documentation/query/pgwire/go.md
import HighlyAvailableReads from "../../partials/pgwire/_highly_available_reads.partial.mdx" import KnownLimitations from "../../partials/pgwire/_known_limitations.partial.mdx" import ConnectionIssues from "../../partials/pgwire/_connection_issues.partial.mdx" import QueryErrors from "../../partials/pgwire/_query_errors.partial.mdx" import TimestampConfusion from "../../partials/pgwire/_timestamp_confusion.partial.mdx"
QuestDB is tested with the following Go client:
Other Go clients that are compatible with the PostgreSQL wire protocol should also work with QuestDB, but we do not test them. If you find a client that does not work, please open an issue.
QuestDB is a high-performance database. The PGWire protocol has many flavors, and some of them are not optimized for performance. For best performance when querying data from QuestDB with Go, we recommend using pgx.
:::tip
For data ingestion, we recommend using QuestDB's first-party clients with the InfluxDB Line Protocol (ILP) instead of PGWire. PGWire should primarily be used for querying data in QuestDB. QuestDB provides an official Go client for data ingestion using ILP.
:::
QuestDB supports the PostgreSQL Wire Protocol (PGWire) for querying data. This compatibility allows you to use standard Go PostgreSQL clients with QuestDB's high-performance time-series database.
It's important to note that QuestDB's underlying storage model differs from PostgreSQL's, which means some PostgreSQL features may not be available in QuestDB.
The pgx client needs the following connection parameters to connect to QuestDB:
localhost)8812)admin)quest)qdb)pgx is a pure Go PostgreSQL driver and toolkit. It provides more features and better
performance than the older lib/pq driver, which is now in maintenance mode.
database/sql interface and a more powerful native interfaceTo use pgx in your Go project, you need to install it using Go modules:
# For the latest version (v5)
go get github.com/jackc/pgx/v5
pgx provides multiple ways to connect to QuestDB:
package main
import (
"context"
"fmt"
"log"
"github.com/jackc/pgx/v5"
)
func main() {
// Connection string
connString := "postgres://admin:quest@localhost:8812/qdb"
// Create a context
ctx := context.Background()
// Connect to QuestDB
conn, err := pgx.Connect(ctx, connString)
if err != nil {
log.Fatalf("Unable to connect to database: %v", err)
}
defer conn.Close(ctx)
// Verify connection
err = conn.Ping(ctx)
if err != nil {
log.Fatalf("Unable to ping database: %v", err)
}
fmt.Println("Successfully connected to QuestDB!")
}
For more control over connection parameters:
package main
import (
"context"
"fmt"
"github.com/jackc/pgx/v5"
"log"
"time"
)
func main() {
// Create configuration
config, err := pgx.ParseConfig("postgres://admin:quest@localhost:8812/qdb")
if err != nil {
log.Fatalf("Unable to parse config: %v", err)
}
config.ConnectTimeout = time.Second * 10
// Connect with config
ctx := context.Background()
conn, err := pgx.ConnectConfig(ctx, config)
if err != nil {
log.Fatalf("Unable to connect to database: %v", err)
}
defer conn.Close(ctx)
fmt.Println("Successfully connected to QuestDB with custom configuration!")
