Documentation

Migration scripts

How to review your pending changes, export the T-SQL script, and what it contains before you run it yourself.

Review your changes

Go to Workspace → Pending changes. It lists every difference between the schema as you read it and the schema as you have it now, and marks the changes that drop an object or a column.

The count stays until you refresh the project from a database that has your changes. Saving the project does not apply anything.

Export and run it

1. Select Export to download a .sql file.
2. Read the script. The header says how many statements drop data.
3. Run it in SSMS, or through your deployment tool.

WoodFireERD never runs the script against your database, and needs no write permission on it.

What is in the script

• Statements in the order they need to run: foreign keys dropped before their tables, tables created before the keys that reference them, functions before views and procedures.
• Views, functions and procedures as CREATE OR ALTER.
• Everything wrapped in a transaction with SET XACT_ABORT ON, so a failure rolls back.

Dropped tables and columns lose their data. Read the script before you run it.

When a column becomes NOT NULL, or you add a NOT NULL column to a table that already has rows, the script fills the empty rows first. It uses the column's default if it has one. Otherwise it uses blank text, 0, the current date and time, or a new GUID for each row. Every fill is listed in Pending changes, with the value it writes.

Manual steps

Adding, removing or retargeting identity on an existing column cannot be done in T-SQL. Those changes appear at the end of the script as commented instructions, and the header says the script is incomplete without them. Applying them means rebuilding the table by hand.

Identity on a new table or a new column is scripted normally.

Plan

Export is part of Pro and Unlimited. On Free, the Pending changes page tells you so and no script is produced. See Plans and limits.