LabVIEW osnove vol.

2
TM


Lesson 1
Common Design Techniques
A. Design Patterns
B. Single Loop Design Patterns
C. Multiple Loop Design Patterns
D. Timing a Design Pattern

Lesson 2
Synchronization Techniques
A. Variables
B. Notifiers
C. Queues

Lesson 3
Event Programming
A. Events
B. Event-Driven Programming
C. Caveats and Recommendations
D. Event-Based Design Patterns

Lesson 4
Error Handling
A. Importance of Error Handling
B. Detect and Report Errors
C. Errors and Warnings
D. Ranges of Error Codes
E. Error Handlers

Lesson 5
Controlling the User Interface
A. Property Nodes
B. Invoke Nodes
C. VI Server Architecture
D. Control References

© National Instruments Corporation iii LabVIEW Core 2 Course Manual

Contents

Lesson 6
File I/O Techniques
A. File Formats
B. Binary Files
C. TDMS Files

Lesson 7
Improving an Existing VI
A. Refactoring Inherited Code
B. Typical Refactoring Issues
C. Comparing VIs

Lesson 8
Creating and Distributing Applications
A. Preparing the Files
B. Build Specifications
C. Building the Application and Installer

Appendix A
Using Variables
A. Parallelism
B. Variables
C. Functional Global Variables
D. Race Conditions

LabVIEW Core 2 Course Manual iv ni.com

Common Design Techniques
1
You can develop better programs in LabVIEW and in other programming
languages if you follow consistent programming techniques. Design
patterns represent techniques that have proved themselves useful time and
time again. To facilitate development, LabVIEW provides templates for
several common design patterns. This lesson discusses two different
categories of programming design patterns—single loop and multiple loops.

Single loop design patterns include the simple VI, the general VI, and the
state machine.

Multiple loop design patterns include the parallel loop VI, the master/slave,
and the producer/consumer.

Understanding the appropriate use of each design pattern helps you create
more efficient LabVIEW VIs.

Topics
A. Design Patterns
B. Single Loop Design Patterns
C. Multiple Loop Design Patterns
D. Timing a Design Pattern

© National Instruments Corporation 1-1 LabVIEW Core 2 Course Manual

Lesson 1 Common Design Techniques

A. Design Patterns
Application design patterns represent LabVIEW code implementations and
techniques that are solutions to specific problems in software design. Design
patterns typically evolve through the efforts of many developers and are
fine-tuned for simplicity, maintainability, and readability. Design patterns
represent the techniques that have proved themselves useful over time.
Furthermore, as a pattern gains acceptance, it becomes easier to
recognize—this recognition alone helps you to read and make changes
to your code.

LabVIEW includes several built-in VI templates for several standard design

patterns that you will learn about in later lessons. To access the design
patterns, select File»New to display the New dialog box. The design
patterns are available in the VI»From Template»Frameworks»Design
Patterns folder. Listed below are the different kinds of design patterns.
• Master/Slave Design Pattern—Use this template to build a master/slave
design pattern. The master loop always executes. It notifies one or more
slave loops to execute their code. The slave loop or loops continue
executing until they complete, then wait for another notification.
Contrast this with the producer/consumer pattern in which the consumer
loops execute only when they have data in their queue.
• Producer/Consumer Design Pattern (Data)—Use this template to build
a producer/consumer design pattern. Use this template when you need
to execute a process, such as data analysis, when a data source, such as
a triggered acquisition, produces data at an uneven rate and you need to
execute the process when the data becomes available.
• Producer/Consumer Design Pattern (Events)—Use this template to
build a producer/consumer design pattern with events to produce queue
items. Use this design pattern instead of the user interface event handler
pattern for user interfaces when you want to execute code
asynchronously in response to an event without slowing the user
interface responsiveness.
• Queued Message Handler—Use this template to build a queued message
handler design pattern, in which each message handling code can queue
any number of new messages.
• Standard State Machine—Use this template to build a standard state
machine design pattern. Each state executes code and determines which
state to transition to. Contrast this design pattern with the user interface
event pattern, in which code executes in response to user actions.
Contrast this design pattern also with the queued message handler
pattern, in which each message handling code can queue any number of
messages.
4

LabVIEW Core 2 Course Manual 1-2 ni.com

 Lesson 1 Common Design Techniques

• User Interface Event Handler—Use this template to build a user

interface event handler design pattern. Use this pattern for dialog boxes
and other user interfaces in which code executes in response to user
actions. You also can create and execute user-defined events that the VI
can handle the same way as user interface events.

B. Single Loop Design Patterns

You learned to design three different types of design patterns
the simple architecture, the general architecture,
and the state machine.

Simple VI Design Patterns

When performing calculations or making quick lab measurements, you do
not need a complicated architecture. Your program might consist of a single
VI that takes a measurement, performs calculations, and either displays the
results or records them to disk. The simple VI design pattern usually does
not require a specific start or stop action from the user. The user just clicks
the Run button. Use this architecture for simple applications or for
functional components within larger applications. You can convert these
simple VIs into subVIs that you use as building blocks for larger
applications.

Figure 1-1 displays the block diagram of the Determine Warnings VI
This VI performs a single task—it determines what warning to output
dependent on a set of inputs. You can use this VI as a subVI whenever
you must determine thewarning level.

Notice that the VI in Figure 1-1 contains no start or stop actions from the
user. In this VI all block diagram objects are connected through data flow.
You can determine the overall order of operations by following the flow of
data. For example, the Not Equal function cannot execute until the Greater
Than or Equal function, the Less Than or Equal function, and both Select
functions have executed.

© National Instruments Corporation 1-3 LabVIEW Core 2 Course Manual

Lesson 1 Common Design Techniques

Figure 1-1. Simple VI Architecture

General VI Design Patterns

A general VI design pattern has three main phases—startup, main
application, and shutdown. Each of the following phases may contain code
that uses another type of design pattern.
• Startup—Initializes hardware, reads configuration information from
files, or prompts the user for data file locations.
• Main Application—Consists of at least one loop that repeats until the
user decides to exit the program or the program terminates for other
reasons such as I/O completion.
• Shutdown—Closes files, writes configuration information to disk, or
resets I/O to the default state.

Figure 1-2 shows the general VI design pattern.

Figure 1-2. General VI Design Pattern

LabVIEW Core 2 Course Manual 1-4 ni.com

 Lesson 1 Common Design Techniques

In Figure 1-2, the error cluster wires control the execution order of the three
sections. The While Loop does not execute until the Start Up VI finishes
running and returns the error cluster data. Consequently, the Shut Down VI
cannot run until the main application in the While Loop finishes and the
error cluster data leaves the loop.

Tip Most loops require a Wait function, especially if that loop monitors user input on
the front panel. Without the Wait function, the loop might run continuously and use all of
the computer system resources. The Wait function forces the loop to run asynchronously
even if you specify 0 milliseconds as the wait period. If the operations inside the main
loop react to user inputs, you can increase the wait period to a level acceptable for
reaction times. A wait of 100 to 200 ms is usually good because most users cannot detect
that amount of delay between clicking a button on the front panel and the subsequent
event execution.

For simple applications, the main application loop is obvious and contains
code that uses the simple VI design pattern. When the application incudes
complicated user interfaces or multiple tasks such as user actions, I/O
triggers, and so on, the main application phase gets more complicated.

