README
¶
go-bigq 
go-bigq is a wrapper to make querying BigQuery easier for the Go programming language.
Even though there is an official package to interact with BigQuery the API of that library is sort of arcane, so I ended up making a layer on top of it to make the experience of querying BigQuery more pleasant.
This package only performs reads, if you are looking for a way to write to BigQuery I recommend you go-bqstreamer.
Usage
// Create the service
service, err := bigq.New(bigq.WithConfigFile("/path/to/token.json"), bigq.Config{
ProjectID: "my-project",
DatasetID: "my-dataset",
})
handleErr(err)
// Perform the query starting at 0 index and with 100 results per page
q, err := service.Query("SELECT foo FROM bar WHERE baz", 0, 100)
handleErr(err)
// Get the next page of results
rows, err := q.NextPage()
handleErr(err)
doSomethingWith(rows)
Loop through results using an iterator
// Create the service
service, err := bigq.New(bigq.WithConfigFile("/path/to/token.json"), bigq.Config{
ProjectID: "my-project",
DatasetID: "my-dataset",
})
handleErr(err)
// Perform the query starting at 0 index and with 100 results per page
q, err := service.Query("SELECT foo FROM bar WHERE baz", 0, 100)
handleErr(err)
// Get iterator
iter := q.Iter()
var myStruct struct {
FirstCol string
SecondCol int
ThirdCol float64
FourthCol bool
}
for iter.Next(&myStruct) {
use(myStruct)
}
handleErr(iter.Err())
TODO
Servicemethod to perform queries with large results (queries whose results have to go to another table).
Documentation
¶
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
Types ¶
type ClientOptions ¶
type ClientOptions interface {
// Service returns the client service
Service() (*bigquery.Service, error)
}
ClientOptions will construct the client service based on the options it was passed.
func WithConfigFile ¶
func WithConfigFile(path string) ClientOptions
WithConfigFile returns a ClientOptions that will construct the client service using a config file.
func WithJWTConfig ¶
func WithJWTConfig(config *jwt.Config) ClientOptions
WithJWTConfig returns a ClientOptions that will construct the client service using the given jwt config.
type Iter ¶
type Iter interface {
// Next fetches the next row and fills the fields of the given
// struct pointer with the columns of the row in appearance order.
// For example, given:
// struct {
// Foo int
// Bar int
// }
//
// The row 1, 2 would result in struct{Foo: 1, Bar:2}
// This method returns a boolean reporting if the operation was
// successful.
Next(interface{}) bool
// Err returns the latest error that happened.
Err() error
}
Iter is a structure to loop through query results.
type Query ¶
type Query interface {
// NextPage returns the next page of rows in the query resultset. It returns up
// to the max results that were given to the query on its creation. Note that
// BigQuery has a limit of 10MB, that is, when your page reaches the 10MB limit
// it will yield and will not return the max number of results instead.
NextPage() ([][]interface{}, error)
// Iter returns an iterator to retrieve the query results.
// Using this method sets the query in "iter" mode, that is,
// the NextPage method can't be used after using Iter, but it
// can be used before retrieving the iterator.
Iter() Iter
}
Query contains all the context of a query execution and has methods to retrieve the rows in pages. The Query instance can be seen as a cursor, it can't go back and it can't re-read the same page again. It is not thread safe and it is not intended to be.
type Service ¶
type Service struct {
// contains filtered or unexported fields
}
Service instances will be able to make queries. A Service is basically a Query constructor that holds the connection with BigQuery.
func New ¶
func New(clientOptions ClientOptions, config Config) (*Service, error)
New creates a new Service with the given client options and config.
func (*Service) Query ¶
Query creates a new query with the SQL sentence passed and a series of arguments. You can pass none, which means no additional parameters. The first parameter passed will be the start, that is, the offset in the resultset. The second parameter passed will be the max results allowed per page.
Source Files