SnappyDB

SnappyDB is a key-value database for Android it's an alternative for SQLite if you want to use a NoSQL approach.

It allows you to store and get primitive types, but also a Serializable object or array in a type-safe way.

SnappyDB can outperform SQLite in read/write operations. benchmark

SnappyDB is based on leveldb and use snappy compression algorithm, on redundant content you could achieve a good compression ratio

Check out the Demo App

Usage

try {
   DB snappydb = DBFactory.open(context); //create or open an existing database using the default name
   
   snappydb.put("name", "Jack Reacher"); 
   snappydb.putInt("age", 42);  
   snappydb.putBoolean("single", true);
   snappydb.put("books", new String[]{"One Shot", "Tripwire", "61 Hours"}); 
   
   String 	 name   =  snappydb.get("name");
   int 	   age    =  snappydb.getInt("age");
   boolean  single =  snappydb.getBoolean("single");
   String[] books  =  snappydb.getArray("books", String.class);// get array of string
   	
   snappydb.close();
   
   } catch (SnappydbException e) {
   }

For more recipes please take a look at the Cookbook.

With SnappyDB you could seamlessly store and retrieve your object/array, it uses Kryo serialization which is faster than the regular Java serialization.

Installation

SnappyDB uses native code for performance, it's available as an Android Library Project AAR.

dependencies {
    compile 'com.snappydb:snappydb-lib:0.5.2'
    compile 'com.esotericsoftware.kryo:kryo:2.24.0'
}

Manual:

Download JAR file and other folders from here
Put all the files and folders in the libs subfolder of your Android project

libs
├───|── snappydb-0.5.2.jar
    |── armeabi
    │   └── libsnappydb-native.so
    ├── armeabi-v7a
    │   └── libsnappydb-native.so
    ├── mips
    │   └── libsnappydb-native.so
    └── x86
        └── libsnappydb-native.so

Cookbook

Common tasks snippets

Open/Create database
Close database
Destroy database
Insert primitive types
Read primitive types
Insert Serializable
Insert Object
Read Serializable
Read Object
Insert Array
Read Array
Check key
Delete key
Keys Search
Keys Count
Iterators

Create database

Create using the default name

 DB snappydb = DBFactory.open(context);

Create with a given name

 DB snappydb = DBFactory.open(context, "books");

SnappyDB use the internal storage to create your database. It creates a directory containing all the necessary files Ex: /data/data/com.snappydb/files/mydatabse

Using the builder pattern

 DB snappyDB = new SnappyDB.Builder(context)
                    .directory(Environment.getExternalStorageDirectory().getAbsolutePath()) //optional
                    .name("books")//optional
                    .build();

directory Specify the location of the database (sdcard in this example)

Close database

snappydb.close();

Destroy database

snappydb.destroy();

Insert primitive types

snappyDB.put("quote", "bazinga!");

snappyDB.putShort("myshort", (short)32768);

snappyDB.putInt("max_int", Integer.MAX_VALUE);

snappyDB.putLong("max_long", Long.MAX_VALUE);

snappyDB.putDouble("max_double", Double.MAX_VALUE);

snappyDB.putFloat("myfloat", 10.30f);

snappyDB.putBoolean("myboolean", true);

Read primitive types

String quote      = snappyDB.get("quote");

short myshort     = snappyDB.getShort("myshort");

int maxInt        = snappyDB.getInt("max_int");

long maxLong      = snappyDB.getLong("max_long");

double maxDouble  = snappyDB.getDouble("max_double");

float myFloat     = snappyDB.getFloat("myfloat");

boolean myBoolean = snappyDB.getBoolean("myboolean");

Insert Serializable

AtomicInteger objAtomicInt = new AtomicInteger (42);
snappyDB.put("atomic integer", objAtomicInt);

Insert Object

MyPojo pojo = new MyPojo ();
snappyDB.put("my_pojo", pojo);

Note: MyPojo doesn't have to implement Serializable interface

Read Serializable

 AtomicInteger myObject = snappyDB.get("atomic integer", AtomicInteger.class);

Read Object

MyPojo myObject = snappyDB.getObject("non_serializable", MyPojo.class);

Note: MyPojo doesn't have to implement Serializable interface

Insert Array

Number[] array = {new AtomicInteger (42), new BigDecimal("10E8"), Double.valueOf(Math.PI)};

snappyDB.put("array", array);

Read Array

Number [] numbers = snappyDB.getObjectArray("array", Number.class);

Check Key

boolean isKeyExist = snappyDB.exists("key");

Delete Key

snappyDB.del("key");

Keys Search

By Prefix

snappyDB.put("android:03", "Cupcake"); // adding 0 to maintain lexicographical order
snappyDB.put("android:04", "Donut");
snappyDB.put("android:05", "Eclair");
snappyDB.put("android:08", "Froyo");
snappyDB.put("android:09", "Gingerbread");
snappyDB.put("android:11", "Honeycomb");
snappyDB.put("android:14", "Ice Cream Sandwich");
snappyDB.put("android:16", "Jelly Bean");
snappyDB.put("android:19", "KitKat");