State Machine Design Pattern

The state machine design pattern is a modification of the general design
pattern. It usually has a start up and shut down phase. However, the main
application phase consists of a Case structure embedded in the loop. This
architecture allows you to run different code each time the loop executes,
depending upon some condition. Each case defines a state of the machine,
hence the name, state machine. Use this design pattern for VIs that are easily
divided into several simpler tasks, such as VIs that act as a user interface.

A state machine in LabVIEW consists of a While Loop, a Case structure,

and a shift register. Each state of the state machine is a separate case in
the Case structure. You place VIs and other code that the state should
execute within the appropriate case. A shift register stores the state that
should execute upon the next iteration of the loop. The block diagram of a
state machine VI with five states appears in Figure 1-3. Figure 1-4 shows the
other cases, or states, of the state machine.

© National Instruments Corporation 1-5 LabVIEW Core 2 Course Manual

Lesson 1 Common Design Techniques

Figure 1-3. State Machine with Startup State

Figure 1-4. Idle (Default), Event 1, Event 2, and Shutdown States

In the state machine design pattern, you design the list of possible tasks, or
states, and then map them to each case. For the VI in the previous example,
the possible states are Startup, Idle, Event 1, Event 2, and Shutdown. An
enumerated constant stores8 the states. Each state has its own case in the Case

LabVIEW Core 2 Course Manual 1-6 ni.com

 Lesson 1 Common Design Techniques

structure. The outcome of one case determines which case to execute next.
The shift register stores the value that determines which case to execute next.

The state machine design pattern can make the block diagram much smaller,
and therefore, easier to read and debug. Another advantage of the state
machine architecture is that each case determines the next state, unlike
Sequence structures that must execute every frame in sequence.

A disadvantage of the state machine design pattern is that with the approach
in the previous example, it is possible to skip states. If two states in the
structure are called at the same time, this model handles only one state, and
the other state does not execute. Skipping states can lead to errors that are
difficult to debug because they are difficult to reproduce. More complex
versions of the state machine design pattern contain extra code that creates
a queue of events, or states, so that you do not miss a state. Refer to
Lesson 2, Synchronization Techniques, for more information about
queue-based state machines.

C. Multiple Loop Design Patterns

This section describes the following multiple loop design patterns—parallel
loop, master/slave, and producer/consumer data.

Parallel Loop Design Pattern

Some applications must respond to and run several tasks concurrently.
One way to improve parallelism is to assign a different loop to each task.
For example, you might have a different loop for each button on the front
panel and for every other kind of task, such as a menu selection, I/O trigger,
and so on. Figure 1-5 shows this parallel loop design pattern.

Figure 1-5. Parallel Loop Design Pattern

9

© National Instruments Corporation 1-7 LabVIEW Core 2 Course Manual

Lesson 1 Common Design Techniques

This structure is straightforward and appropriate for some simple menu VIs,
where you expect a user to select from one of several buttons that perform
different actions. The parallel loop design pattern lets you handle multiple,
simultaneous, independent tasks. In this design pattern, responding to
one action does not prevent the VI from responding to another action. For
example, if a user clicks a button that displays a dialog box, parallel loops
can continue to respond to I/O tasks.

However, the parallel loop design pattern requires you to coordinate and
communicate between different loops. You cannot use wires to pass data
between loops because doing so prevents the loops from running in parallel.
Instead, you must use a messaging technique for passing information among
processes. Refer to Appendix A, Using Variables, for more information
about using local variables to message among parallel loops. Refer to
Lesson 2, Synchronization Techniques, for messaging techniques using
notifiers and queues.

Master/Slave Design Pattern

The master/slave design pattern consists of multiple parallel loops. Each of
the loops may execute tasks at different rates. One loop acts as the master,
and the other loops act as slaves. The master loop controls all the slave loops
and communicates with them using messaging techniques, as shown in
Figure 1-6.

Figure 1-6. Master/Slave Design Pattern

Use the master/slave design pattern when you need a VI to respond to user
interface controls while simultaneously collecting data. For example, you
want to create a VI that measures and logs a slowly changing voltage once
every five seconds. The VI acquires a waveform from a transmission line
and displays it on a graph every 100 ms. The VI also provides a user
interface that allows the user to change parameters for each acquisition. The
master/slave design pattern is well suited for this acquisition application.
For this application, the master loop contains the user interface. The voltage
10

LabVIEW Core 2 Course Manual 1-8 ni.com

 Lesson 1 Common Design Techniques

acquisition occurs in one slave loop, while the graphing occurs in another
slave loop.

Using the standard master/slave design pattern approach to this VI, you
would put the acquisition processes in two separate While Loops, both of
them driven by a master loop that receives inputs from the user interface
controls. This ensures that the separate acquisition processes do not affect
each other, and that any delays caused by the user interface, such as
displaying a dialog box, do not delay any iterations of the acquisition
processes.

VIs that involve control also benefit from the use of master/slave design
patterns. Consider a VI where a user controls a free-motion robotic arm
using buttons on a front panel. This type of VI requires efficient, accurate,
and responsive control because of the physical damage to the arm or
surroundings that might occur if control is mishandled. For example, if the
user instructs the arm to stop its downward motion, but the program is
occupied with the arm swivel control, the robotic arm might collide with the
support platform. Apply the master/slave design pattern to the application to
avoid these problems. In this case, the master loop handles the user
interface, and each controllable section of the robotic arm has its own slave
loop. Because each controllable section of the arm has its own loop and its
own piece of processing time, the user interface has more responsive control
of the robotic arm.

With a master/slave design pattern, it is important that no two While Loops

write to the same shared data. Ensure that no more than one While Loop
may write to any given piece of shared data. Refer to Lesson 2,
Synchronization Techniques, for more information about implementing an
application based on the master/slave design pattern.

The slave must not take too long to respond to the master. If the slave is
processing a signal from the master and the master sends more than
one message to the slave, the slave receives only the latest message. This
use of the master/slave architecture could cause a loss of data. Use a
master/slave architecture only if you are certain that each slave task takes
less time to execute than the master loop.

Producer/Consumer Design Pattern

The producer/consumer design pattern is based on the master/slave design
pattern and improves data sharing among multiple loops running at different
rates. Similar to the master/slave design pattern, the producer/consumer
design pattern separates tasks that produce and consume data at different
rates. The parallel loops in the producer/consumer design pattern are
separated into two categories—those that produce data and those that
consume the data produced. Data queues communicate data among the
11

© National Instruments Corporation 1-9 LabVIEW Core 2 Course Manual

Lesson 1 Common Design Techniques

loops. The data queues also buffer data among the producer and consumer
loops.

Tip A buffer is a memory device that stores temporary data among two devices, or in
this case, multiple loops.

Use the producer/consumer design pattern when you must acquire multiple
sets of data that must be processed in order. Suppose you want to create a VI
that accepts data while processing the data sets in the order they were
received. The producer/consumer pattern is ideal for this type of VI because
queuing (producing) the data occurs much faster than the data can be
processed (consumed). You could put the producer and consumer in the
same loop for this application, but the processing queue could not receive
additional data until the first piece of data was completely processed. The
producer/consumer approach to this VI queues the data in the producer loop
and processes the data in the consumer loop, as shown in Figure 1-7.

