Validation

FireODM integrates Zod under the hood to enforce schema rules at runtime. Every field decorator you use builds a Zod schema that validates data before it’s written to Firestore (during save(), update() or model instantiation).

1. How It Works

• Field Decorators (@StringField(), @NumberField(), @EmailField(), etc.) register individual Zod schemas for each property.
• FireODM composes these into a single Zod object schema for the whole model.
• On save or update, FireODM calls schema.parse(data):
  • If validation passes, operation proceeds.
  • If validation fails, FireODM throws a ValidationError containing detailed issues.

2. Built-in Decorators & Options

Decorator	Zod Type	Options
@StringField(opts?)	z.string()	min, max, required, message
@EmailField(message?)	z.string().email()	required, custom message
@NumberField(opts?)	z.number()	min, max, required
@EnumField(enumObj, opts?)	z.nativeEnum(enumObj)	required, defaultValue
@BooleanField(opts?)	z.boolean()	required, defaultValue
@TimestampField(opts?)	z.instanceof(Timestamp)	required, defaultNow
@GeoPointField(opts?)	z.instanceof(GeoPoint)	required
@ArrayField(schema, opts?)	z.array(schema)	required
@MapField(schema, opts?)	z.record(z.string(), schema)	required
@DocumentReferenceField(opts?)	union of DocumentReference, BaseModel, null	required

3. Example: Handling Validation Errors

  import { ValidationError } from 'fireodm'
  import { User } from './models'

  async function createUser() {
    try {
      const u = new User({ name: 'A', email: 'not-an-email' })
      await u.save()
    } catch (err) {
      if (err instanceof ValidationError) {
        console.error('Validation failed:', err.issues)
      } else {
        throw err
      }
    }
  }

The ValidationError.issues array includes objects with:

• path: property path (e.g. ['email'])
• message: human-readable error
• code: Zod error code (e.g. invalid_type, too_small)

Next: CRUD API — learn how to create, read, update, and delete documents with FireODM.