}
package main
import (
"context"
"fmt"
"log"
"github.com/jackc/pgx/v5"
)
func main() {
ctx := context.Background()
conn, err := pgx.Connect(ctx, "postgres://admin:quest@localhost:8812/qdb")
if err != nil {
log.Fatalf("Unable to connect to database: %v", err)
}
defer conn.Close(ctx)
rows, err := conn.Query(ctx, "SELECT * FROM trades LIMIT 10")
if err != nil {
log.Fatalf("Query failed: %v", err)
}
defer rows.Close()
for rows.Next() {
// Create a map to hold the row data
values, err := rows.Values()
if err != nil {
log.Fatalf("Error scanning row: %v", err)
}
// Print the row data
fmt.Println(values)
}
if err := rows.Err(); err != nil {
log.Fatalf("Error iterating rows: %v", err)
}
}
For more type safety, you can scan results into structs:
package main
import (
"context"
"fmt"
"log"
"time"
"github.com/jackc/pgx/v5"
)
// Trade represents a row in the trades table
type Trade struct {
Timestamp time.Time `db:"timestamp"`
Symbol string `db:"symbol"`
Price float64 `db:"price"`
Amount float64 `db:"amount"`
}
func main() {
ctx := context.Background()
conn, err := pgx.Connect(ctx, "postgres://admin:quest@localhost:8812/qdb")
if err != nil {
log.Fatalf("Unable to connect to database: %v", err)
}
defer conn.Close(ctx)
// Execute a query
rows, err := conn.Query(ctx, "SELECT timestamp, symbol, price, amount FROM trades LIMIT 10")
if err != nil {
log.Fatalf("Query failed: %v", err)
}
defer rows.Close()
// Collect results
var trades []Trade
for rows.Next() {
var t Trade
if err := rows.Scan(&t.Timestamp, &t.Symbol, &t.Price, &t.Amount); err != nil {
log.Fatalf("Error scanning row: %v", err)
}
trades = append(trades, t)
}
// Check for errors from iterating over rows
if err := rows.Err(); err != nil {
log.Fatalf("Error iterating rows: %v", err)
}
// Print trades
for _, t := range trades {
fmt.Printf("Timestamp: %s, Symbol: %s, Price: %.2f, Amount: %.6f\n",
t.Timestamp, t.Symbol, t.Price, t.Amount)
}
}
Parameterized queries help prevent SQL injection and can improve performance:
package main
import (
"context"
"fmt"
"log"
"time"
"github.com/jackc/pgx/v5"
)
func main() {
ctx := context.Background()
conn, err := pgx.Connect(ctx, "postgres://admin:quest@localhost:8812/qdb")
if err != nil {
log.Fatalf("Unable to connect to database: %v", err)
}
defer conn.Close(ctx)
symbol := "BTC-USD"
startTime := time.Now().Add(-7 * 24 * time.Hour) // 7 days ago
rows, err := conn.Query(ctx,
"SELECT timestamp, symbol, price, amount FROM trades WHERE symbol = $1 AND timestamp >= $2 ORDER BY timestamp DESC LIMIT 10",
symbol, startTime)
if err != nil {
log.Fatalf("Query failed: %v", err)
}
defer rows.Close()
fmt.Printf("Recent %s trades:\n", symbol)
for rows.Next() {
var ts time.Time
var sym string
var price, amount float64
if err := rows.Scan(&ts, &sym, &price, &amount); err != nil {
log.Fatalf("Error scanning row: %v", err)
}
fmt.Printf(" %s: Price: %.2f, Amount: %.6f\n", ts, price, amount)
}
if err := rows.Err(); err != nil {
log.Fatalf("Error iterating rows: %v", err)
}
}
For applications that need to execute multiple queries concurrently, connection pooling is recommended:
package main
import (
"context"
"fmt"
"log"
"sync"
"time"
"github.com/jackc/pgx/v5/pgxpool"
)
func main() {
ctx := context.Background()
poolConfig, err := pgxpool.ParseConfig("postgres://admin:quest@localhost:8812/qdb")
if err != nil {
log.Fatalf("Unable to parse pool config: %v", err)
}
poolConfig.MaxConns = 10
poolConfig.MinConns = 1
poolConfig.MaxConnLifetime = time.Hour
poolConfig.MaxConnIdleTime = 30 * time.Minute
pool, err := pgxpool.NewWithConfig(ctx, poolConfig)
if err != nil {
log.Fatalf("Unable to create connection pool: %v", err)
}
defer pool.Close()
if err := pool.Ping(ctx); err != nil {
log.Fatalf("Unable to ping database: %v", err)
}
symbols := []string{"BTC-USD", "ETH-USD", "SOL-USD", "ADA-USD", "XRP-USD"}
var wg sync.WaitGroup
for _, symbol := range symbols {
wg.Add(1)
go func(sym string) {
defer wg.Done()
rows, err := pool.Query(ctx,
"SELECT timestamp, price FROM trades WHERE symbol = $1 ORDER BY timestamp DESC LIMIT 5",
sym)
if err != nil {
log.Printf("Query failed for symbol %s: %v", sym, err)
return
}
defer rows.Close()
fmt.Printf("Latest trades for %s:\n", sym)
for rows.Next() {
var ts time.Time
var price float64
if err := rows.Scan(&ts, &price); err != nil {
log.Printf("Error scanning row: %v", err)
return
}
fmt.Printf(" %s: %.2f\n", ts, price)
}
if err := rows.Err(); err != nil {
log.Printf("Error iterating rows: %v", err)
}
fmt.Println()
}(symbol)
}
wg.Wait()
}
QuestDB provides specialized time-series functions that can be used with pgx:
package main
import (
"context"
"fmt"
"log"
"time"
"github.com/jackc/pgx/v5"
)
func main() {
ctx := context.Background()
conn, err := pgx.Connect(ctx, "postgres://admin:quest@localhost:8812/qdb")
if err != nil {
log.Fatalf("Unable to connect to database: %v", err)
}
defer conn.Close(ctx)
// SAMPLE BY query (time-based downsampling)
sampleByQuery := `
SELECT
timestamp,
symbol,
avg(price) as avg_price,
min(price) as min_price,
max(price) as max_price
FROM trades
WHERE timestamp >= dateadd('d', -7, now())
SAMPLE BY 1h
`
fmt.Println("Executing SAMPLE BY query...")