Tip Queue functions allow you to store a set of data that can be passed among multiple
loops running simultaneously or among VIs. Refer to Lesson 2, Synchronization
Techniques, for more information about queues and implementing applications using the
producer/consumer design pattern.

Figure 1-7.12Producer/Consumer Design Pattern

LabVIEW Core 2 Course Manual 1-10 ni.com

 Lesson 1 Common Design Techniques

This design pattern allows the consumer loop to process the data at its own
pace, while the producer loop continues to queue additional data.

You also can use the producer/consumer design pattern to create a VI that
analyzes network communication. This type of VI requires two processes to
operate at the same time and at different speeds. The first process constantly
polls the network line and retrieves packets. The second process analyzes
the packets retrieved by the first process.

In this example, the first process acts as the producer because it supplies data
to the second process, which acts as the consumer. The producer/consumer
design pattern is an effective architecture for this VI. The parallel producer
and consumer loops handle the retrieval and analysis of data off the network,
and the queued communication between the two loops allows buffering of
the network packets retrieved. Buffering can become important when
network communication is busy. With buffering, packets can be retrieved
and communicated faster than they can be analyzed.

D. Timing a Design Pattern

This section discusses two forms of timing—execution timing and software
control timing. Execution timing uses timing functions to give the processor
time to complete other tasks. Software control timing involves timing a
real-world operation to perform within a set time period.

Execution Timing
Execution timing involves timing a design pattern explicitly or based on
events that occur within the VI. Explicit timing uses a function that
specifically allows the processor time to complete other tasks, such as the
Wait Until Next ms Multiple function. When timing is based on events, the
design pattern waits for some action to occur before continuing and allows
the processor to complete other tasks while it waits.

Use explicit timing for design patterns such as the master/slave,

producer/consumer, and state machine. These design patterns perform some
type of polling while they execute.

Tip Polling is the process of making continuous requests for data from another device.
In LabVIEW, this generally means that the block diagram continuously asks if there is
data available, usually from the user interface.

For example, the master/slave design pattern shown in Figure 1-8 uses a
While Loop and a Case structure to implement the master loop. The master
executes continuously and polls for an event of some type, such as the user
clicking a button. When the event occurs, the master sends a message to the
slave. You need to time the
13 master so it does not take over the execution of

© National Instruments Corporation 1-11 LabVIEW Core 2 Course Manual

Lesson 1 Common Design Techniques

the processor. In this case, you typically use the Wait (ms) function to
regulate how frequently the master polls.

Tip Always use a timing function such as the Wait (ms) function or the Wait Until Next
ms Multiple function in any design pattern that continually executes and needs to be
regulated. If you do not use a timing function in a continuously executing structure,
LabVIEW uses all the processor time, and background processes may not run.

Figure 1-8. Master/Slave Design Pattern

Notice that the slave loop does not contain any form of timing. The use of
Synchronization functions, such as queues and notifiers, to pass messages
provides an inherent form of timing in the slave loop because the slave loop
waits for the Notifier function to receive a message. After the Notifier
function receives a message, the slave executes on the message. This creates
an efficient block diagram that does not waste processor cycles by polling
for messages. This is an example of execution timing by waiting for an
event.

When you implement design patterns where the timing is based on the
occurrence of events, you do not have to determine the correct timing
frequency because the execution of the design pattern occurs only when an
14

LabVIEW Core 2 Course Manual 1-12 ni.com

 Lesson 1 Common Design Techniques

event occurs. In other words, the design pattern executes only when it
receives an event.

Software Control Timing

Many applications that you create must execute an operation for a specified
amount of time. Consider implementing a state machine design pattern for
a temperature data acquisition system. If the specifications require that the
system acquire temperature data for 5 minutes, you could remain in the
acquisition state for 5 minutes. However, during that time you cannot
process any user interface actions such as stopping the VI. To process user
interface actions, you must implement timing so that the VI continually
executes for specified time. Implementing this type of timing involves
keeping the application executing while monitoring a real-time clock.

In YRO, you implemented software control timing to

monitor the time until the VI should acquire the next piece of data, as shown
in Figure 1-9. Notice the use of the Elapsed Time Express VI to keep track
of a clock.

Figure 1-9. Use of the Elapsed Time Express VI

If you use the Wait (ms) function or the Wait Until Next ms Multiple
function to perform software timing, the execution of the function you are
timing does not occur until the wait function finishes. These timing
functions are not the preferred method for performing software control
timing, especially for VIs where the system must continually execute. A
better method for software control timing utilizes the Get Date/Time In
Seconds function to get the current time and track it using shift registers.

15

© National Instruments Corporation 1-13 LabVIEW Core 2 Course Manual

Lesson 1 Common Design Techniques

Figure 1-10. Software Timing Using the Get Date/Time In Seconds Function

The Get Date/Time In Seconds function, connected to the left terminal of the
shift register, initializes the shift register with the current system time. Each
state uses another Get Date/Time In Seconds function and compares the
current time to the start time. If the difference in these two times is greater
or equal to the wait time, the state finishes executing and the rest of the
application executes.

Tip Always use the Get Date/Time In Seconds function instead of the Tick Count
function for this type of comparison because the value of the Tick Count function can
rollover to 0 during execution.

Refer to Appendix A, Using Variables, for more information on creating a

timing functional global variable to make the timing functionality modular
and reusable.

16

LabVIEW Core 2 Course Manual 1-14 ni.com

 Lesson 1 Common Design Techniques

Self-Review: Quiz
1. The state machine is an example of a design pattern.
a. True
b. False

2. Which of the following are reasons for using a multiple loop design
pattern?
a. Execute multiple tasks concurrently
b. Execute different states in a state machine
c. Execute tasks at different rates
d. Execute start up code, main loop, and shutdown code

3. Software control timing allows the processor time to complete other

tasks.
a. True
b. False

17

© National Instruments Corporation 1-15 LabVIEW Core 2 Course Manual

18
 Lesson 1 Common Design Techniques

Self-Review: Quiz Answers

1. The state machine is an example of a design pattern.
a. True
b. False

2. Which of the following are reasons for using a multiple loop design
pattern?
a. Execute multiple tasks concurrently
b. Execute different states in a state machine
c. Execute tasks at different rates
d. Execute start up code, main loop, and shutdown code

3. Software control timing allows the processor time to complete other

tasks.
a. True
b. False

19

© National Instruments Corporation 1-17 LabVIEW Core 2 Course Manual

Lesson 1 Common Design Techniques

Notes

20

LabVIEW Core 2 Course Manual 1-18 ni.com

Synchronization Techniques
2
Variables are useful in LabVIEW for passing data between parallel
processes. However, when using variables it is often difficult to synchronize
data transfers and you must take care to avoid race conditions. This lesson
introduces notifiers and queues as alternative methods for passing data
between parallel processes. Notifiers and queues have advantages over
using variables because of the ability to synchronize the transfer of data.

