LevelDOWN (RocksDB backend)

A Low-level Node.js RocksDB binding

RocksDB

This is a RocksDB fork of LevelDOWN. It uses almost the same exact code as LevelDOWN. The only thing that has changed is the build process. The original LevelDOWN resides here: https://github.com/Level/leveldown.

LevelDOWN

LevelDOWN was extracted from LevelUP and now serves as a stand-alone binding for LevelDB.

It is strongly recommended that you use LevelUP in preference to LevelDOWN unless you have measurable performance reasons to do so. LevelUP is optimised for usability and safety. Although we are working to improve the safety of the LevelDOWN interface it is still easy to crash your Node process if you don't do things in just the right way.

See the section on safety below for details of known unsafe operations with LevelDOWN.

Tested & supported platforms

• Linux (including ARM platforms such as Raspberry Pi and Kindle!)
• Mac OS
• Solaris (SmartOS & Nodejitsu)
• FreeBSD
• Windows
  • Node 0.10 and above only, see issue #5 for more info
  • See installation instructions for node-gyp dependencies here, you'll need these (free) components from Microsoft to compile and run any native Node add-on in Windows.

API

• leveldown()
• leveldown#open()
• leveldown#close()
• leveldown#put()
• leveldown#get()
• leveldown#del()
• leveldown#batch()
• leveldown#approximateSize()
• leveldown#getProperty()
• leveldown#iterator()
• iterator#next()
• iterator#seek()
• iterator#end()
• leveldown.destroy()
• leveldown.repair()

leveldown(location)

leveldown() returns a new LevelDOWN instance. location is a String pointing to the LevelDB location to be opened.

leveldown#open([options, ]callback)

open() is an instance method on an existing database object.

The callback function will be called with no arguments when the database has been successfully opened, or with a single error argument if the open operation failed for any reason.

options

The optional options argument may contain:

• 'createIfMissing' (boolean, default: true): If true, will initialise an empty database at the specified location if one doesn't already exist. If false and a database doesn't exist you will receive an error in your open() callback and your database won't open.

• 'errorIfExists' (boolean, default: false): If true, you will receive an error in your open() callback if the database exists at the specified location.

• 'compression' (boolean, default: true): If true, all compressible data will be run through the Snappy compression algorithm before being stored. Snappy is very fast and shouldn't gain much speed by disabling so leave this on unless you have good reason to turn it off.

• 'cacheSize' (number, default: 8 * 1024 * 1024 = 8MB): The size (in bytes) of the in-memory LRU cache with frequently used uncompressed block contents.

Advanced options

The following options are for advanced performance tuning. Modify them only if you can prove actual benefit for your particular application.

• 'writeBufferSize' (number, default: 4 * 1024 * 1024 = 4MB): The maximum size (in bytes) of the log (in memory and stored in the .log file on disk). Beyond this size, LevelDB will convert the log data to the first level of sorted table files. From the LevelDB documentation:

Larger values increase performance, especially during bulk loads. Up to two write buffers may be held in memory at the same time, so you may wish to adjust this parameter to control memory usage. Also, a larger write buffer will result in a longer recovery time the next time the database is opened.

• 'blockSize' (number, default 4096 = 4K): The approximate size of the blocks that make up the table files. The size related to uncompressed data (hence "approximate"). Blocks are indexed in the table file and entry-lookups involve reading an entire block and parsing to discover the required entry.

• 'maxOpenFiles' (number, default: 1000): The maximum number of files that LevelDB is allowed to have open at a time. If your data store is likely to have a large working set, you may increase this value to prevent file descriptor churn. To calculate the number of files required for your working set, divide your total data by 2MB, as each table file is a maximum of 2MB.

• 'blockRestartInterval' (number, default: 16): The number of entries before restarting the "delta encoding" of keys within blocks. Each "restart" point stores the full key for the entry, between restarts, the common prefix of the keys for those entries is omitted. Restarts are similar to the concept of keyframes in video encoding and are used to minimise the amount of space required to store keys. This is particularly helpful when using deep namespacing / prefixing in your keys.

leveldown#close(callback)

close() is an instance method on an existing database object. The underlying LevelDB database will be closed and the callback function will be called with no arguments if the operation is successful or with a single error argument if the operation failed for any reason.

leveldown#put(key, value[, options], callback)

put() is an instance method on an existing database object, used to store new entries, or overwrite existing entries in the LevelDB store.

The key and value objects may either be Strings or Node.js Buffer objects. Other object types are converted to JavaScript Strings with the toString() method. Keys may not be null or undefined and objects converted with toString() should not result in an empty-string. Values of null, undefined, '', [] and new Buffer(0) (and any object resulting in a toString() of one of these) will be stored as a zero-length character array and will therefore be retrieved as either '' or new Buffer(0) depending on the type requested.

A richer set of data-types are catered for in LevelUP.

options

The only property currently available on the options object is 'sync' (boolean, default: false). If you provide a 'sync' value of true in your options object, LevelDB will perform a synchronous write of the data; although the operation will be asynchronous as far as Node is concerned. Normally, LevelDB passes the data to the operating system for writing and returns immediately, however a synchronous write will use fsync() or equivalent so your callback won't be triggered until the data is actually on disk. Synchronous filesystem writes are significantly slower than asynchronous writes but if you want to be absolutely sure that the data is flushed then you can use 'sync': true.

