Variant To Flattened Byte Array (G Dataflow)

Each wire and terminal on the diagram represents a data type. A type descriptor is an internal structure that keeps track of each type with a sequence of 32-bit integers.

A type can be aggregate, meaning it can consist of different types. For example, in an array of Booleans, the array is a type and its elements are another type (Boolean). All type descriptors for such complex types are stored in a list.

32 bits nTDs
        TDs
var size nTypesUsed
var size TypesUsed

• nTDs—number of elemental type descriptors contained in the buffer
• TDs—list of nTDs type descriptors
• nTypesUsed—actual number of type descriptors types used by the entity referencing the buffer
• TypesUsed—list of indexes for all the types of TDs used

Variable size fields are stored in either 16 or 32 bits, depending on the value being represented. If the value is less than 32768, it is stored in 2 bytes. If the value is greater than or equal to 32768, it is stored in 4 bytes with the high bit set to 1.

A type descriptor uses the following format:

[size][type code][type code specific info]

Note  

Some type descriptors have additional information following the type code. Arrays and clusters are structured, or aggregate, data types because they include references to other types. For example, the cluster type contains additional information about the type of each of its elements.

The following tables list numeric and non-numeric data types, type codes, and type descriptors.

n=length; x=reserved; k=number; k pstrs=number of Pascal strings; k base-exp=number of base-exponent pairs.

The minimum value in the size field of a type descriptor is 4, as shown in the table. However, any type descriptor can have a name (a Pascal string) appended, in which case the size field is larger by the length of the name rounded up to a multiple of two.

n=length; x=reserved; k=number; k pstrs=number of Pascal strings; k base-exp=number of base-exponent pairs. A dim is a 32-bit integer. The following sections have examples of data and type descriptors.

In the following example of an enumerated 8-bit integer for the items am, fm, and fm stereo, each group of characters represents a 16-bit word. The space enclosed in quotation marks (" ") represents an ASCII space.

0016 0015 0003 02a m02 fm 09f m" " st er eo

• 0016—indicates 22 bytes total
• 0015—indicates an enumerated 8-bit integer
• 0003— indicates there are three items

In the following example of a double-precision physical quantity with units m/s, each group represents a 16-bit word.

000E 001A 0002 0002 FFFF 0003 0001

• 000E—indicates 14 bytes total
• 001A—indicates this is a double-precision physical quantity
• 0002—indicates two base-exponent pairs
• 0002—denotes the seconds base index
• FFFF (-1)—the exponent of seconds
• 0003—denotes the meters base index
• 0001—the exponent of meters

The node stores all physical quantities internally in terms of base units, regardless of the units used to display them.

The following table shows the nine bases that are represented by indexes 0 through 8 for radians through candela.

The string, path, and picture data types have a 32-bit length, similar to the array dimension size. However, the only value currently encoded is FFFFFFFF (-1), which indicates variable sized. Currently, all strings, paths, and pictures are variable sized. The actual length is stored with the data.

The array and cluster data types each have their own type code. They also contain additional information about the data types of their elements and the dimensionality for arrays or number of elements for clusters.

The type code for an array is 40. A word that contains the number of dimensions of the array immediately follows the type code. Then, for each dimension, an integer contains the size in elements of that dimension. Finally, after all of the dimension sizes, the type descriptor for the element appears. The element type can be any type except an array. Currently all sizes are FFFFFFFF (-1), which means the array dimension size is variable. The node stores the actual dimension size, which is always greater than or equal to zero, with the data.

The following example is a type descriptor for a 1D array of double-precision floating-point numbers:

The following example is a type descriptor for a 2D array of Boolean values:

The type code for a cluster is 50. A word that contains the number of items in the cluster immediately follows the type code. The type descriptor for each element in cluster order follows this word. For example, consider a cluster of two integers, a 16-bit signed integer and a 32-bit unsigned integer:

Because array and cluster type descriptors contain other type descriptors, they might become deeply nested. For example, the following is a type descriptor for a multi-plot graph. The numeric types can vary.

Note that duplicate types are not included in the type list. The following example is a type descriptor list of a 1D array of double-precision floating-point numbers and a cluster with one double-precision floating-point number.

Note that the type referenced in the cluster at index 2 must be looked up in the types used array to make sense. The second element in types used is 0, which corresponds to type descriptor 0.

The type code for a waveform is 54. A word that contains the type of waveform immediately follows the type code. The type descriptor of the cluster of the waveforms elements follows this word. There are five categories of waveforms:

• Analog
• Digital
• Digital Data
• Time Stamp
• Dynamic

The following table lists the types of analog waveforms.

For example, consider a double-precision waveform:

0006 0054 0003          //size of waveform, type code, and subtype code
0080 0050 0005          //size of cluster, cluster code, and number of elements
<type descriptor of time stamp>
<type descriptor of dt> 
<type descriptor of array of element type>  
<type descriptor of unused error cluster>
<type descriptor of attributes>

The type descriptor of the other types of analog waveforms is similar to the double-precision waveform. The difference is a different type for the array element and a different subtype.

The digital waveform is the only type of waveform that has a subtype code 8.

For example:

00B6 0054 0008           //size of digital waveform, type code, and subtype code 
<type descriptor of time stamp> 
<type descriptor of dt>
<type descriptor of digital data>
<type descriptor of unused error cluster> 
<type descriptor of attributes>

Digital data is the only data type with a waveform type code and a subtype code of 7.

For example:

003E 0054 0007          //size of digital data, type code, and subtype code 
<type descriptor of transitions>
<type descriptor of data>

The time stamp has a subtype code of 6. The time stamp is a 128-bit fixed-point number with a 64-bit radix. The node stores a time stamp as a cluster of four integers where the first two signed integers (64 bits) represent the number of whole seconds after 12:00 a.m., Friday, January 1, 1904, Universal Time [01-01-1904 00:00:00]. The next two unsigned integers (64 bits) represent the fractions of seconds.

For example:

001C 0054 0006
<type descriptor of an analog waveform>