A super fast key-value store with SQLite that uses bun:sqlite and v8 as a fast JSON replacement.
Bun's lightning-fast SQLite implementation makes Bun-SQLite-Key-Value perfect for a fast and reliable storage and cache solution with TTL support. You need Bun to be able to use this package.
The ideas for the implementation come from bun-sqlite-cache and bun-kv. Thank you very much!
set()
get()
getSet()
getValues()
getValuesAsSet()
--> Set with valuessetItems()
getItem()
getItems()
getItemsAsObject()
--> Object with itemsgetItemsAsMap()
--> Map with itemshSet()
hGet()
hmSet()
hmGet()
hHasField()
hGetCount()
hGetFields()
hGetValues()
hDelete()
hIncr()
--> Increments the value of a field.hDecr()
--> Decrements the value of a field.lPush()
--> Insert elements at the begin of the list.rPush()
--> Insert elements at the end of the list.lPop()
--> Returns elements at the begin of the list and removes them.rPop()
--> Returns elements at the end of the list and removes them.lIndex()
--> Returns the element at the specified index of the list.lLen()
--> Returns the length of the list.lSet()
--> Updates the element at the specified index of the list.addTag()
--> Adds a tag to an item.deleteTag()
--> Deletes a tag of an item.deleteTags()
--> Deletes multiple tags or all tags of an item.deleteTaggedItems()
--> Deletes tagged items.getTaggedKeys()
--> Returns tagged keys.getTaggedValues()
--> Returns tagged values.getTaggedItems()
--> Returns tagged items.bun add bun-sqlite-key-value
Using this key value store is dead simple: create a new BunSqliteKeyValue instance and you're set. And if you want to save the data permanently, enter the path to the database.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
// Use regular methods to write and read values.
store.set("myKey", [1, 2, 3, 4])
store.get("myKey") // --> [ 1, 2, 3, 4 ]
// Or use the `data` object for simple write and read access.
store.data.myKey = "Hello world!"
store.data.myKey // --> "Hello World"
const store = new BunSqliteKeyValue(filename?, options?)
Opens and creates the SQLite database either in memory or on the file system.
The full path of the SQLite database to open.
Pass an empty string (""
) or ":memory:"
or undefined
for an in-memory database.
readonly?: boolean
:
Open the database as read-only (default: false).
create?: boolean
:
Allow creating a new database (default: true).
If the database folder does not exist, it will be created.
readwrite?: boolean
:
Open the database as read-write (default: true).
ttlMs?: boolean
:
Default time span in milliseconds before an item
written to the DB becomes invalid and is marked for deletion.
maxExpiringItemsInDb?: number
:
Default value that specifies the maximum number of
expiring items that may be in the database.
Is used by the deleteOldExpiringItems()
method as default value.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
// In-memory
const store1 = new BunSqliteKeyValue()
// In-memory with 30 seconds default expiration timeout
const store2 = new BunSqliteKeyValue(":memory:", {ttlMs: 30000})
// Store items in file system
const store3 = new BunSqliteKeyValue("./store3.sqlite")
close()
Closes database and removes .sqlite-shm and .sqlite-wal files.
set(key: string | undefined, value: any, ttlMs?: number): Key
data.<key> = <value>
data[<key>] = <value>
Writes a value into the database and returns the key.
The key
can be a string or undefined
.
If the key
is undefined
, a UUID is used instead.
The value can be any object that can be serialized with v8. This means that not only simple data types (string, number) are possible, but also more complex types such as sets or maps. You can find a list of the supported data types here.
"Time to live" in milliseconds. After this time, the item becomes invalid and is deleted from the database the next time it is accessed or when the application is started. Set the value to 0 if you want to explicitly deactivate the process.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
// Stays in database
store.set("myKey1", "my-value")
store.data.myKey2 = "my-value"
store.data["myKey3"] = "my-value"
// Becomes invalid after 30 seconds
store.set("myKey6", "item-with-ttl", 30000)
get(key: string): any
data.<key>: any
data[<key>]: any
Reads a value from the database.
