Execution

In this section we'll learn how to execute queries in various modes as well as how we unmarshall the data into a Go structure. We'll also learn how to execute multiple query blocks

Execute Query#

In order to execute a query we must run the Execute method. Every time we call Execute a new transaction is created, and it is committed upon success.

response, err := db.
    Query(dqlx.HasFn("name")).
    Select(`
        uid
        name
        species
    `).
    Filter(
        dqlx.EqFn("name", "Ollie"),
    ).
    Execute(ctx)

Manual Transaction#

if you want control over the transaction you can certainly do that. We do that passing dqlx.WithTnx() function to the Execute method

tnx := db.NewTnx()

response, err := db.
    Query(dqlx.HasFn("name")).
    Select(`
        uid
        name
        species
    `).
    Filter(
        dqlx.EqFn("name", "Ollie"),
    ).
    Execute(ctx, dqlx.WithTnx(tnx))

tnx.Commit()

Read Only Transaction#

Similarly, we use WithReadOnly if we want to mark this query to be read-only

response, err := db.
    Query(dqlx.HasFn("name")).
    Select(`
        uid
        name
        species
    `).
    Filter(
        dqlx.EqFn("name", "Ollie"),
    ).
    Execute(ctx, dqlx.WithReadOnly())

Unmarshalling#

Once the query is executed successfully, we want to grab the result set and Unmarshall it into a go data structure.

We can Unmarshall the results by calling Unmarshall on the response object.

response, err := db.
    Query(dqlx.HasFn("name")).
    Select(`
        uid
        name
        species
    `).
    Filter(
        dqlx.EqFn("name", "Ollie"),
    ).
    Execute(ctx)

var data []map[string]interface{}

err := response.Unmarshall(&data)

if err != nil {
    panic(err)
}

// data is now populated!

Another way to achieve the same as above but with fewer lines of code is to use the UnmarshallInto method during the building phase.

var data []map[string]interface{}

response, err := db.
    Query(dqlx.HasFn("name")).
    Select(`
        uid
        name
        species
    `).
    Filter(
        dqlx.EqFn("name", "Ollie"),
    ).
    UnmarshallInto(&data)
    Execute(ctx)

Structs#

You can also Unmarshall into well-defined structs, which is usually the recommended way. The marshaller takes in consideration the json: tag of the field to automatically map the data.

type Animal struct {
    Uid     string `json:"uid"`
    Name    string `json:"name"`
    Species string `json:"species"`
}

var animals []Animal

response, err := db.
    Query(dqlx.HasFn("name")).
    Select(`
        uid
        name
        species
    `).
    Filter(
        dqlx.EqFn("name", "Ollie"),
    ).
    UnmarshallInto(&animals)
    Execute(ctx)