KSQL was created to offer an actually simple and satisfactory tool for interacting with SQL Databases in Golang.
The core goal of KSQL is not to offer new features that are unavailable on other libraries (although we do have some), but to offer a well-thought and well-planned API so that users have an easier time, learning, debugging, and avoiding common pitfalls.
KSQL is also decoupled from its backend so that
the actual communication with the database is performed by
well-known and trusted technologies, namely: pgx
and database/sql
.
You can even create your own backend adapter for KSQL which is
useful in some situations.
In this README you will find examples for "Getting Started" with the library, for more advanced use-cases please read our Wiki.
- Every operation returns errors a single time, so its easier to handle them
- Helper functions for everyday operations, namely: Insert, Patch and Delete
- Generic and powerful functions for Querying and Scanning data into structs
- Works on top of existing battle-tested libraries such as
database/sql
andpgx
- Supports
sql.Scanner
andsql.Valuer
and also allpgx
special types (when usingkpgx
) - And many other features designed to make your life easier
This short example below is a TLDR version to illustrate how easy it is to use KSQL.
You will find more complete examples in the sections below.
package main
import (
"context"
"fmt"
"log"
"os"
"github.com/vingarcia/ksql"
"github.com/vingarcia/ksql/adapters/kpgx"
)
var UsersTable = ksql.NewTable("users", "user_id")
type User struct {
ID int `ksql:"user_id"`
Name string `ksql:"name"`
Type string `ksql:"type"`
}
func main() {
ctx := context.Background()
db, err := kpgx.New(ctx, os.Getenv("POSTGRES_URL"), ksql.Config{})
if err != nil {
log.Fatalf("unable connect to database: %s", err)
}
defer db.Close()
// For querying only some attributes you can
// create a custom struct like this:
var count []struct {
Count string `ksql:"count"`
Type string `ksql:"type"`
}
err = db.Query(ctx, &count, "SELECT type, count(*) as count FROM users WHERE type = $1 GROUP BY type", "admin")
if err != nil {
log.Fatalf("unable to query users: %s", err)
}
fmt.Println("number of users by type:", count)
// For loading entities from the database KSQL can build
// the SELECT part of the query for you if you omit it like this:
var users []User
err = db.Query(ctx, &users, "FROM users WHERE type = $1", "admin")
if err != nil {
log.Fatalf("unable to query users: %s", err)
}
fmt.Println("users:", users)
}
Note: In the example above we are using the
$1
,$2
and$3
as placeholders on the query because this example is meant to run on top of Postgres.If you are running on top of MySQL or SQLite use
?
instead, and if you are running on top of SQLServer use@p1
,@p2
and@p3
instead.
We support a few different adapters,
one of them is illustrated above (kpgx
),
the other ones have the exact same signature
but work on different databases or driver versions,
they are:
-
kpgx.New(ctx, os.Getenv("DATABASE_URL"), ksql.Config{})
for Postgres, it works on top ofpgxpool
and pgx version 4, download it with:go get github.com/vingarcia/ksql/adapters/kpgx
-
kpgx5.New(ctx, os.Getenv("DATABASE_URL"), ksql.Config{})
for Postgres, it works on top ofpgxpool
and pgx version 5, download it with:go get github.com/vingarcia/ksql/adapters/kpgx5
-
kmysql.New(ctx, os.Getenv("DATABASE_URL"), ksql.Config{})
for MySQL, it works on top ofdatabase/sql
, download it with:go get github.com/vingarcia/ksql/adapters/kmysql
-
ksqlserver.New(ctx, os.Getenv("DATABASE_URL"), ksql.Config{})
for SQLServer, it works on top ofdatabase/sql
, download it with:go get github.com/vingarcia/ksql/adapters/ksqlserver
-
ksqlite3.New(ctx, os.Getenv("DATBAASE_PATH"), ksql.Config{})
for SQLite3, it works on top ofdatabase/sql
and mattn/go-sqlite3 which relies on CGO, download it with:go get github.com/vingarcia/ksql/adapters/ksqlite3
-
ksqlite.New(ctx, os.Getenv("DATABASE_PATH"), ksql.Config{})
for SQLite, it works on top ofdatabase/sql
and modernc.org/sqlite which does not require CGO, download it with:go get github.com/vingarcia/ksql/adapters/modernc-ksqlite
For more detailed examples see:
./examples/all_adapters/all_adapters.go
The current interface contains the methods the users are expected to use, and it is also used for making it easy to mock the whole library if needed.