String [] keys = snappyDB.findKeys("android");
assert keys.length == 9;

keys = snappyDB.findKeys("android:0");
assert keys.length == 5;

assert snappyDB.get(keys[0]).equals("Cupcake");
assert snappyDB.get(keys[1]).equals("Donut");
assert snappyDB.get(keys[2]).equals("Eclair");
assert snappyDB.get(keys[3]).equals("Froyo");
assert snappyDB.get(keys[4]).equals("Gingerbread");

keys = snappyDB.findKeys("android:1");
assert keys.length == 4;

assert snappyDB.get(keys[0]).equals("Honeycomb");
assert snappyDB.get(keys[1]).equals("Ice Cream Sandwich");
assert snappyDB.get(keys[2]).equals("Jelly Bean");
assert snappyDB.get(keys[3]).equals("KitKat");

By Range [from .. to]

both 'FROM' & 'TO' keys exist

keys = snappyDB.findKeysBetween("android:08", "android:11");
assertEquals(3, keys.length);
assertEquals("android:08", keys[0]);
assertEquals("android:09", keys[1]);
assertEquals("android:11", keys[2]);

'FROM' key exist, but not the `TO

keys = snappyDB.findKeysBetween("android:05", "android:10");
assertEquals(3, keys.length);
assertEquals("android:05", keys[0]);
assertEquals("android:08", keys[1]);
assertEquals("android:09", keys[2]);

'FROM' key doesn't exist but the 'TO' key do

keys = snappyDB.findKeysBetween("android:07", "android:09");
assertEquals(2, keys.length);
assertEquals("android:08", keys[0]);
assertEquals("android:09", keys[1]);

both 'FROM' & 'TO' keys doesn't exist

keys = snappyDB.findKeysBetween("android:13", "android:99");
assertEquals(3, keys.length);
assertEquals("android:14", keys[0]);
assertEquals("android:16", keys[1]);
assertEquals("android:19", keys[2]);

With offset and limit

//return all keys starting with "android" after the first 5
keys = snappyDB.findKeys("android", 5);
assertEquals(4, keys.length);
assertEquals("android:11", keys[0]);
assertEquals("android:14", keys[1]);
assertEquals("android:16", keys[2]);
assertEquals("android:19", keys[3]);

//return 3 first keys starting with "android"
keys = snappyDB.findKeys("android", 0, 3);
assertEquals(3, keys.length);
assertEquals("android:03", keys[0]);
assertEquals("android:04", keys[1]);
assertEquals("android:05", keys[2]);

//return the fourth key starting with "android" (offset 3, limit 1)
keys = snappyDB.findKeys("android", 3, 1);
assertEquals(1, keys.length);
assertEquals("android:08", keys[0]);

//return the two first keys between android:14 and android:99
keys = snappyDB.findKeysBetween("android:14", "android:99", 0, 2);
assertEquals(2, keys.length);
assertEquals("android:14", keys[0]);
assertEquals("android:16", keys[1]);

//return the third key (offset 2, limit 1) after android:10 before android:99
keys = snappyDB.findKeysBetween("android:10", "android:99", 2, 1);
assertEquals(1, keys.length);
assertEquals("android:16", keys[0]);

Keys Count

Counting is quicker than extracting values (if you don't need them). Especially on very big collections.

By Prefix

assertEquals(9, snappyDB.countKeys("android"));
assertEquals(5, snappyDB.countKeys("android:0"));

By Range [from .. to]

assertEquals(3, snappyDB.countKeysBetween("android:08", "android:11"));
assertEquals(3, snappyDB.countKeysBetween("android:13", "android:99"));

Iterators

Each time you use the offset & limit arguments, the engine makes the query and then scrolls to your offset. Which means that the bigger the offset is, the longer the query will take. This is not a problem on small collections, but on very large collections, it is.

An iterator keeps it's position in the key collection and can be asked for the next key at any time. It is therefore better to use an iterator on very large collections.

Iterators work on DB snapshot, which means that if you add or delete value in / from the DB, the iterators will not see those changes.

Please note that iterators given by the SnappyDB are closeable and need to be closed once finishe

SnappyDB

Install / Use

README

SnappyDB

Usage

Installation

Cookbook

Create database

Create using the default name

Create with a given name

Using the builder pattern

Close database

Destroy database

Insert primitive types

Read primitive types

Insert Serializable

Insert Object

Read Serializable

Read Object

Insert Array

Read Array

Check Key

Delete Key

Keys Search

By Prefix

By Range [from .. to]

With offset and limit

Keys Count

By Prefix

By Range [from .. to]

Iterators