Schema Evolution

Changing a table's schema in Revisium automatically transforms existing data. No manual data migration needed — the platform handles it. All changes happen in Draft, nothing is finalized until you commit.

Schema changes are applied as JSON Patches (add, remove, replace, move) — the same format used in the REST API and migrations.

Add Field

When you add a field to an existing table, all rows get the field's default value automatically.

Data Before

{
  "title": "iPhone 16 Pro",
  "price": 999
}

Data After

{
  "title": "iPhone 16 Pro",
  "price": 999,
  "inStock": false
}

New field inStock added with default false — all existing rows updated.

Patch

[
  { "op": "add", "path": "/properties/inStock", "value": { "type": "boolean", "default": false } }
]

The required array is updated automatically.

Review Changes — field inStock added with default false (boolean)

Remove Field

When you remove a field, the data is cleaned from all existing rows.

Data Before

{
  "title": "iPhone 16 Pro",
  "price": 999,
  "legacyCode": "PROD-001"
}

Data After

{
  "title": "iPhone 16 Pro",
  "price": 999
}

Field legacyCode removed — data cleaned from all rows.

Patch

[
  { "op": "remove", "path": "/properties/legacyCode" }
]

The field is also removed from the required array automatically.

Review Changes — field legacyCode was removed

Change Type

When you change a field's type, existing data is converted automatically.

Data Before

{
  "title": "iPhone 16 Pro",
  "price": "999"
}

price is a string.

Data After

{
  "title": "iPhone 16 Pro",
  "price": 999
}

price converted from string to number — "999" → 999.

Patch

[
  { "op": "replace", "path": "/properties/price", "value": { "type": "number", "default": 0 } }
]

Conversion Rules

From → To	Conversion
String → Number	Parsed as number, 0 if invalid
String → Boolean	"true" → true, everything else → false
Number → String	Number as string (42 → "42")
Number → Boolean	0 → false, non-zero → true
Boolean → String	true → "true", false → "false"
Boolean → Number	true → 1, false → 0

Move Field

Move a field from one location to another — between root and nested objects, or between different nesting levels. Data moves with the field. In Admin UI, drag the field to another object.

Schema Editor — drag to move weight field to another object

Data Before

{
  "title": "iPhone 16 Pro",
  "weight": 199,
  "specs": {
    "color": "Desert Titanium"
  }
}

Data After

{
  "title": "iPhone 16 Pro",
  "specs": {
    "color": "Desert Titanium",
    "weight": 199
  }
}

weight moved from root into specs — data relocated automatically.

Patch

[
  { "op": "move", "from": "/properties/weight", "path": "/properties/specs/properties/weight" }
]

Wrap / Unwrap Array

Convert between a primitive field and an array.

Wrap (string → array)

Before: "tag": "electronics"

After: "tags": ["electronics"]

The old value becomes the first element of the new array.

Unwrap (array → string)

Before: "tags": ["electronics", "phones"]

After: "tag": "electronics"

Only the first element is kept — remaining elements are lost. Review changes before committing.

Rename Table

When you rename a table, all foreign key references in other schemas are updated automatically.

Rename: "categories" → "product-categories"

products schema:  "foreignKey": "categories"  →  "foreignKey": "product-categories"
orders schema:    "foreignKey": "categories"  →  "foreignKey": "product-categories"

Rename Row

When you rename a row ID, all foreign key values pointing to that row are updated automatically.

Rename row in categories: "electronics" → "electronics-devices"

products/iphone-16:  category: "electronics"  →  category: "electronics-devices"
products/macbook-m4: category: "electronics"  →  category: "electronics-devices"

Review Changes

When you edit a schema in the Admin UI, the platform generates patches describing what changed. Before applying, a review dialog shows:

• Fields added (with type and default)
• Fields removed
• Type changes
• Fields moved

Schema Evolution — review changes dialog showing field added and field removed

You can review, revert individual changes, or apply all at once. After applying, the data transforms are visible in the Changes & Diff view before commit.

Nested Schema Changes

Schema evolution works at any nesting depth:

• Adding a field to a nested object → updates that object in all rows
• Changing an array item type → converts every element in every row
• Removing a nested field → cleaned from all rows at that path
• Moving a field between nesting levels → data relocated

Migrations

Schema evolution automatically produces migrations (init, update, rename, remove). Migrations can be exported and applied on other instances — useful when independent instances are used instead of a single instance with branches (e.g., dev → staging → production).