This interface is declared in the project as ksql.Provider
and is displayed below.
We plan on keeping it very simple with a small number of well-thought functions that cover all use cases, so don't expect many additions:
// Provider describes the KSQL public behavior
//
// The Insert, Patch, Delete and QueryOne functions return `ksql.ErrRecordNotFound`
// if no record was found or no rows were changed during the operation.
type Provider interface {
Insert(ctx context.Context, table Table, record interface{}) error
Patch(ctx context.Context, table Table, record interface{}) error
Delete(ctx context.Context, table Table, idOrRecord interface{}) error
Query(ctx context.Context, records interface{}, query string, params ...interface{}) error
QueryOne(ctx context.Context, record interface{}, query string, params ...interface{}) error
QueryChunks(ctx context.Context, parser ChunkParser) error
Exec(ctx context.Context, query string, params ...interface{}) (Result, error)
Transaction(ctx context.Context, fn func(Provider) error) error
}
In the example below we'll cover all the most common use cases such as:
- Inserting records
- Updating records
- Deleting records
- Querying one or many records
- Making transactions
More advanced use cases are illustrated on their own pages on our Wiki:
- Querying in Chunks for Big Queries
- Avoiding Code Duplication with the Select Builder
- Reusing Existing Structs on Queries with JOINs
- Testing Tools and
ksql.Mock
For the more common use cases please read the example below, which is also available here if you want to compile it yourself.
package main
import (
"context"
"fmt"
"time"
"github.com/vingarcia/ksql"
"github.com/vingarcia/ksql/adapters/ksqlite3"
"github.com/vingarcia/ksql/nullable"
)
type User struct {
ID int `ksql:"id"`
Name string `ksql:"name"`
Age int `ksql:"age"`
// The following attributes are making use of the KSQL Modifiers,
// you can find more about them on our Wiki:
//
// - https://github.com/VinGarcia/ksql/wiki/Modifiers
//
// The `json` modifier will save the address as JSON in the database
Address Address `ksql:"address,json"`
// The timeNowUTC modifier will set this field to `time.Now().UTC()` before saving it:
UpdatedAt time.Time `ksql:"updated_at,timeNowUTC"`
// The timeNowUTC/skipUpdates modifier will set this field to `time.Now().UTC()` only
// when first creating it and ignore it during updates.
CreatedAt time.Time `ksql:"created_at,timeNowUTC/skipUpdates"`
}
type PartialUpdateUser struct {
ID int `ksql:"id"`
Name *string `ksql:"name"`
Age *int `ksql:"age"`
Address *Address `ksql:"address,json"`
}
type Address struct {
State string `json:"state"`
City string `json:"city"`
}
// UsersTable informs KSQL the name of the table and that it can
// use the default value for the primary key column name: "id"
var UsersTable = ksql.NewTable("users")
func main() {
ctx := context.Background()
// In this example we'll use sqlite3:
db, err := ksqlite3.New(ctx, "/tmp/hello.sqlite", ksql.Config{
MaxOpenConns: 1,
})
if err != nil {
panic(err.Error())
}
defer db.Close()
// In the definition below, please note that BLOB is
// the only type we can use in sqlite for storing JSON.
_, err = db.Exec(ctx, `CREATE TABLE IF NOT EXISTS users (
id INTEGER PRIMARY KEY,
age INTEGER,
name TEXT,
address BLOB,
created_at DATETIME,
updated_at DATETIME
)`)
if err != nil {
panic(err.Error())
}
var alison = User{
Name: "Alison",
Age: 22,
Address: Address{
State: "MG",
},
}
err = db.Insert(ctx, UsersTable, &alison)
if err != nil {
panic(err.Error())
}
fmt.Println("Alison ID:", alison.ID)
// Inserting inline:
err = db.Insert(ctx, UsersTable, &User{
Name: "Cristina",
Age: 27,
Address: Address{
State: "SP",
},
})
if err != nil {
panic(err.Error())
}
// Deleting Alison:
err = db.Delete(ctx, UsersTable, alison.ID)
if err != nil {
panic(err.Error())
}
// Retrieving Cristina, note that if you omit the SELECT part of the query
// KSQL will build it for you (efficiently) based on the fields from the struct:
var cris User
err = db.QueryOne(ctx, &cris, "FROM users WHERE name = ? ORDER BY id", "Cristina")
if err != nil {
panic(err.Error())
}
fmt.Printf("Cristina: %#v\n", cris)
// Updating all fields from Cristina:
cris.Name = "Cris"
err = db.Patch(ctx, UsersTable, cris)
// Changing the age of Cristina but not touching any other fields:
// Partial update technique 1:
err = db.Patch(ctx, UsersTable, struct {
ID int `ksql:"id"`
Age int `ksql:"age"`
}{ID: cris.ID, Age: 28})
if err != nil {
panic(err.Error())
}
// Partial update technique 2:
err = db.Patch(ctx, UsersTable, PartialUpdateUser{
ID: cris.ID,
Age: nullable.Int(28), // (just a pointer to an int, if null it won't be updated)
})
if err != nil {
panic(err.Error())
}
// Listing first 10 users from the database
// (each time you run this example a new Cristina is created)
//
// Note: Using this function it is recommended to set a LIMIT, since
// not doing so can load too many users on your computer's memory or
// cause an Out Of Memory Kill.