Topics
A. Variables
B. Notifiers
C. Queues

21

© National Instruments Corporation 2-1 LabVIEW Core 2 Course Manual

Lesson 2 Synchronization Techniques

A. Variables
For parallel loops to communicate, you must use some form of globally
available shared data. Using a variable breaks the LabVIEW dataflow
paradigm, allows for race conditions, and incurs more overhead than
passing the data by wire. Refer to Appendix A, Using Variables, for more
information about using variables for communicating among multiple
loops.

The example shown in Figure 2-1 is a less effective implementation of a

master/slave design pattern. This example uses a variable, which causes
two problems—there is no timing between the master and the slave, and the
variable can cause race conditions. The master cannot signal the slave that
data is available, so the slave loop must continually poll the variable to
determine if the data changes.

Figure 2-1. Master/Slave Architecture Using Global Variables

B. Notifiers
A more effective implementation of the master/slave design pattern uses
notifiers to synchronize data transfer. A notifier sends data along with a
notification that the data is available. Using a notifier to pass data from the
master to the slave removes any issues with race conditions. Using notifiers
also provides a synchronization advantage because the master and slave are
timed when data is available, providing for an elegant implementation of the
master/slave design pattern. Figure 2-2 shows the master/slave design
pattern using notifiers.
22

LabVIEW Core 2 Course Manual 2-2 ni.com

 Lesson 2 Synchronization Techniques

Figure 2-2. Master/Slave Design Pattern Using Notifiers

The notifier is created before the loops begin using the Obtain Notifier
function. The master loop uses the Send Notification function to notify the
slave loop through the Wait on Notification function. After the VI has
finished using the notifiers, the Release Notifier function releases the
notifiers.

The following benefits result from using notifiers in the master/slave design
pattern:
• Both loops are synchronized to the master loop. The slave loop only
executes when the master loop sends a notification.
• You can use notifiers to create globally available data. Thus, you can
send data with a notification. For example, in Figure 2-2, the Send
Notification function sends the string instruction.
• Using notifiers creates efficient code. You need not use polling to
determine when data is available from the master loop.

However, using notifiers can have drawbacks. A notifier does not buffer the
data. If the master loop sends another piece of data before the slave loop(s)
reads the first piece of data, that data is overwritten and lost.
23

© National Instruments Corporation 2-3 LabVIEW Core 2 Course Manual

Lesson 2 Synchronization Techniques

C. Queues
Queues are similar to notifiers, except that a queue can store multiple pieces
of data. By default, queues work in a first in, first out (FIFO) manner.
Therefore, the first piece of data inserted into the queue is the first piece of
data that is removed from the queue. Use a queue when you want to process
all data placed in the queue. Use a notifier if you want to process only the
current data.

When used, the producer/consumer design pattern, queues pass data and
synchronize the loops.

Figure 2-3. Producer/Consumer Design Pattern Using Queues

The queue is created before the loops begin using the Obtain Queue
function. The producer loop uses the Enqueue Element function to add data
to the queue. The consumer loop removes data from the queue using the
Dequeue Element function. The consumer loop does not execute until data
is available in the queue. After the VI has finished using the queues, the
Release Queue function releases the queues. When the queue releases, the
Dequeue Element function generates an error, effectively stopping the
consumer loop. This eliminates the need to use a variable to stop the loops.

24

LabVIEW Core 2 Course Manual 2-4 ni.com

 Lesson 2 Synchronization Techniques

The following benefits result from using queues in the producer/consumer

design pattern:
• Both loops are synchronized to the producer loop. The consumer loop
only executes when data is available in the queue.
• You can use queues to create globally available data that is queued,
removing the possibility of losing the data in the queue when new data
is added to the queue.
• Using queues creates efficient code. You need not use polling to
determine when data is available from the producer loop.

Queues are also useful for holding state requests in a state machine. In the
implementation of a state machine that you have learned, if two states are
requested simultaneously, you might lose one of the state requests. A queue
stores the second state request and executes it when the first has finished.

Case Study: Weather Station Project

The weather station project acquires temperature and wind speed data,
and analyzes it to determine if the situation requires a warning. If the
temperature is too high or too low, it alerts the user to a danger of heatstroke
or freezing. It also monitors the wind speed to generate a high wind warning
when appropriate.

The block diagram consists of two parallel loops, which are synchronized
using queues. One loop acquires data for temperature and wind speed and
the other loop analyzes the data. The loops in the block diagram use the
producer/consumer design pattern and pass the data through the queue.
Queues help process every reading acquired from the DAQ Assistant.

Code for acquiring temperature and wind speed is placed in the producer
loop. Code containing the state machine for analysis of temperature-weather
conditions is within the no error case of the consumer loop. The code using
a queue is more readable and efficient than the code using only state
machine architecture. The Obtain Queue function creates the queue
reference. The producer loop uses the Enqueue Element function to add data
obtained from the DAQ Assistant to the queue. The consumer loop uses the
Dequeue Element function to get the data from the queue and provide it to
the state machine for analysis. The Release Queue function marks the end
of queue by destroying it. The use of queues also eliminates the need for a
shared variable to stop the loops because the Dequeue Element function
stops the consumer loop when the queue is released.

25

© National Instruments Corporation 2-5 LabVIEW Core 2 Course Manual

Lesson 2 Synchronization Techniques

Figure 2-4 shows the block diagram consisting of a producer and a

consumer loop. Data transfer and synchronization between the loops is
achieved by the queue functions.

Figure 2-4. Data Transfer and Synchronization of Parallel Loops Using Queues

26

LabVIEW Core 2 Course Manual 2-6 ni.com

 Lesson 2 Synchronization Techniques

Self-Review: Quiz
1. Which of the following buffer data?
a. Notifiers
b. Queues
c. Global Variables
d. Local Variables

2. Match the following:

Obtain Queue Destroys the queue reference

Get Queue Status Assigns the data type of the queue

Release Queue Adds an element to the back of the queue

Enqueue Element Determines the number of elements currently

in the queue

3. Which of the following are valid data types for Queues and Notifiers?
a. String
b. Numeric
c. Enum
d. Array of Booleans
e. Cluster of a String and a Numeric

27

© National Instruments Corporation 2-7 LabVIEW Core 2 Course Manual

28
 Lesson 2 Synchronization Techniques

Self-Review: Quiz Answers

1. Which of the following buffer data?
a. Notifiers
b. Queues
c. Global Variables
d. Local Variables

2. Match the following:

Obtain Queue Assigns the data type of the queue

Get Queue Status Determines the number of elements

currently in the queue

Release Queue Destroys the queue reference

Enqueue Element Adds an element to the back of the queue

3. Which of the following are valid data types for Queues and Notifiers?
a. String
b. Numeric
c. Enum
d. Array of Booleans
e. Cluster of a String and a Numeric

29

© National Instruments Corporation 2-9 LabVIEW Core 2 Course Manual

Lesson 2 Synchronization Techniques

Notes

30

LabVIEW Core 2 Course Manual 2-10 ni.com

Event Programming
3
Event-based design patterns allow you to create more efficient and flexible
applications. Event-based design patterns use the Event structure to respond
directly to the user or other events. This lesson describes event-driven
programming using the Event structure and design patterns that use the
Event structure.