rows, err := conn.Query(ctx, sampleByQuery)
if err != nil {
log.Fatalf("SAMPLE BY query failed: %v", err)
}
for rows.Next() {
var ts time.Time
var symbol string
var avgPrice, minPrice, maxPrice float64
if err := rows.Scan(&ts, &symbol, &avgPrice, &minPrice, &maxPrice); err != nil {
log.Fatalf("Error scanning row: %v", err)
}
fmt.Printf("Time: %s, Symbol: %s, Avg Price: %.2f, Range: %.2f - %.2f\n",
ts, symbol, avgPrice, minPrice, maxPrice)
}
rows.Close()
// LATEST ON query (last value per group)
fmt.Println("\nExecuting LATEST ON query...")
latestByQuery := "SELECT timestamp, symbol, price, amount FROM trades LATEST ON timestamp PARTITION BY symbol"
rows, err = conn.Query(ctx, latestByQuery)
if err != nil {
log.Fatalf("LATEST ON query failed: %v", err)
}
for rows.Next() {
var ts time.Time
var symbol string
var price, amount float64
// Add additional fields as needed based on your table structure
if err := rows.Scan(&ts, &symbol, &price, &amount); err != nil {
log.Fatalf("Error scanning row: %v", err)
}
fmt.Printf("Symbol: %s, Latest Price: %.2f at %s\n", symbol, price, ts)
}
rows.Close()
}
For larger applications, it's a good practice to create a reusable database client:
package main
import (
"context"
"fmt"
"log"
"time"
"github.com/jackc/pgx/v5/pgxpool"
)
// QuestDBClient provides a simplified interface for interacting with QuestDB
type QuestDBClient struct {
pool *pgxpool.Pool
}
// NewQuestDBClient creates a new QuestDB client with connection pooling
func NewQuestDBClient(ctx context.Context, connString string) (*QuestDBClient, error) {
poolConfig, err := pgxpool.ParseConfig(connString)
if err != nil {
return nil, fmt.Errorf("unable to parse connection string: %w", err)
}
poolConfig.MaxConns = 10
poolConfig.MinConns = 1
poolConfig.MaxConnLifetime = time.Hour
poolConfig.MaxConnIdleTime = 30 * time.Minute
pool, err := pgxpool.NewWithConfig(ctx, poolConfig)
if err != nil {
return nil, fmt.Errorf("unable to create connection pool: %w", err)
}
if err := pool.Ping(ctx); err != nil {
pool.Close()
return nil, fmt.Errorf("unable to ping database: %w", err)
}
return &QuestDBClient{pool: pool}, nil
}
func (c *QuestDBClient) Close() {
if c.pool != nil {
c.pool.Close()
}
}
// GetRecentTrades fetches recent trades for a given symbol
func (c *QuestDBClient) GetRecentTrades(ctx context.Context, symbol string, limit int) ([]Trade, error) {
query := `
SELECT timestamp, symbol, price, amount
FROM trades
WHERE symbol = $1
ORDER BY timestamp DESC
LIMIT $2
`
rows, err := c.pool.Query(ctx, query, symbol, limit)
if err != nil {
return nil, fmt.Errorf("query failed: %w", err)
}
defer rows.Close()
var trades []Trade
for rows.Next() {
var t Trade
if err := rows.Scan(&t.Timestamp, &t.Symbol, &t.Price, &t.Amount); err != nil {
return nil, fmt.Errorf("scan failed: %w", err)
}
trades = append(trades, t)
}
if err := rows.Err(); err != nil {
return nil, fmt.Errorf("error iterating rows: %w", err)
}
return trades, nil
}
// GetSampledData fetches downsampled price data
