How to Effectively Scan Nested Structs Using SQLX?

Structs Within Structs: A Guide to SQLX's Advanced Scanning

While using SQLX, you might encounter challenges scanning data into nested structs. Let's consider this example:

type Customer struct {
   Id     int    `json:"id" db:"id"`
   Name   string `json:"name" db:"name"`
   Address Address `json:"adress"` // Oops, missing embedded property
}
type Address struct {
   Street string `json:"street" db:"street"`
   City   string `json:"city" db:"city"`
}

When attempting to scan data using this definition, you'll encounter this error:

missing destination name street in *models.Customer

Solution: Embrace Embedded Structs

The key to resolving this issue lies in understanding SQLX's deep scanning capabilities. As the documentation suggests, it supports embedding structs, promoting their fields into the parent struct. To achieve this, simply embed Address into Customer:

type Customer struct {
   Id     int    `json:"id" db:"id"`
   Name   string `json:"name" db:"name"`
   Address Address
}

Note that we removed the Address field's own db tag, as it's no longer a separate entity.

Caution: Flatten the JSON Output

However, embedding Address flattens the JSON output of the Customer struct, as both Name and City are now direct properties:

{
    "id": 1,
    "name": "foo",
    "street": "bar",
    "city": "baz"
}

Possible Alternatives

If this is not desired, there are several alternative approaches:

1. Remapping: Scan data into a map[string]interface{} and remap it using custom logic.
2. Defining Interface Types: Create an interface representing the desired data structure and implement it with both the Address and Customer structs.

The above is the detailed content of How to Effectively Scan Nested Structs Using SQLX?. For more information, please follow other related articles on the PHP Chinese website!