Skip to end of metadata
Go to start of metadata


In this section, you'll learn about using the Buffer object and the Codec module to interact with binary data streams.


The Titanium.Buffer is a container for bytes. It differs from Titanium.Blob in that a `Buffer` is mutable and resizable.

The Titanium.Codec provides utility methods to convert to and from binary data.

Buffer and Codec are primarily used in conjunction with Titanium.IOStream objects, including Titanium.Network.Socket.TCP and Titanium.Filesystem.FileStream.

The Titanium.Stream module provides additional utility methods for working with streams.


Buffers are created from the Titanium module using Titanium.createBuffer(). Buffers are similar to byte arrays in other programming languages but have the ability to be resized. Buffers may be grown by setting the length property to a larger value, or when using the insert() or append() methods. Their length may also be reduced by setting the length property to a smaller value or using the release() method.

You can create an empty buffer with a specified initial size, or use a string or numeric value as the initial value of a buffer. The following code excerpt show several ways to initialize a buffer.

The following code excerpt shows some other manipulations on a buffer, including appending one buffer to another and truncating buffers.

Buffers can also be addressed directly as if they were arrays of byte values, as shown in the following code excerpt.

The Buffer object's append, copy, and insert methods resize the buffer as needed. However, when you access a buffer as an array, the buffer is not automatically extended like a JavaScript array. For example, the following code throws an exception, because the value is being assigned to a position past the end of the buffer.

Also note that when assigning values to the buffer this way, the values must be byte values.


The Titanium.Codec module provides methods for encoding and decoding binary data from buffers into JavaScript primitives. The Codec module supports conversion of two basic primitive types: Numbers and Strings. Codec also handles byte order conversion.

The Codec module methods do not resize the target buffer.

Numeric conversions

Numeric conversion is provided for the following types: byte, short, int, long, float, and double.

The following code excerpt encodes a 4-byte integer into the buffer, starting at index 10 (the 11th byte of the integer).

The following code excerpt decodes the 4-byte integer written to the buffer in the previous example.

String conversions

String conversion is provided using the following character sets: ISO_LATIN1, UTF8, UTF16, UTF16BE, and UTF16LE. The default is UTF-8. Titanium.Codec also suports both Big Endian and Little Endian byte ordering.

The following excerpt encodes a string into a buffer, then trims the buffer to the length of the encoded string.

The following excerpt decodes a UTF-8 string from a buffer.

Buffer creation with encode

Titanium.createBuffer() has convenience options for creating a buffer from existing data. These options are just a convenient alternative to calling `createBuffer` and `encodeString` or `encodeNumber` separately, as shown in the following excerpt.