The callback function will be called with no arguments if the operation is successful or with a single error argument if the operation failed for any reason.

leveldown#get(key[, options], callback)

get() is an instance method on an existing database object, used to fetch individual entries from the LevelDB store.

The key object may either be a String or a Node.js Buffer object and cannot be undefined or null. Other object types are converted to JavaScript Strings with the toString() method and the resulting String may not be a zero-length. A richer set of data-types are catered for in LevelUP.

Values fetched via get() that are stored as zero-length character arrays (null, undefined, '', [], new Buffer(0)) will return as empty-String ('') or new Buffer(0) when fetched with asBuffer: true (see below).

options

The optional options object may contain:

• 'fillCache' (boolean, default: true): LevelDB will by default fill the in-memory LRU Cache with data from a call to get. Disabling this is done by setting fillCache to false.

• 'asBuffer' (boolean, default: true): Used to determine whether to return the value of the entry as a String or a Node.js Buffer object. Note that converting from a Buffer to a String incurs a cost so if you need a String (and the value can legitimately become a UTF8 string) then you should fetch it as one with asBuffer: true and you'll avoid this conversion cost.

The callback function will be called with a single error if the operation failed for any reason. If successful the first argument will be null and the second argument will be the value as a String or Buffer depending on the asBuffer option.

leveldown#del(key[, options], callback)

del() is an instance method on an existing database object, used to delete entries from the LevelDB store.

The key object may either be a String or a Node.js Buffer object and cannot be undefined or null. Other object types are converted to JavaScript Strings with the toString() method and the resulting String may not be a zero-length. A richer set of data-types are catered for in LevelUP.

options

The only property currently available on the options object is 'sync' (boolean, default: false). See leveldown#put() for details about this option.

The callback function will be called with no arguments if the operation is successful or with a single error argument if the operation failed for any reason.

leveldown#batch(operations[, options], callback)

batch() is an instance method on an existing database object. Used for very fast bulk-write operations (both put and delete). The operations argument should be an Array containing a list of operations to be executed sequentially, although as a whole they are performed as an atomic operation inside LevelDB.

Each operation is contained in an object having the following properties: type, key, value, where the type is either 'put' or 'del'. In the case of 'del' the 'value' property is ignored. Any entries with a 'key' of null or undefined will cause an error to be returned on the callback. Any entries where the type is 'put' that have a 'value' of undefined, null, [], '' or new Buffer(0) will be stored as a zero-length character array and therefore be fetched during reads as either '' or new Buffer(0) depending on how they are requested.

See LevelUP for full documentation on how this works in practice.

options

The only property currently available on the options object is 'sync' (boolean, default: false). See leveldown#put() for details about this option.

The callback function will be called with no arguments if the operation is successful or with a single error argument if the operation failed for any reason.

leveldown#approximateSize(start, end, callback)

approximateSize() is an instance method on an existing database object. Used to get the approximate number of bytes of file system space used by the range [start..end). The result may not include recently written data.

The start and end parameters may be either String or Node.js Buffer objects representing keys in the LevelDB store.

The callback function will be called with no arguments if the operation is successful or with a single error argument if the operation failed for any reason.

leveldown#getProperty(property)

getProperty can be used to get internal details from LevelDB. When issued with a valid property string, a readable string will be returned (this method is synchronous).

Currently, the only valid properties are:

• 'leveldb.num-files-at-levelN': return the number of files at level N, where N is an integer representing a valid level (e.g. "0").

• 'leveldb.stats': returns a multi-line string describing statistics about LevelDB's internal operation.

• 'leveldb.sstables': returns a multi-line string describing all of the sstables that make up contents of the current database.

leveldown#iterator([options])

iterator() is an instance method on an existing database object. It returns a new Iterator instance.

options

The optional options object may contain:

• 'gt' (greater than), 'gte' (greater than or equal) define the lower bound of the values to be fetched and will determine the starting point where 'reverse' is not true. Only records where the key is greater than (or equal to) this option will be included in the range. When 'reverse' is true the order will be reversed, but the records returned will be the same.

• 'lt' (less than), 'lte' (less than or equal) define the higher bound of the range to be fetched and will determine the starting point where 'reverse' is not true. Only records where the key is less than (or equal to) this option will be included in the range. When 'reverse' is true the order will be reversed, but the records returned will be the same.

• 'start', 'end' legacy ranges - instead use 'gte', 'lte'

• 'reverse' (boolean, default: false): a boolean, set to true if you want the stream to go in reverse order. Beware that due to the way LevelDB works, a reverse seek will be slower than a forward seek.

• 'keys' (boolean, default: true): whether the callback to the next() method should receive a non-null key. There is a small efficiency gain if you ultimately don't care what the keys are as they don't need to be converted and copied into JavaScript.