Topics
A. Events
B. Event-Driven Programming
C. Caveats and Recommendations
D. Event-Based Design Patterns

31

© National Instruments Corporation 3-1 LabVIEW Core 2 Course Manual

Lesson 3 Event Programming

A. Events
LabVIEW is a dataflow programming environment where the flow of data
determines the execution order of block diagram elements. Event-driven
programming features extend the LabVIEW dataflow environment to allow
the user’s direct interaction with the front panel and other asynchronous
activity to further influence block diagram execution.

Note Event-driven programming features are available only in the LabVIEW Full and
Professional Development Systems. You can run a VI built with these features in the
LabVIEW Base Package, but you cannot reconfigure the event-handling components.

What Are Events?

An event is an asynchronous notification that something has occurred.
Events can originate from the user interface, external I/O, or other parts of
the program. User interface events include mouse clicks, key presses, and so
on. External I/O events include hardware timers or triggers that signal when
data acquisition completes or when an error condition occurs. Other types
of events can be generated programmatically and used to communicate with
different parts of the program. LabVIEW supports user interface and
programmatically generated events. LabVIEW also supports ActiveX and
.NET generated events, which are external I/O events.

In an event-driven program, events that occur in the system directly

influence the execution flow. In contrast, a procedural program executes in
a pre-determined, sequential order. Event-driven programs usually include
a loop that waits for an event to occur, executes code to respond to the event,
and reiterates to wait for the next event. How the program responds to each
event depends on the code written for that specific event. The order in which
an event-driven program executes depends on which events occur and on the
order in which they occur. Some sections of the program might execute
frequently because the events they handle occur frequently, and other
sections of the program might not execute at all because the events never
occur.

Why Use Events?

Use user interface events in LabVIEW to synchronize user actions on the
front panel with block diagram execution. Events allow you to execute a
specific event-handling case each time a user performs a specific action.
Without events, the block diagram must poll the state of front panel objects
in a loop, checking to see if any change has occurred. Polling the front panel
requires a significant amount of CPU time and can fail to detect changes if
they occur too quickly. By using events to respond to specific user actions,
you eliminate the need to poll the front panel to determine which actions the
user performed. Instead, LabVIEW actively notifies the block diagram each
32

LabVIEW Core 2 Course Manual 3-2 ni.com

 Lesson 3 Event Programming

time an interaction you specified occurs. Using events reduces the CPU
requirements of the program, simplifies the block diagram code, and
guarantees that the block diagram can respond to all interactions the user
makes.

Use programmatically generated events to communicate among different

parts of the program that have no dataflow dependency. Programmatically
generated events have many of the same advantages as user interface events
and can share the same event-handling code, making it easy to implement
advanced architectures, such as queued state machines using events.

B. Event-Driven Programming
In Lesson 1, Common Design Techniques, you learned how event-driven
programming extends the LabVIEW dataflow environment to allow
user interaction with the front panel. You also learned about
programmatically-generated events allow for easy implementation of
advanced architectures, such as queued state machines.

Event Structure Components

Use the Event structure, shown as follows, to handle events in a VI.

The Event structure works like a Case structure with a built-in Wait on
Notification function. The Event structure can have multiple cases, each of
which is a separate event-handling routine. You can configure each case to
handle one or more events, but only one of these events can occur at a time.
When the Event structure executes, it waits until one of the configured
events occur, then executes the corresponding case for that event. The Event
structure completes execution after handling exactly one event. It does not
implicitly loop to handle multiple events. Like a Wait on Notification
function, the Event structure can time out while waiting for notification of
an event. When this occurs, a specific Timeout case executes.

The event selector label at the top of the Event structure indicates which
events cause the currently displayed case to execute.

View other event cases by clicking the down arrow next to the case name and
selecting another case from the shortcut menu.

The Timeout terminal at the top left corner of the Event structure specifies
the number of milliseconds to wait for an event before timing out.
33

© National Instruments Corporation 3-3 LabVIEW Core 2 Course Manual

Lesson 3 Event Programming

The default is –1, which specifies to wait indefinitely for an event to occur.
If you wire a value to the Timeout terminal, you must provide a Timeout
case.

The Event Data Node behaves similarly to the Unbundle By Name function.

This node is attached to the inside left border of each event case. The node
identifies the data LabVIEW provides when an event occurs. You can resize
this node vertically to add more data items, and you can set each data item
in the node to access any event data element. The node provides different
data elements in each case of the Event structure depending on which
event(s) you configure that case to handle. If you configure a single case to
handle multiple events, the Event Data Node provides only the event data
elements that are common to all the events configured for that case.

The Event Filter Node is similar to the Event Data Node.

This node is attached to the inside right border of filter event cases. The node
identifies the subset of data available in the Event Data Node that the event
case can modify. The node displays different data depending on which
event(s) you configure that case to handle. By default, these items are
inplace to the corresponding data items in the Event Data Node. If you do
not wire a value to a data item of an Event Filter Node, that data item
remains unchanged.

Refer to the Notify and Filter Events section of this lesson for more
information about filter events.

The dynamic event terminals are available by right-clicking the Event

structure and selecting Show Dynamic Event Terminals from the shortcut
menu.

These terminals are used only for dynamic event registration.

Refer to the Dynamic Event Registration topic and the Modifying

Registration Dynamically topic of the LabVIEW Help for more information
about using these terminals.

Note Like a Case structure, the Event structure supports tunnels. However, by default
you do not have to wire Event structure output tunnels in every case. All unwired tunnels
use the default value for the tunnel data type. Right-click a tunnel and deselect Use
Default If Unwired from the shortcut menu to revert to the default Case structure
behavior where tunnels must be wired in all cases. You also can configure the tunnels to
wire the input and output tunnels automatically in unwired cases.

Refer to the LabVIEW Help for information about the default values for data
types. 34

LabVIEW Core 2 Course Manual 3-4 ni.com

 Lesson 3 Event Programming

Notify and Filter Events

Notify events are an indication that a user action has already occurred, such
as when the user has changed the value of a control. Use notify events to
respond to an event after it has occurred and LabVIEW has processed it. You
can configure any number of Event structures to respond to the same notify
event on a specific object. When the event occurs, LabVIEW sends a copy
of the event to each Event structure configured to handle the event in
parallel.

Filter events inform you that the user has performed an action before
LabVIEW processes it, which allows you to customize how the program
responds to interactions with the user interface. Use filter events to
participate in the handling of the event, possibly overriding the default
behavior for the event. In an Event structure case for a filter event, you can
validate or change the event data before LabVIEW finishes processing it, or
you can discard the event entirely to prevent the change from affecting the
VI. For example, you can configure an Event structure to discard the Panel
Close? event, which prevents the user from interactively closing the front
panel of the VI.

Filter events have names that end with a question mark, such as Panel
Close?, to help you distinguish them from notify events. Most filter events
have an associated notify event of the same name, but without the question
mark, which LabVIEW generates after the filter event if no event case
discarded the event.

For example, you can use the Mouse Down? and Shortcut Menu Activation?
filter events to display a context menu when you left-click a control. To
perform this action, modify the data returned by the Button event data field
of the Mouse Down? filter event. The value of the left mouse button is 1, and
the value of the right mouse button is 2. In order to display the context menu
when you left-click a control, change the Button event data field to 2 so that
LabVIEW treats a left-click like a right-click. Refer to the Left-click
Shortcut Menu VI in the labview\examples\general directory for an
example of using filter events.