func (c *QuestDBClient) GetSampledData(ctx context.Context, symbol string, days int) ([]PriceSample, error) {
query := `
SELECT
timestamp,
symbol,
avg(price) as avg_price,
min(price) as min_price,
max(price) as max_price
FROM trades
WHERE symbol = $1 AND timestamp >= dateadd('d', $2, now())
SAMPLE BY 1h
`
rows, err := c.pool.Query(ctx, query, symbol, -days)
if err != nil {
return nil, fmt.Errorf("query failed: %w", err)
}
defer rows.Close()
var samples []PriceSample
for rows.Next() {
var s PriceSample
if err := rows.Scan(&s.Timestamp, &s.Symbol, &s.AvgPrice, &s.MinPrice, &s.MaxPrice); err != nil {
return nil, fmt.Errorf("scan failed: %w", err)
}
samples = append(samples, s)
}
if err := rows.Err(); err != nil {
return nil, fmt.Errorf("error iterating rows: %w", err)
}
return samples, nil
}
// GetLatestPrices fetches the latest price for each symbol
func (c *QuestDBClient) GetLatestPrices(ctx context.Context) ([]Trade, error) {
query := `SELECT timestamp, symbol, price, amount FROM trades LATEST ON timestamp PARTITION BY symbol`
rows, err := c.pool.Query(ctx, query)
if err != nil {
return nil, fmt.Errorf("query failed: %w", err)
}
defer rows.Close()
var trades []Trade
for rows.Next() {
var t Trade
if err := rows.Scan(&t.Timestamp, &t.Symbol, &t.Price, &t.Amount); err != nil {
return nil, fmt.Errorf("scan failed: %w", err)
}
trades = append(trades, t)
}
if err := rows.Err(); err != nil {
return nil, fmt.Errorf("error iterating rows: %w", err)
}
return trades, nil
}
type Trade struct {
Timestamp time.Time
Symbol string
Price float64
Amount float64
}
type PriceSample struct {
Timestamp time.Time
Symbol string
AvgPrice float64
MinPrice float64
MaxPrice float64
}
func main() {
ctx := context.Background()
client, err := NewQuestDBClient(ctx, "postgres://admin:quest@localhost:8812/qdb")
if err != nil {
log.Fatalf("Failed to create QuestDB client: %v", err)
}
defer client.Close()
// Get recent trades
trades, err := client.GetRecentTrades(ctx, "BTC-USD", 5)
if err != nil {
log.Fatalf("Failed to get recent trades: %v", err)
}
fmt.Println("Recent BTC-USD trades:")
for _, t := range trades {
fmt.Printf(" %s: Price: %.2f, Amount: %.6f\n", t.Timestamp, t.Price, t.Amount)
}
// Get sampled data
samples, err := client.GetSampledData(ctx, "BTC-USD", 1)
if err != nil {
log.Fatalf("Failed to get sampled data: %v", err)
}
fmt.Println("\nHourly BTC-USD price samples (last 24 hours):")
for _, s := range samples {
fmt.Printf(" %s: Avg: %.2f, Range: %.2f - %.2f\n",
s.Timestamp, s.AvgPrice, s.MinPrice, s.MaxPrice)
}
// Get latest prices
latestPrices, err := client.GetLatestPrices(ctx)
if err != nil {
log.Fatalf("Failed to get latest prices: %v", err)
}
fmt.Println("\nLatest prices for all symbols:")
for _, t := range latestPrices {
fmt.Printf(" %s: %.2f\n", t.Symbol, t.Price)
}
}
Here's an example of integrating QuestDB with a Go HTTP server:
package main
import (
"context"
"encoding/json"
"fmt"
"log"
"net/http"
"strconv"
"github.com/jackc/pgx/v5/pgxpool"
)
// QuestDBClient, Trade and PriceSample as defined in the previous example
// ...