//
// If you need to query very big numbers of users we recommend using
// the `QueryChunks` function.
var users []User
err = db.Query(ctx, &users, "FROM users LIMIT 10")
if err != nil {
panic(err.Error())
}
fmt.Printf("Users: %#v\n", users)
// Making transactions:
err = db.Transaction(ctx, func(db ksql.Provider) error {
var cris2 User
err = db.QueryOne(ctx, &cris2, "FROM users WHERE id = ?", cris.ID)
if err != nil {
// This will cause an automatic rollback:
return err
}
err = db.Patch(ctx, UsersTable, PartialUpdateUser{
ID: cris2.ID,
Age: nullable.Int(29),
})
if err != nil {
// This will also cause an automatic rollback and then panic again
// so that we don't hide the panic inside the KSQL library
panic(err.Error())
}
// Commits the transaction
return nil
})
if err != nil {
panic(err.Error())
}
}
The results of the benchmark are good for KSQL, but not flawless.
The next section summarizes the results so its more comprehensible, but if you prefer to read the raw benchmark data just scroll down to the Benchmark Results section.
For transparency purposes this summary will focus at the benchmark showing the worst results for KSQL which is querying multiple lines, this is the summary:
Comparing KSQL running on top of database/sql
with sqlx
, sqlx
is
5% faster than KSQL, which is in practical terms an insignificant difference.
And if KSQL is running on top of pgx
then KSQL becomes 42% faster
because pgx
is significantly faster than sqlx
.
Finally if you are using sqlx
with prepared statements everytime
then sqlx
is 7.5% faster than KSQL on top of pgx
.
So between KSQL vs sqlx
the performance difference is very small, and
if you are using Postgres odds are KSQL will be much faster.
Comparing KSQL running on top of pgx
with pgx
itself, KSQL
is 13.66% slower (on average), which is not insignificant but isn't much either.
Comparing KSQL running on top pgx
with gorm
, KSQL is
11.87% faster than gorm
or inversely gorm
is 13.4% slower.
It is worth noting that KSQL is only caching of prepared statements when using postgres, because this is performed by
pgx
, and this means that when using MySQL, SQLServer or SQLite, if you plan on also using prepared statements other libaries such assqlx
will be significantly faster than KSQL.We are working on adding support for cached prepared statements for these other databases in the future.
To understand the benchmark below you must know that all tests are performed using Postgres 12.1 and that we are comparing the following tools:
- KSQL using the adapter that wraps
database/sql
- KSQL using the adapter that wraps
pgx
database/sql
sqlx
pgx
(withpgxpool
)gorm
sqlc
sqlboiler
For each of these tools, we are running 3 different queries:
The insert-one
query looks like this:
INSERT INTO users (name, age) VALUES ($1, $2) RETURNING id
The single-row
query looks like this:
SELECT id, name, age FROM users OFFSET $1 LIMIT 1
The multiple-rows
query looks like this:
SELECT id, name, age FROM users OFFSET $1 LIMIT 10
Keep in mind that some of the tools tested (like GORM) actually build the queries internally so the actual code used for the benchmark might differ a little bit from the example ones above.