The key must be a string.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("myKey", "my-value")
store.get("myKey") // --> "my-value"
store.data.myKey // --> "my-value"
store.data["myKey"] // --> "my-value"
getSet(key: string, value: any, ttlMs?: number)
Atomically sets key to value and returns the old value stored at key. Inspired by: https://docs.keydb.dev/docs/commands/#getset
The key must be a string.
The value can be any object that can be serialized with v8. This means that not only simple data types (string, number) are possible, but also more complex types such as sets or maps. You can find a list of the supported data types here.
"Time to live" in milliseconds. After this time, the item becomes invalid and is deleted from the database the next time it is accessed or when the application is started. Set the value to 0 if you want to explicitly deactivate the process.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("key-1", "string-value-1")
store.getSet("key-1", "string-value-2")) // --> "string-value-1"
store.get("key-1") // --> "string-value-2"
getValues(startsWithOrKeys?: string | string[]): any[]
<store>.values
Reads the data from the database and returns an array with the values.
undefined
: Returns an array with all values.
string
: Returns an array with all values whose keys begin with the passed string.
If you plan the names of the keys well, more complex data can be stored.
It is advisable to divide keys into ranges using separators.
For example "language:de"
, "language:en"
, "language:it"
.
A search for "language:"
would return all languages.
string[]
: Array with keys. The returned array is exactly
the same size as the passed array.
Items that are not found are returned as undefined
.
Only exact matches with the keys are returned.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("language:de", "German")
store.set("language:en", "English")
store.set("language:it", "Italian")
store.getValues() // --> [ "German", "English", "Italian" ]
store.getValues("language:") // --> [ "German", "English", "Italian" ]
store.values // --> [ "German", "English", "Italian" ]
setItems(items: {key: string, value: T, ttlMs?: number}[])
Adds a large number of items to the database and takes only
a small fraction of the time that set()
would take individually.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
// Add many records
store.setItems([
{key: "a:1", value: "test-value-1"},
{key: "a:2", value: "test-value-2"},
])
Reads the key and the value from the database.
getItem(key: string): {key: string, value: any}
The key must be a string.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("my-key", "my-value")
const item = store.getItem("my-key")
console.log(item) // --> {key: "my-key", value: "my-value"}
getItems(startsWithOrKeys?: string | string[]): {key: string, value: any}[]
<store>.items
Reads the data from the database and returns items in an array as key-value pairs.
undefined
: Returns all items (key, value) in an array.
string
: Returns all items (key, value) in an array whose keys begin with
the passed string.
If you plan the names of the keys well, more complex data can be stored.
It is advisable to divide keys into ranges using separators.
For example "language:de"
, "language:en"
, "language:it"
.
A search for "language:"
would return all languages.
string[]
: Array with keys. The returned array is exactly
the same size as the passed array.
Items that are not found are returned as undefined
.
Only exact matches with the keys are returned.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("language:de", "German")
store.set("language:en", "English")
store.set("language:it", "Italian")
store.getItems("language:") // --> [
// {key: "language:de", value: "German"},
// {key: "language:en", value: "English"},
// {key: "language:it", value: "Italian"}
// ]
store.items // --> [
// {key: "language:de", value: "German"},
// {key: "language:en", value: "English"},
// {key: "language:it", value: "Italian"}
// ]
SQLite has no problem with images and other binaries. The maximum size of a binary file is 2 GB.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
// Read file from filesystem
const sourceFile = Bun.file("<Source File Path>")
// Write ArrayBuffer into database (async !!!)
store.set("my-image", await sourceFile.arrayBuffer())
// Read ArrayBuffer from database
const targetArrayBuffer = store.get("my-image")
// Write target file to filesystem (async !!!)