As with notify events, you can configure any number of Event structures to
respond to the same filter event on a specific object. However, LabVIEW
sends filter events sequentially to each Event structure configured for the
event. The order in which LabVIEW sends the event to each Event structure
depends on the order in which the events were registered. Each Event
structure must complete its event case for the event before LabVIEW can
notify the next Event structure. If an Event structure case changes any of the
event data, LabVIEW passes the changed data to subsequent Event
structures in the chain. If an Event structure in the chain discards the event,
LabVIEW does not pass the event to any Event structures remaining in the
35

© National Instruments Corporation 3-5 LabVIEW Core 2 Course Manual

Lesson 3 Event Programming

chain. LabVIEW completes processing the user action which triggered the
event only after all configured Event structures handle the event without
discarding it.

Note National Instruments recommends you use filter events only when you want to
take part in the handling of the user action, either by discarding the event or by modifying
the event data. If you only want to know that the user performed a particular action, use
notify events.

Event structure cases that handle filter events have an Event Filter Node.
You can change the event data by wiring new values to these terminals. If
you do not wire a value to the data item of the Event Filter Node, the default
value equals the value that the corresponding item in the Event Data Node
returns. You can completely discard an event by wiring a TRUE value to the
Discard? terminal.

Note A single case in the Event structure cannot handle both notify and filter events. A
case can handle multiple notify events but can handle multiple filter events only if the
event data items are identical for all events.

Refer to the Using Events in LabVIEW section of this lesson for more
information about event registration.

Tip In the Edit Events dialog box, notify events are signified by a green arrow, and filter
events are signified by a red arrow.

Using Events in LabVIEW

LabVIEW can generate many different events. To avoid generating
unwanted events, use event registration to specify which events you want
LabVIEW to notify you about. LabVIEW supports two models for event
registration—static and dynamic.

Static registration allows you to specify which events on the front panel of
a VI you want to handle in each Event structure case on the block diagram
of that VI. LabVIEW registers these events automatically when the VI runs,
so the Event structure begins waiting for events as soon as the VI begins
running. Each event is associated with a control on the front panel of the VI,
the front panel window of the VI as a whole, or the LabVIEW application.
You cannot statically configure an Event structure to handle events for the
front panel of a different VI. Configuration is static because you cannot
change at run time which events the Event structure handles.

Dynamic event registration avoids the limitations of static registration by

integrating event registration with the VI Server, which allows you to use
Application, VI, and control references to specify at run time the objects for
36

LabVIEW Core 2 Course Manual 3-6 ni.com

 Lesson 3 Event Programming

which you want to generate events. Dynamic registration provides more

flexibility in controlling what events LabVIEW generates and when it
generates them. However, dynamic registration is more complex than static
registration because it requires using VI Server references with block
diagram functions to explicitly register and unregister for events rather than
handling registration automatically using the information you configured in
the Event structure.

Note In general, LabVIEW generates user interface events only as a result of direct user
interaction with the active front panel. LabVIEW does not generate events, such as Value
Change, when you use shared variables, global variables, local variables, DataSocket,
and so on. However, you can use the Value (Signaling) property to generate a Value
Change event programmatically. In many cases, you can use programmatically generated
events instead of queues and notifiers.

The event data provided by a LabVIEW event always include a time stamp,
an enumeration that indicates which event occurred, and a VI Server
reference to the object that triggered the event. The time stamp is a
millisecond counter you can use to compute the time elapsed between
two events or to determine the order of occurrence. The reference to the
object that generated the event is strictly typed to the VI Server class of that
object. Events are grouped into classes according to what type of object
generates the event, such as Application, VI, or Control. If a single case
handles multiple events for objects of different VI Server classes, the
reference type is the common parent class of all objects. For example, if you
configure a single case in the Event structure to handle events for a numeric
control and a color ramp control, the type of the control reference of the
event source is Numeric because the numeric and color ramp controls are in
the Numeric class. If you register for the same event on both the VI and
Control class, LabVIEW generates the VI event first.

Note Clusters are the only container objects for which you can generate events.
LabVIEW generates Control events for clusters, before it generates events for the objects
they contain, except in the case of the Value Change event. The Value Change event
generates the event on an element in the cluster, then on the cluster itself. If the Event
structure case for a VI event or for a Control event on a container object discards the
event, LabVIEW does not generate further events.

Each Event structure and Register For Events function on the block diagram
owns a queue that LabVIEW uses to store events. When an event occurs,
LabVIEW places a copy of the event into each queue registered for that
event. An Event structure handles all events in its queue and the events in
the queues of any Register For Events functions that you wired to the
dynamic event terminals of the Event structure. LabVIEW uses these queues

37

© National Instruments Corporation 3-7 LabVIEW Core 2 Course Manual

Lesson 3 Event Programming

to ensure that events are reliably delivered to each registered Event structure
in the order the events occur.

By default, when an event enters a queue, LabVIEW locks the front panel
that contains the object that generated that event. LabVIEW keeps the front
panel locked until all Event structures finish handling the event. While the
front panel is locked, LabVIEW does not process front panel activity but
places those interactions in a buffer and handles them when the front panel
is unlocked.

For example, a user might anticipate that an event case launches an

application that requires text entry. Since the user already knows text entry
is needed, he might begin typing before the application appears on the front
panel. If the Lock front panel until the event case for this event
completes option is enabled, once the application launches and appears on
the front panel, it processes the key presses in the order in which they
occurred. If the Lock front panel until the event case for this event
completes option is disabled, the key presses might be processed elsewhere
on the front panel, since LabVIEW does not queue their execution to depend
on the completion of the event case.

Front panel locking does not affect certain actions, such as moving the
window, interacting with the scroll bars, and clicking the Abort button.

LabVIEW can generate events even when no Event structure is waiting to

handle them. Because the Event structure handles only one event each time
it executes, place the Event structure in a While Loop to ensure that an Event
structure can handle all events that occur.

Caution If no Event structure executes to handle an event and front panel locking is
enabled, the user interface of the VI becomes unresponsive. If this occurs, click the
Abort button to stop the VI. You can disable front panel locking by right-clicking the
Event structure and removing the checkmark from the Lock front panel until the event
case for this event completes checkbox in the Edit Events dialog box. You cannot turn
off front panel locking for filter events.

Static Event Registration

Static event registration is available only for user interface events. Use the
Edit Events dialog box to configure an Event structure to handle a statically
registered event. Select the event source, which can be the application, the
VI, or an individual control. Select a specific event the event source can
generate, such as Panel Resize, Value Change, and so on. Edit the case to
handle the event data according to the application requirements.

LabVIEW statically registers events automatically and transparently when

you run a VI that contains an Event structure. LabVIEW generates events
38

LabVIEW Core 2 Course Manual 3-8 ni.com

 Lesson 3 Event Programming

for a VI only while that VI is running or when another running VI calls the
VI as a subVI.

When you run a VI, LabVIEW sets that top-level VI and the hierarchy of
subVIs the VI calls on its block diagram to an execution state called
reserved. You cannot edit a VI or click the Run button while the VI is in the
reserved state because the VI can be called as a subVI at any time while its
parent VI runs. When LabVIEW sets a VI to the reserved state, it
automatically registers the events you statically configured in all Event
structures on the block diagram of that VI. When the top-level VI finishes
running, LabVIEW sets it and its subVI hierarchy to the idle execution state
and automatically unregisters the events.