Without further ado, here are the results:
$ make bench TIME=5s
sqlc generate
go test -bench=. -benchtime=5s
goos: linux
goarch: amd64
pkg: github.com/vingarcia/ksql/benchmarks
cpu: Intel(R) Core(TM) i7-10750H CPU @ 2.60GHz
BenchmarkInsert/ksql/sql-adapter/insert-one-12 9711 618727 ns/op
BenchmarkInsert/ksql/pgx-adapter/insert-one-12 10000 555967 ns/op
BenchmarkInsert/sql/insert-one-12 9450 624334 ns/op
BenchmarkInsert/sql/prep-stmt/insert-one-12 10000 555119 ns/op
BenchmarkInsert/sqlx/insert-one-12 9552 632986 ns/op
BenchmarkInsert/sqlx/prep-stmt/insert-one-12 10000 560244 ns/op
BenchmarkInsert/pgxpool/insert-one-12 10000 553535 ns/op
BenchmarkInsert/gorm/insert-one-12 9231 668423 ns/op
BenchmarkInsert/sqlc/insert-one-12 9589 632277 ns/op
BenchmarkInsert/sqlc/prep-stmt/insert-one-12 10803 560301 ns/op
BenchmarkInsert/sqlboiler/insert-one-12 9790 631464 ns/op
BenchmarkQuery/ksql/sql-adapter/single-row-12 44436 131191 ns/op
BenchmarkQuery/ksql/sql-adapter/multiple-rows-12 42087 143795 ns/op
BenchmarkQuery/ksql/pgx-adapter/single-row-12 86192 65447 ns/op
BenchmarkQuery/ksql/pgx-adapter/multiple-rows-12 74106 79004 ns/op
BenchmarkQuery/sql/single-row-12 44719 134491 ns/op
BenchmarkQuery/sql/multiple-rows-12 43218 138309 ns/op
BenchmarkQuery/sql/prep-stmt/single-row-12 89328 64162 ns/op
BenchmarkQuery/sql/prep-stmt/multiple-rows-12 84282 71454 ns/op
BenchmarkQuery/sqlx/single-row-12 44118 132928 ns/op
BenchmarkQuery/sqlx/multiple-rows-12 43824 137235 ns/op
BenchmarkQuery/sqlx/prep-stmt/single-row-12 87570 66610 ns/op
BenchmarkQuery/sqlx/prep-stmt/multiple-rows-12 82202 72660 ns/op
BenchmarkQuery/pgxpool/single-row-12 94034 63373 ns/op
BenchmarkQuery/pgxpool/multiple-rows-12 86275 70275 ns/op
BenchmarkQuery/gorm/single-row-12 83052 71539 ns/op
BenchmarkQuery/gorm/multiple-rows-12 62636 89652 ns/op
BenchmarkQuery/sqlc/single-row-12 44329 132659 ns/op
BenchmarkQuery/sqlc/multiple-rows-12 44440 139026 ns/op
BenchmarkQuery/sqlc/prep-stmt/single-row-12 91486 66679 ns/op
BenchmarkQuery/sqlc/prep-stmt/multiple-rows-12 78583 72583 ns/op
BenchmarkQuery/sqlboiler/single-row-12 70030 87089 ns/op
BenchmarkQuery/sqlboiler/multiple-rows-12 69961 84376 ns/op
PASS
ok github.com/vingarcia/ksql/benchmarks 221.596s
Benchmark executed at: 2023-10-22
Benchmark executed on commit: 35b6882317e82de7773fb3908332e8ac3d127010
The tests use docker-test
for setting up all the supported databases,
which means that:
-
You need to have
docker
installed -
You must be able to run docker without
sudo
, i.e. if you are not root you should add yourself to the docker group, e.g.:$ sudo usermod <your_username> -aG docker
And then restart your login session (or just reboot)
-
Finally run
make pre-download-all-images
only once so your tests don't timeout downloading the database images.
After that, you can just run the tests by using:
make test
- Add an
Upsert
helper method - Try to implement an automatic prepared statements cache like pgx does.
- Update
ksqltest.FillStructWith
to work withksql:"..,json"
tagged attributes - Improve error messages (ongoing)
- Finish the
kbuilder
package
- Test if using a pointer on the field info is faster or not
- Consider passing the cached structInfo as an argument for all the functions that use it, so that we don't need to get it more than once in the same call.
- Use a cache to store often-used queries (like pgx)
- Preload the insert method for all dialects inside
ksql.NewTable()
- Use prepared statements for the helper functions,
Update
,Insert
andDelete
.
- Change the
.Transaction(db ksql.Provider)
to a.Transaction(ctx context.Context)
- Make the
.Query()
method to return atype Query interface { One(); All(); Chunks(); }
- Have an
Update()
method that updates without ignoring NULLs asPatch()
does- Have a new Modifier
skipNullUpdates
so that the Update function will do the job of thePatch
- Remove the
Patch
function.
- Have a new Modifier
- Rename
NewTable()
to justTable()
so it feels right to declare it inline when convenient