Running the Tool
Using the Snowpark Migration Accelerator (SMA)
Now that the SMA is installed and the codebase is ready, it’s time to execute. If you still have the application open, head back over to it. If not, open the application.
Project Setup
If you see the Start/Activate page select "Start":
If you have ever used the SMA before, you may not see the above page. You may be taken immediately to the project page. If you do see the above page, choosing "Start" will take you to the project page:
Choose “New Project” in the menu. (If you have already created a project, you can open it by selecting “Open Project”.)
This will take you to the “Project Creation” page. Each execution of the Snowpark Migration Accelerator, whether you are doing an assessment or converting code, is tied to a “project” file that you create in this step. This project is a project file (with a .snowct extension) created in the source directory you previously chose on the project creation page. To tie multiple executions to a single project, you must open an existing project file. This is recorded both locally and in the shared database.
All fields here are essential for both the setup of the assessment tool and the project management that happens after you run the tool. The slide image below overviews each option.
A couple of notes on the final two fields on the project creation menu:
The first field (Email Address) is related to the email you want the email update sent to with the execution ID associated with this run. We recommend that you enter your own email address here. Note that if you put in your Snowflake Email Address, the output reports will be generated locally. If you put in a non-Snowflake email, you must query the database to view the detailed report. (More on this later.)
The second field (Customer’s Company) is for the customer’s organization regardless of the email address put in the previous field. This allows you to connect a project with an organization both locally and in the shared database.
Before you start the assessment, recall the importance of ensuring the input directory has the correct codebase in the correct file type, as discussed in the previous step (Extraction, Preprocessing, and Code Preparation).
When ready, select “SAVE & START ASSESSMENT”:
Execution and Assessment Output
Once you start the assessment process, the SMA will begin analyzing the source code. It runs a basic code analyzer that does a lite scan of the code designed to inventory every file and every keyword detected in the codebase. It then parses the source code based on the source language and builds a semantic model representing what that code actually does. This model is then queried for the rest of the information (such as the Spark Reference Inventory and the Import Library Analysis) and will output the converted code in conversion mode.
This entire process is represented by a few progress markers (Loading Source Code, Analyzing Source Code, and Writing Results) that will illuminate on this screen:
When the analysis finishes, select “View Results” to evaluate the output.
Last updated