Refer to the labview\examples\general\uievents.llb for

examples of using static event registration.

Configuring Events
Before you configure events for the Event structure to handle, refer to the
Caveats and Recommendations when Using Events in LabVIEW topic of the
LabVIEW Help.

Complete the following steps to configure an Event structure case to handle

an event.
1. (Optional) If you want to configure the Event structure to handle a user
event, a Boolean control within a radio buttons control, or a user
interface event that is generated based on a reference to an application,
VI, or control, you first must dynamically register that event. Refer to
the Dynamically Registering Events topic of the LabVIEW Help for
more information about using dynamic events.
2. Right-click the border of the Event structure and select Edit Events
Handled by This Case from the shortcut menu to display the Edit
Events dialog box to edit the current case. You also can select Add
Event Case from the shortcut menu to create a new case.
3. Specify an event source in the Event Sources pane.
4. Select the event you want to configure for the event source, such as Key
Down, Timeout, or Value Change from the Events list. When you
select a dynamic event source from the Event Sources list, the Events
list displays that event. This is the same event you selected when you
registered the event. If you have registered for events dynamically and
wired event reg refnum out to the dynamic event terminal, the sources
appear in the Dynamic section.
5. If you want to add additional events for the current case to handle, click
the + button and repeat steps 3 and 4 to specify each additional event.
The Event Specifiers section at the top of the dialog box lists all the
39

© National Instruments Corporation 3-9 LabVIEW Core 2 Course Manual

Lesson 3 Event Programming

events for the case to handle. When you click an item in this list, the
Event Sources section updates to highlight the event source you
selected. You can repeat steps 3 and 4 to redefine each event or click the
X button to remove the selected event.
6. Click the OK button to save the configuration and close the dialog box.
The event cases you configured appear as selection options in the event
selector label at the top of the Event structure and the Event Data node
displays the data common to all events handled in that case.
7. (Optional) You can use a Timeout event to configure an Event structure
to wait a specified amount of time for an event to occur. Wire a value to
the Timeout terminal at the top left of the Event structure to specify the
number of milliseconds the Event structure should wait for an event to
occur before generating a Timeout event. The default value for the
Timeout terminal is –1, which specifies to wait indefinitely for an event
to occur.
8. Repeat steps 1 through 6 for each event case you want to configure.

Refer to the following VIs for examples of using events:

labview\examples\general\dynamicevents.llb
labview\examples\general\uievents.llb

Event Example
Figure 3-1 shows an Event structure configured with the Menu Selection
(User) event. This VI uses the Event structure to capture menu selections
made using the user-defined menu named sample.rtm. The ItemTag
returns the menu item that was selected and the MenuRef returns the refnum
to the menubar. This information is passed to the Get Menu Item Info
function. Refer to examples\general\uievents.llb for more
examples of using events.

Figure 3-1. Menu Selection (User) Event

Note If you use the Get Menu Selection function with an Event structure configured to
handle the same menu item, the Event structure takes precedence, and LabVIEW ignores
the Get Menu Selection function. In any given VI, use the Event structure or the Get
Menu Selection function to handle menu events, not both.
40

LabVIEW Core 2 Course Manual 3-10 ni.com

 Lesson 3 Event Programming

C. Caveats and Recommendations

The following list describes some of the caveats and recommendations to
consider when incorporating events into LabVIEW applications.
• Avoid using an Event structure outside a loop.
– LabVIEW can generate events even when no Event structure is
waiting to handle them. Because the Event structure handles only
one event each time it executes, place the Event structure in a While
Loop that terminates when the VI is no longer interested in events to
ensure that an Event structure handles all events that occur.
• Remember to read the terminal of a latched Boolean control in its Value
Change event case.
– When you trigger an event on a Boolean control configured with a
latching mechanical action, the Boolean control does not reset to its
default value until the block diagram reads the terminal on the
Boolean control. You must read the terminal inside the event case for
the mechanical action to work correctly.
• Avoid placing two Event structures in one loop.
– National Instruments recommends that you place only one Event
structure in a loop. When an event occurs in this configuration, the
Event structure handles the event, the loop iterates, and the Event
structure waits for the next event to occur. If you place two Event
structures in a single loop, the loop cannot iterate until both Event
structures handle an event. If you have enabled front panel locking
for the Event structures, the user interface of the VI can become
unresponsive depending on how the user interacts with the front
panel.

Refer to the Caveats and Recommendations when Using Events in

LabVIEW topic of the LabVIEW Help for more caveats and
recommendations when you use events in LabVIEW.

41

© National Instruments Corporation 3-11 LabVIEW Core 2 Course Manual

Lesson 3 Event Programming

D. Event-Based Design Patterns

Event-based design patterns provide efficiency gains because they only
respond when an event occurs. When LabVIEW executes the Event
structure, the VI that contains the Event structure sleeps until a registered
events occurs, or generates. When a registered event generates, the Event
structure automatically wakes up and executes the appropriate subdiagram
to handle the event.

User Interface Event Handler Design Pattern

The user interface event handler design pattern provides a powerful and
efficient architecture for handling user interaction with LabVIEW. Use the
user interface event handler for detecting when a user changes the value of
a control, moves or clicks the mouse, or presses a key.

The standard user interface event handler template consists of an Event

structure contained in a While Loop, as shown in Figure 3-2. Configure the
Event structure to have one case for each category of event you want to
detect. Each event case contains the handling code that executes
immediately after an event occurs.

Because the event handler loop wakes up precisely when an event occurs
and sleeps in between events, you do not have to poll or read control values
repeatedly in order to detect when a user clicks a button. The user interface
event handler allows you to minimize processor use without sacrificing
interactivity.

1 Event Structure 3 Timeout Terminal

2 While Loop 4 Event Data Node

Figure 3-2. User Interface Event Handler Design Pattern

42

LabVIEW Core 2 Course Manual 3-12 ni.com

 Lesson 3 Event Programming

A common problem when using the user interface event handler is that it
computes the While Loop termination before the Event structure executes.
This can cause the While Loop to iterate one more time than you expected.
To avoid this situation, compute the While Loop termination within all your
event handling code.

The event handler code must execute quickly, generally within 200 ms.
Anything slower can make it feel as if the user interface is locked up. Also,
if the event handler code takes a long time to execute, the Event structure
might lock. By default, the front panel locks while an event is handled. You
can disable front panel locking for each event case to make the user interface
more responsive. However, any new events that are generated while an event
is being handled will not be handled immediately. So, the user interface will
still seem unresponsive.

Any code that is in an event case cannot be shared with another Event
structure. You must use good code design when using the Event structure.
Modularize code that will be shared between multiple Event structure cases.

The Event structure includes a Timeout event, which allows you to control
when the Timeout event executes. For example, if you set a Timeout of
200 ms, the Timeout event case executes every 200 ms in the absence of
other events. You can use the Timeout event to perform critical timing in
your code.

Producer/Consumer (Events) Design Pattern

One of the most versatile and flexible design patterns combines the
producer/consumer and user interface event handler design patterns.
A VI built using the producer/consumer (events) pattern responds to the
user interface asynchronously, allowing the user interface to continuously
respond to the user. The consumer loop of this pattern responds as events
occur, similar to the consumer loop of the producer/consumer (data) design
pattern.