• 'values' (boolean, default: true): whether the callback to the next() method should receive a non-null value. There is a small efficiency gain if you ultimately don't care what the values are as they don't need to be converted and copied into JavaScript.

• 'limit' (number, default: -1): limit the number of results collected by this iterator. This number represents a maximum number of results and may not be reached if you get to the end of the store or your 'end' value first. A value of -1 means there is no limit.

• 'fillCache' (boolean, default: false): whether LevelDB's LRU-cache should be filled with data read.

• 'keyAsBuffer' (boolean, default: true): Used to determine whether to return the key of each entry as a String or a Node.js Buffer object. Note that converting from a Buffer to a String incurs a cost so if you need a String (and the value can legitimately become a UTF8 string) then you should fetch it as one.

• 'valueAsBuffer' (boolean, default: true): Used to determine whether to return the value of each entry as a String or a Node.js Buffer object.

iterator#next(callback)

next() is an instance method on an existing iterator object, used to increment the underlying LevelDB iterator and return the entry at that location.

the callback function will be called with no arguments in any of the following situations:

• the iterator comes to the end of the store
• the end key has been reached; or
• the limit has been reached

Otherwise, the callback function will be called with the following 3 arguments:

• error - any error that occurs while incrementing the iterator.
• key - either a String or a Node.js Buffer object depending on the keyAsBuffer argument when the iterator() was called.
• value - either a String or a Node.js Buffer object depending on the valueAsBuffer argument when the iterator() was called.

iterator#seek(key)

seek() is an instance method on an existing iterator object, used to seek the underlying LevelDB iterator to a given key.

By calling seek(key), subsequent calls to next(cb) will return key/values larger or smaller than key, based on your reverse setting in the iterator constructor.

iterator#end(callback)

end() is an instance method on an existing iterator object. The underlying LevelDB iterator will be deleted and the callback function will be called with no arguments if the operation is successful or with a single error argument if the operation failed for any reason.

leveldown.destroy(location, callback)

destroy() is used to completely remove an existing LevelDB database directory. You can use this function in place of a full directory rm if you want to be sure to only remove LevelDB-related files. If the directory only contains LevelDB files, the directory itself will be removed as well. If there are additional, non-LevelDB files in the directory, those files, and the directory, will be left alone.

The callback will be called when the destroy operation is complete, with a possible error argument.

leveldown.repair(location, callback)

repair() can be used to attempt a restoration of a damaged LevelDB store. From the LevelDB documentation:

If a DB cannot be opened, you may attempt to call this method to resurrect as much of the contents of the database as possible. Some data may be lost, so be careful when calling this function on a database that contains important information.

You will find information on the repair operation in the LOG file inside the store directory.

A repair() can also be used to perform a compaction of the LevelDB log into table files.

The callback will be called when the repair operation is complete, with a possible error argument.

Safety

Database state

Currently LevelDOWN does not track the state of the underlying LevelDB instance. This means that calling open() on an already open database may result in an error. Likewise, calling any other operation on a non-open database may result in an error.

LevelUP currently tracks and manages state and will prevent out-of-state operations from being send to LevelDOWN. If you use LevelDOWN directly then you must track and manage state for yourself.

Snapshots

LevelDOWN exposes a feature of LevelDB called snapshots. This means that when you do e.g. createReadStream and createWriteStream at the same time, any data modified by the write stream will not affect data emitted from the read stream. In other words, a LevelDB Snapshot captures the latest state at the time the snapshot was created, enabling the snapshot to iterate or read the data without seeing any subsequent writes. Any read not performed on a snapshot will implicitly use the latest state.

Getting support

There are multiple ways you can find help in using LevelDB in Node.js:

• IRC: you'll find an active group of LevelUP users in the ##leveldb channel on Freenode, including most of the contributors to this project.
• Mailing list: there is an active Node.js LevelDB Google Group.
• GitHub: you're welcome to open an issue here on this GitHub repository if you have a question.

Contributing

LevelDOWN is an OPEN Open Source Project. This means that:

Individuals making significant and valuable contributions are given commit-access to the project to contribute as they see fit. This project is more like an open wiki than a standard guarded open source project.

See the contribution guide for more details.

Windows

A large portion of the Windows support comes from code by Krzysztof Kowalczyk @kjk, see his Windows LevelDB port here. If you're using LevelUP on Windows, you should give him your thanks!

Prebuilt binaries

LevelDOWN uses prebuild to support prebuilt binaries. See this list of supported prebuilt platform binaries. When installing LevelDOWN prebuild --download will download prebuilt binaries from GitHub if they exist and fallback to a compile step if they don't.

If you are working on LevelDOWN and want to compile the C++ code you can do:

$ npm run rebuild [--debug] [--verbose]

or

$ npm i --build-from-source [--debug] [--verbose]

License & copyright

Copyright © 2012-2016 LevelDOWN contributors.

LevelDOWN is licensed under the MIT license. All rights not explicitly granted in the MIT license are reserved. See the included LICENSE.md file for more details.

LevelDOWN builds on the excellent work of the LevelDB and Snappy teams from Google and additional contributors. LevelDB and Snappy are both issued under the New BSD Licence.