A node.js high performance, distributed unique id generator library for Azure storage tables. This gives Azure table storage autoincrement fields.
This is a pure javascript implementation of Tatham Oddie's awesome Snowmaker library
// Simplest usage:
var snowmaker = require('snowmaker');
snowmaker.connect(function (err) {
snowmaker.nextId("productorders", function(err,value) {
console.log("Next productorderid is "+value);
});
});
npm install snowmaker
Snowmaker uses async callbacks to connect to the table storage server and set/return identifiers. The first time a table is queried, Snowmaker will create a blob container and a blob to store the next identifier. Snowmaker will "take" a batch of identifiers all at once, which it will then dispense one at a time for every nextId() call until the batch is used. It will then get another batch. This makes the library high-performance and allows concurrency.
These methods are accessible via the snowmaker object, which is a singleton. Only one instance should be running per domain.
Connects to blob storage and creates a new blob container if it doesn't exist that will store all the identifiers for all tables. Optional connectionString is used to supply connection information. If no connection string is used, then it will connect to the local Azure storage emulator client running on your computer.
Checks to see if it has any more identifiers left to dispense and if so returns the next one. If not, it will first connect back to blob storage, and will create a blob if one doesn't exist for the specified table, and will grab another batch of identifiers and will write the max+1 batch identifier to storage.
This is the number of identifiers that Snowmaker will "reserve" at a time.
Snowmaker uses Azure's optimistic concurrency feature to make sure its write operation fails if another client running Snowmaker has already written that same batch identifier to blob storage. This ensures that no two running instances of Snowmaker get the same batches or overwrite each other. Snowmaker will retry the write operation this many times before giving up.
This is the name of the blob container that will store all the identifiers for all the tables in your Azure table storage instance.
Snowmaker will use alphanumeric identifiers by default. If this is set, then Snowmaker will convert the identifier from an int to a base-62 encoded string. This is useful if you want shorter more human friendly identifiers instead of large ints or GUIDs. Set to false to use normal ints. Keep in mind that Azure Table Storage sorts keys alphanumerically, so ints won't appear in numeric order in Table Storage.
| Type | Value |
|---|---|
| GUID | 00ab1985-0000-0000-0000-000000000000 |
| int | 11213189 |
| base62 | L33T |
Plus, large numbers are compressed into small strings using base62. For instance, if one of your table identifiers ever reached 68,289,801,377,242 then that ID would be stored as "johnhamm"!
Snowmaker will use alphanumeric identifiers with both uppercase and lowercase letters if this is set to true.
This is Tatham Oddie's diagram from his blog which explains how Snowmaker handles concurrent clients:
For more information on Snowmaker, please see this blog article on Tatham Oddie's blog.