await Bun.write(Bun.file("<Target File Path>"), targetArrayBuffer)
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
import { readFileSync, writeFileSync } from "node:fs"
const store = new BunSqliteKeyValue()
// Read content from filesystem
const sourceContent = readFileSync("<Source File Path>")
// Write Buffer into database
store.set("my-image", sourceContent)
// Read Buffer from database
const targetBuffer = store.get("my-image")
// Write target file to filesystem
writeFileSync("<Target File Path>", targetBuffer)
has(key: string): boolean
exists(key: string) // --> alias for has()
<key> in <store>.data
Checks if key exists. Returns false
if the item is expired.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.has("my-key") // --> false
console.log("my-key" in store.data) // --> false
getKeys(startsWithOrKeys?: string | string[]): string[]
<store>.keys // --> all keys
Reads the keys from the database and returns an array.
undefined
: Returns all keys in an array.
string
: Returns an array with the keys that begin with the passed string.
If you plan the names of the keys well, more complex data can be stored.
It is advisable to divide keys into ranges using separators.
For example "language:de"
, "language:en"
, "language:it"
.
A search for "language:"
would return all languages.
string[]
: Array with keys.
Only exact matches with the keys are returned.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("language:de", "German")
store.set("language:en", "English")
store.set("language:es", "Esperanto")
store.getKeys() // --> ["language:de", "language:en", "language:es"]
store.keys // --> ["language:de", "language:en", "language:es"]
store.getKeys("language:e") // --> ["language:en", "language:es"]
store.getKeys(["language:de", "language:fr"]) // --> ["language:de"]
rename(oldKey: string, newKey: string): boolean
Renames oldKey
to newKey
.
It returns false
when oldKey
does not exist.
If newKey
already exists it is deleted first.
Inspired by: https://docs.keydb.dev/docs/commands/#rename
delete(keyOrKeys?: string | string[])
clear() // --> delete all items
delete <store>.data.<key>
Deletes all items if no parameter was passed.
key: string
: Deletes the item whose key was passed as a string.
keys: string[]
: Deletes the items whose keys were passed in an array.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
// Delete all items
store.delete()
store.clear()
// Delete one item
store.delete("myKey")
delete store.data.myKey
// Delete multiple items
store.delete(["key1", "key2"])
deleteExpired()
Deletes all expired items.
These are items whose TTL (Time to live) has expired.
These items are not deleted continuously,
but only when they are accessed directly or when the database is opened.
If you want to delete the expired items in between,
you can do this with deleteExpired()
.
deleteOldExpiringItems(maxExpiringItemsInDb?: number)
If there are more expiring items in the database than maxExpiringItemsInDb
,
the oldest items are deleted until there are only maxExpiringItemsInDb
items with
an expiration date in the database.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("static:1", "my-value")
store.set("static:2", "my-value")
store.set("dynamic:1", "my-value", 4000)
store.set("dynamic:2", "my-value", 5000)
store.set("dynamic:3", "my-value", 6000)
store.deleteOldExpiringItems(2)
console.log(store.getKeys("dynamic:"))
// --> [ "dynamic:2", "dynamic:3" ]
getCount(): number
count() // --> alias for getCount()
length // --> getter method for `getCount()`
Returns the number of all items, including those that have already expired.
The fact that possibly expired items are also counted is for reasons of speed.
Use getCountValid()
if you want to get the number of items that have not yet expired.
If you do not use ttlMs
(time to live), getCount()
is faster than getCountValid()
.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("my-key1", "my-value1")
store.set("my-key2", "my-value2")
store.getCount() // --> 2
store.length // --> 2
getCountValid(deleteExpired?: boolean): number
Returns the number of valid (non-expired) items. Can also delete the expired items.
If true
is passed, the expired items are deleted first
before the items are counted.
If the parameter is not specified or false
is passed,
then only the items that have no expiration date or
whose expiration date is in the future are counted.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("my-key1", "my-value1")
store.set("my-key2", "my-value2", 100)
store.getCountValid() // --> 2
await Bun.sleep(500)
store.getCountValid() // --> 1
You can specify a caching period when you open the database.
This period in milliseconds is then added with each write.
If you read the value within this period, the value is returned.
