge.technolutions.com/hc/en-us/articles/360033424111-Importing-Documents-Using-a-Document-Import-Processor-Source-Format

2. Supported in a round-about way: If the .zip file is unzipped and the PDFs are loaded to the SFTP as individual files, Slate can process the rest of the way.

Neither of these methods is compatible with the file produced by WebAdMIT, and there is no reason to expect WebAdMIT to offer this in the future (have asked their help desk).

Ideally, I am looking for a solution which can:

1. Automate/schedule a daily or semi-daily "call" to the WebAdMIT API to generate the .zip file for any newly eligible records since the last run

2. Obtain the generated .zip file and either a) generate an index file based on the file names of the PDFs within it and add the index file to the .zip file and deliver that to Slate's SFTP server (preferred) or b) unzip the file so that the individual PDFs can be delivered to Slate's SFTP server.

I have tested and can confirm that I am able to manually make the call via the WebAdMIT API to generate the .zip file and download it (see attached sample). After manually unzipping it, I can add it to Slate's SFTP server and confirm that it matches to records correctly.

WebAdMIT API retrieving documents instructions: https://help.liaisonedu.com/Integration/WebAdMIT/WebAdMIT_API_Guides/03_Retrieving_Documents_via_the_WebAdMIT_API

Research:

After talking with other members of the SIS team, it seems there are at least two possibilities for supporting this:

1. Using Powershell and a Tufts-hosted server

2. Use SnapLogic

Bruce Krasnoff offered to experiment with the powershell/server solution, and I submitted the above description as a support issue request to the SIS developer team. It was assigned to Mario Gonzalez.

Outcome:

Bruce had success building the Powershell script to execute the steps needed, and agreed to run the process on the server he uses for TUSM for a temporary period of time until the SnapLogic process can be built out by the SIS developer group.

PowerShell

In WebAdMIT:

 1. Create list to identify records which should have the documents exported to Slate.

 2. Create the PDF Manager configurations/exports needed.

 3. Create the Destination Manager

 4. Create the Schedule