Todd Knight
fd0da998e8
SDK methods now return 1 error. To differentiate between Client errors and Server errors I've also implemented the AsServerError function which returns an *api.Error if it is an error originating from the remote server or nil otherwise.
43 lines
1.1 KiB
Go
43 lines
1.1 KiB
Go
package authmethods
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
|
|
"github.com/hashicorp/boundary/api/authtokens"
|
|
)
|
|
|
|
func (c *Client) Authenticate(ctx context.Context, authMethodId string, credentials map[string]interface{}, opt ...Option) (*authtokens.AuthTokenReadResult, error) {
|
|
if c.client == nil {
|
|
return nil, fmt.Errorf("nil client in Authenticate request")
|
|
}
|
|
|
|
_, apiOpts := getOpts(opt...)
|
|
|
|
reqBody := map[string]interface{}{
|
|
"credentials": credentials,
|
|
}
|
|
|
|
req, err := c.client.NewRequest(ctx, "POST", fmt.Sprintf("auth-methods/%s:authenticate", authMethodId), reqBody, apiOpts...)
|
|
if err != nil {
|
|
return nil, fmt.Errorf("error creating Authenticate request: %w", err)
|
|
}
|
|
|
|
resp, err := c.client.Do(req)
|
|
if err != nil {
|
|
return nil, fmt.Errorf("error performing client request during Authenticate call: %w", err)
|
|
}
|
|
|
|
target := new(authtokens.AuthTokenReadResult)
|
|
target.Item = new(authtokens.AuthToken)
|
|
apiErr, err := resp.Decode(target.Item)
|
|
if err != nil {
|
|
return nil, fmt.Errorf("error decoding Authenticate response: %w", err)
|
|
}
|
|
if apiErr != nil {
|
|
return nil, apiErr
|
|
}
|
|
|
|
return target, nil
|
|
}
|