If the value is read after this period, undefined
is returned.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue(":memory:", {ttlMs: 1000})
const KEY = "cache-key"
store.set(KEY, 12345)
await Bun.sleep(500)
console.log(store.get(KEY)) // --> 12345
await Bun.sleep(1000)
console.log(store.get(KEY)) // --> undefined
setTtl(key: string, ttlMs?: number): boolean
Renews or deletes the TTL of the database row.
Returns true
if the key
exists.
The key must be a string.
"Time to live" in milliseconds. After this time,
the item becomes invalid and is deleted from the database
the next time it is accessed or when the application is started.
Uses the global ttlMs
as default value.
Set the value to 0 if you want to delete the TTL.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("my-key", "my-value", 10000)
// Update TTL
store.setTtl("my-key", 10000) // --> true
// Delete TTL
store.setTtl("my-key", 0) // --> true
getTtl(key: string): number | undefined
Returns how long the data record is still valid (in milliseconds).
Returns undefined
if the key
does not exist or no expiration date has been set.
Inspired by: https://docs.keydb.dev/docs/commands/#ttl
The key must be a string.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("my-key", "my-value", 20000)
await Bun.sleep(1)
store.getTtl("my-key") // --> 19999
getRandomValue(): any // --> random value
randomValue() // --> alias for getRandomValue()
Returns a random value or undefined
if no valid item was found.
Inspired by: https://docs.keydb.dev/docs/commands/#randomkey
getRandomItem() // --> random item
randomItem() // --> alias for getRandomItem()
Returns a random item or undefined
if no valid item was found.
Inspired by: https://docs.keydb.dev/docs/commands/#randomkey
getRandomKey() // --> random key
randomKey() // --> alias for getRandomKey()
Returns a random key or undefined
if no valid item was found.
Inspired by: https://docs.keydb.dev/docs/commands/#randomkey
incr(key: string, incrBy: number = 1, ttlMs?: number): number
Increments the saved number by incrBy
(default = 1),
saves the new number and returns it.
If the key does not yet exist in the database,
the value is set to 0 before being incremented by incrBy
.
If a string is stored in the database that can be converted into a number,
this is converted first.
If the stored value cannot be converted into a number, NaN
is returned.
The key must be a string.
The stored number is increased by this value.
"Time to live" in milliseconds. After this time, the item becomes invalid and is deleted from the database the next time it is accessed or when the application is started. Set the value to 0 if you want to explicitly deactivate the process.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.incr("my-key") // --> 1
store.incr("my-key") // --> 2
decr(key: string, decrBy: number = 1, ttlMs?: number): number
Decrements the saved number by decrBy
(default = 1),
saves the new number and returns it.
If the key does not yet exist in the database,
the value is set to 0 before being decremented by decrBy
.
If a string is stored in the database that can be converted into a number,
this is converted first.
If the stored value cannot be converted into a number, NaN
is returned.
The key must be a string.
The stored number is decreased by this value.
"Time to live" in milliseconds. After this time, the item becomes invalid and is deleted from the database the next time it is accessed or when the application is started. Set the value to 0 if you want to explicitly deactivate the process.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.set("my-key", 10)
store.decr("my-key") // --> 9
store.decr("my-key") // --> 8
append(key: string, value: string, ttlMs?: number): number
If key already exists, this command appends the value at the end of the string.
If key does not exist it is created and set as an empty string,
so append()
will be similar to set()
in this special case.
Inspired by: https://docs.keydb.dev/docs/commands/#append
Returns the length of the string after the append operation.
The key must be a string.
The string that is appended to the existing string.
"Time to live" in milliseconds. After this time, the item becomes invalid and is deleted from the database the next time it is accessed or when the application is started. Set the value to 0 if you want to explicitly deactivate the process.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.append("my-key", "Hello!") // --> 6
store.append("my-key", "World!") // --> 12
store.get("my-key") // --> "Hello!World!"
hSet(key: string, field: string, value: any, ttlMs?: number)
First the
JavaScript Map Object
is read from the database.
If the data record does not yet exist, a new "Map Object" is created.
Then the entry marked with field
is added to the "Map Object" or overwritten.
Finally, the modified "Map Object" is written back to the database.