// API handlers
func main() {
ctx := context.Background()
// Create a database connection pool
pool, err := pgxpool.New(ctx, "postgres://admin:quest@localhost:8812/qdb")
if err != nil {
log.Fatalf("Unable to create connection pool: %v", err)
}
defer pool.Close()
client := &QuestDBClient{pool: pool}
http.HandleFunc("/api/trades", func(w http.ResponseWriter, r *http.Request) {
handleTrades(w, r, client)
})
http.HandleFunc("/api/sampled", func(w http.ResponseWriter, r *http.Request) {
handleSampledData(w, r, client)
})
http.HandleFunc("/api/latest", func(w http.ResponseWriter, r *http.Request) {
handleLatestPrices(w, r, client)
})
port := 8080
fmt.Printf("Starting server on port %d...\n", port)
log.Fatal(http.ListenAndServe(fmt.Sprintf(":%d", port), nil))
}
func handleTrades(w http.ResponseWriter, r *http.Request, client *QuestDBClient) {
// Parse query parameters
symbol := r.URL.Query().Get("symbol")
limitStr := r.URL.Query().Get("limit")
if symbol == "" {
symbol = "BTC-USD" // Default symbol
}
limit := 10 // Default
if limitStr != "" {
var err error
limit, err = strconv.Atoi(limitStr)
if err != nil || limit <= 0 {
http.Error(w, "Invalid limit parameter", http.StatusBadRequest)
return
}
}
trades, err := client.GetRecentTrades(r.Context(), symbol, limit)
if err != nil {
http.Error(w, fmt.Sprintf("Error fetching trades: %v", err), http.StatusInternalServerError)
return
}
w.Header().Set("Content-Type", "application/json")
if err := json.NewEncoder(w).Encode(trades); err != nil {
http.Error(w, fmt.Sprintf("Error encoding response: %v", err), http.StatusInternalServerError)
}
}
func handleSampledData(w http.ResponseWriter, r *http.Request, client *QuestDBClient) {
symbol := r.URL.Query().Get("symbol")
if symbol == "" {
http.Error(w, "Symbol parameter is required", http.StatusBadRequest)
return
}
daysStr := r.URL.Query().Get("days")
days := 7 // Default
if daysStr != "" {
var err error
days, err = strconv.Atoi(daysStr)
if err != nil || days <= 0 {
http.Error(w, "Invalid days parameter", http.StatusBadRequest)
return
}
}
samples, err := client.GetSampledData(r.Context(), symbol, days)
if err != nil {
http.Error(w, fmt.Sprintf("Error fetching sampled data: %v", err), http.StatusInternalServerError)
return
}
w.Header().Set("Content-Type", "application/json")
if err := json.NewEncoder(w).Encode(samples); err != nil {
http.Error(w, fmt.Sprintf("Error encoding response: %v", err), http.StatusInternalServerError)
}
}
func handleLatestPrices(w http.ResponseWriter, r *http.Request, client *QuestDBClient) {
latestPrices, err := client.GetLatestPrices(r.Context())
if err != nil {
http.Error(w, fmt.Sprintf("Error fetching latest prices: %v", err), http.StatusInternalServerError)
return
}
w.Header().Set("Content-Type", "application/json")
if err := json.NewEncoder(w).Encode(latestPrices); err != nil {
http.Error(w, fmt.Sprintf("Error encoding response: %v", err), http.StatusInternalServerError)
}
}
When using pgx with QuestDB, be aware of these limitations:
DECLARE CURSOR and subsequent operations.pgxpool package.SAMPLE BY and LATEST ON for
efficient queries.QuestDB provides specialized time-series functions that can be used with pgx:
SAMPLE BY is used for time-based downsampling:
SELECT timestamp,
symbol,
avg(price) as avg_price,
min(price) as min_price,
max(price) as max_price
FROM trades
WHERE timestamp >= dateadd('d', -7, now()) SAMPLE BY 1h;
LATEST ON is an efficient way to get the most recent values:
SELECT *
FROM trades
WHERE timestamp IN today()
LATEST ON timestamp PARTITION BY symbol;
pgx provides a robust way to connect Go applications to QuestDB through the PostgreSQL Wire Protocol. By following the guidelines in this documentation, you can effectively query time-series data from QuestDB and integrate it with various Go applications.
For data ingestion, it's recommended to use QuestDB's first-party clients with the InfluxDB Line Protocol (ILP) for high-throughput data insertion.
QuestDB's SQL extensions for time-series data, such as SAMPLE BY and LATEST ON, provide powerful tools for analyzing
time-series data that can be easily accessed through pgx.