Inspired by: https://docs.keydb.dev/docs/commands/#hset
Do not use the hash functions with several very large amounts (megabytes) of data or blobs. This is because the entire data record with all fields is always read and written. It is better to use
setValues()
andgetValues()
for large amounts of data.
The key must be a string.
The field name must be a string.
The value can be any object that can be serialized with v8. This means that not only simple data types (string, number) are possible, but also more complex types such as sets or maps. You can find a list of the supported data types here.
"Time to live" in milliseconds (for the database line, marked with key
).
After this time, the item becomes invalid and is deleted from the database
the next time it is accessed or when the application is started.
Set the value to 0 if you want to explicitly deactivate the process.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.hSet("key-1", "field-1", "value-1")
store.hSet("key-1", "field-2", "value-2")
store.get("key-1") // --> Map(2) {
"name-1": "value-1",
"name-2": "value-2",
}
hGet(key: string, field: string)
First the
JavaScript Map Object
is read from the database.
If the data record (marked with key
) does not exist, undefined
is returned.
If the field (marked with field
) does not exist in the "Map Object", undefined
is returned.
Do not use the hash functions with several very large amounts (megabytes) of data or blobs. This is because the entire data record with all fields is always read and written. It is better to use
setValues()
andgetValues()
for large amounts of data.
Inspired by: https://docs.keydb.dev/docs/commands/#hget
The key must be a string.
The field name must be a string.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.hSet("key-1", "field-1", "value-1")
store.hGet("key-1", "field-1") // --> "value-1"
store.hGet("key-1", "field-2") // --> undefined
hmSet(key: string, fields: {[field: string]: T}, ttlMs?: number)
Like hSet()
, with the difference that several fields
are written to the database in one go.
Do not use the hash functions with several very large amounts (megabytes) of data or blobs. This is because the entire data record with all fields is always read and written. It is better to use
setValues()
andgetValues()
for large amounts of data.
Inspired by: https://docs.keydb.dev/docs/commands/#hmset
The key must be a string.
Object with field names (keys) and values.
"Time to live" in milliseconds (for the database line, marked with key
).
After this time, the item becomes invalid and is deleted from the database
the next time it is accessed or when the application is started.
Set the value to 0 if you want to explicitly deactivate the process.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.hmSet("my-key", {
"field-1": "value-1",
"field-2": "value-2",
"field-3": "value-3"
})
hmGet(key: string, fields: fields?: string[])
Like hGet()
, with the difference that several fields are read in one go.
Do not use the hash functions with several very large amounts (megabytes) of data or blobs. This is because the entire data record with all fields is always read and written. It is better to use
setValues()
andgetValues()
for large amounts of data.
Inspired by: https://docs.keydb.dev/docs/commands/#hmget
The key must be a string.
Array with field names. If the parameter is not specified, all fields are returned.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.hmSet("my-key", {
"field-1": "value-1",
"field-2": "value-2"
})
store.hmGet(KEY_1, ["field-1", "field-100"]) // --> {
// "field-1": "value-1",
// "field-100": undefined,
// }
hHasField(key: string, field: string)
Returns if field
is an existing field in the hash stored at key
.
Do not use the hash functions with several very large amounts (megabytes) of data or blobs. This is because the entire data record with all fields is always read and written. It is better to use
setValues()
andgetValues()
for large amounts of data.
Inspired by: https://docs.keydb.dev/docs/commands/#hexists
The key must be a string.
The field name must be a string.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.hSet("key-1", "field-1", "value-1")
store.hHasField("key-1", "field-1") // --> true
store.hHasField("key-1", "field-1") // --> undefined
hGetCount(key: string)
Returns the number of fields contained in the hash stored at key
.
Do not use the hash functions with several very large amounts (megabytes) of data or blobs. This is because the entire data record with all fields is always read and written. It is better to use
setValues()
andgetValues()
for large amounts of data.
Inspired by: https://docs.keydb.dev/docs/commands/#hlen
The key must be a string.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.hGetCount("key-1") // --> undefined
store.hSet("key-1", "field-1", "value-1")
store.hGetCount("key-1") // --> 1
hGetFields(key: string)
Returns the field names contained in the hash stored at key
.
Use hmGet()
to read field names and values.
Do not use the hash functions with several very large amounts (megabytes) of data or blobs. This is because the entire data record with all fields is always read and written. It is better to use
setValues()
andgetValues()
for large amounts of data.
Inspired by: https://docs.keydb.dev/docs/commands/#hkeys
The key must be a string.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.hmSet("key-1", {
"field-1": "value-1",
"field-2": "value-2"
})
store.hGetFields("key-1") // --> ["field-1", "field-2"]
hGetValues(key: string)
Returns the values contained in the hash stored at key
.
Use hmGet()
to read field names and values.
Do not use the hash functions with several very large amounts (megabytes) of data or blobs. This is because the entire data record with all fields is always read and written. It is better to use
setValues()
andgetValues()
for large amounts of data.
Inspired by: https://docs.keydb.dev/docs/commands/#hvals
The key must be a string.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.hmSet("key-1", {
"field-1": "value-1",
"field-2": "value-2"
})
store.hGetValues("key-1") // --> ["value-1", "value-2"]
hDelete(key: string, field: string)
Deletes a field of the map object.
undefined
if the key does not exist.true
if the field existed and was deleted.false
if the field did not exist.Do not use the hash functions with several very large amounts (megabytes) of data or blobs. This is because the entire data record with all fields is always read and written. It is better to use
setValues()
andgetValues()
for large amounts of data.
Inspired by: https://docs.keydb.dev/docs/commands/#hdel
The key must be a string.
The field name must be a string.
It is no problem at all to use several databases and access them at the same time.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
import { join } from "node:path"
const dbDir = join(__dirname, "databases")
const settingsPath = join(dbDir, "settings.sqlite")
const languagesPath = join(dbDir, "languages.sqlite")
const settingsStore = new BunSqliteKeyValue(settingsPath)
const languagesStore = new BunSqliteKeyValue(languagesPath)
// Write settings
settingsStore.set("language", "de")
settingsStore.set("page-size", "A4")
settingsStore.set("screen-position", {top: 100, left: 100})
// Write languages
languagesStore.set("de", "German")
languagesStore.set("en", "English")
languagesStore.set("it", "Italian")
// Read all settings
const settingItems = settingsStore.getItems()
console.log(settingItems) // --> [
// {key: "language", value: "de"},
// {key: "page-size", value: "A4"},
// {key: "screen-position", value: {top: 100, left: 100}},
// ]
// Read all languages
const languageValues = languagesStore.getValues()
console.log(languageValues) // --> [ "German", "English", "Italian" ]
// Read current language
const languageKey = settingsStore.get("language")
const currentLanguage = languagesStore.get(languageKey)
console.log(`Current language: "${currentLanguage}"`) // --> Current language: "German"
// Close DBs
settingsStore.close()
languagesStore.close()
Transactions can be used to combine several database statements. These combined database statements are processed much faster than if they were executed individually. The more database statements are combined, the greater the speed advantage. You can find more infos in the Bun documentation.
import { BunSqliteKeyValue } from "bun-sqlite-key-value"
const store = new BunSqliteKeyValue()
store.db.transaction(() => {
store.set("key1", "100")
store.set("key2", "200")
store.set("key3", "300")
})()
store.db.transaction(() => {
const value1 = store.get("key1")
const value2 = store.get("key2")
const value3 = store.get("key3")
const total = value1 + value2 + value3
store.set("total1", total)
})()
SQLite provides a solid and well-tested foundation. SQLite reliably takes care of saving and reading data - both for short strings and for larger BLOBs. It provides a robust foundation on which to build. Even if SQLite is not fully utilized and no relations between tables are required, this is not a disadvantage.
Please give this GitHub project a ⭐ if this project is useful to you. Thank you very much! And if you speak German, here is my business homepage: GP-Softwaretechnik Maybe you will find something interesting for you there. 😃