The producer/consumer (events) design pattern uses the same

implementation as the producer/consumer (data) design pattern except the
producer loop uses an Event structure to respond to user interface events, as
shown in Figure 3-3. The Event structure enables continuous response to
user interaction.

43

© National Instruments Corporation 3-13 LabVIEW Core 2 Course Manual

Lesson 3 Event Programming

Figure 3-3. Producer/Consumer (Events) Design Pattern

Figure 3-3 shows how you can use Synchronization VIs and functions to
add functionality to the design pattern.

Queues have the ability to transfer any data type. The data type transferred
in Figure 3-3 is a string. A string is not the most efficient data type for
passing data in design patterns. A more efficient data type for passing data
in design patterns is a cluster consisting of an enumerated type control and
a variant.

44

LabVIEW Core 2 Course Manual 3-14 ni.com

 Lesson 3 Event Programming

Self Review: Quiz

1. Using user interface events allows you to synchronize user actions on
the front panel with block diagram execution.
a. True
b. False

2. The Event structure handles only one event each time it executes.
a. True
b. False

3. Which of the following are examples of user interface events?

a. Mouse click
b. Keystroke
c. Event Filter Node
d. Value change of a control

4. Which of the following operations will generate a Value Change event

for a numeric control?
a. Click inside the digital display window and enter a number from the
keyboard
b. Click the increment or decrement arrow buttons
c. Place the cursor to the right of the digit you want to change and press
the up or down arrow keys
d. Update the numeric control using a local variable

45

© National Instruments Corporation 3-15 LabVIEW Core 2 Course Manual

46
 Lesson 3 Event Programming

Self Review: Quiz Answers

1. Using user interface events allows you to synchronize user actions on
the front panel with block diagram execution.
a. True
b. False

2. The Event structure handles only one event each time it executes.
a. True
b. False

3. Which of the following are examples of user interface events?

a. Mouse click
b. Keystroke
c. Event Filter Node
d. Value change of a control

4. Which of the following operations will generate a Value Change event

for a numeric control?
a. Click inside the digital display window and enter a number from
the keyboard
b. Click the increment or decrement arrow buttons
c. Place the cursor to the right of the digit you want to change and
press the up or down arrow keys
d. Update the numeric control using a local variable

47

© National Instruments Corporation 3-17 LabVIEW Core 2 Course Manual

Lesson 3 Event Programming

Notes

48

LabVIEW Core 2 Course Manual 3-18 ni.com

Error Handling
4
By default, LabVIEW automatically handles any error when a VI runs by
suspending execution, highlighting the subVI or function where the error
occurred, and displaying an error dialog box. Automatic error handling is
convenient for quick prototypes and proof-of-concept development, but not
recommended for professional application development. If you rely on
automatic error handling your application might stop in a critical section of
your code because of an error dialog box. The user might be unable to
continue running the application or fix the problem.

By manually implementing error handling, you control when popup dialogs

occur. If you plan to create a stand-alone application, you must incorporate
manual error handling because LabVIEW does not display automatic error
handling dialog boxes in the LabVIEW Run-Time Engine.

Topics
A. Importance of Error Handling
B. Detect and Report Errors
C. Errors and Warnings
D. Ranges of Error Codes
E. Error Handlers

49

© National Instruments Corporation 4-1 LabVIEW Core 2 Course Manual

Lesson 4 Error Handling

A. Importance of Error Handling

Error handling is the mechanism for anticipation, detection, and resolution
of warnings and errors. Error handling is an essential component in your
LabVIEW application development. With error handling you quickly
pinpoint the source of programming errors. Without it, you might observe
unexpected behavior but struggle to find the source of the problem.

Error handling is also extremely valuable when you test your application to
ensure that your error reporting is meaningful and that the error handling
code safely stops your application when an error occurs. For example,
during stress testing you are setting values or conditions that are beyond the
normal operational capacity of your application which often result in errors.
When such errors occur, you want to ensure proper shutdown of your
application.

Error handling continues to be important after an application is deployed.

Error handling can help detect system and environment differences—such
as differences in file systems, memory, and disk resources.

B. Detect and Report Errors

To implement good error handling, you must determine the actions to take
when an error occurs at every point in your application. To begin with, you
must utilize the error terminals on functions and VIs. Since the error cluster
is implemented as a flow-through parameter, you should propagate errors by
wiring the error out cluster of the first node you want to execute to the error
in cluster of the next node you want to execute. You must continue to do this
for sequences of nodes.

As the VI runs, LabVIEW tests for errors at each node. If LabVIEW does
not find any errors, the node executes normally. If LabVIEW detects an
error, the node passes the error to the next node without executing that part
of the code. Any subVIs that you create should also implement this
flow-through behavior.

Use the Merge Error VI to merge the error out cluster values from parallel
sequences. Refer to Figure 4-1 for an example of merging error information
from parallel node sequences.

50

LabVIEW Core 2 Course Manual 4-2 ni.com

 Lesson 4 Error Handling

Figure 4-1. Merge Errors From Multiple Sources

At the end of your application after all error sources are merged into one
error cluster, you must report errors to the user using the Simple Error
Handler VI or another error reporting mechanism.

C. Errors and Warnings

Recall that the error in and error out clusters include the following
components of information:
• status is a Boolean value that reports TRUE if an error occurred.
• code is a 32-bit signed integer that identifies the error numerically. A
nonzero error code coupled with a status of FALSE signals a warning
rather than a error.
• source is a string that identifies where the error occurred.

Notice that an error is defined as status value of TRUE, regardless of the

code value. A nonzero code with a status of FALSE is considered a warning.
Although most errors have negative code values and warnings have positive
code values, this is not universally true. Therefore you should rely on the
status value and the code value to detect errors and warnings.

Warnings are typically considered less severe than errors. Some APIs and
functions, such as the Regular Expression function, only report errors.
However, other APIs such as the VISA API for controlling stand-alone
instruments often reports warnings. A common VISA warning occurs when
calling the VISA Read function and specifying the number of bytes to read.
In this case, VISA returns a warning with the following description: The
number of bytes transferred is equal to the requested
input count. More data might be available.

Unlike when an error occurs, nodes execute normally even when LabVIEW
detects a warning. Even though code executes normally, it is important that
during development you monitor warnings to ensure proper behavior of
your application. To ensure that warning information is propagated
correctly, it is important to51use shift-registers for error cluster wires in loops

© National Instruments Corporation 4-3 LabVIEW Core 2 Course Manual

Lesson 4 Error Handling

so that the warning information is propagated through all iterations. Refer

to Figure 4-2 for proper use of the shift-register to propagate errors and
warnings to successive loop iterations.

Figure 4-2. Use Shift Registers to Propagate Errors and Warnings

By default, the Simple Error Handler VI displays a dialog with a description

of any errors that occurred and does not report warnings. However, the
Simple Error Handler VI can be configured for other error handling
behavior. You can select the type of dialog from the following options:
• No dialog—Displays no dialog box. This is useful if you want to have
programmatic control over handling errors.
• OK message (default)—Displays a dialog box with a single OK button.
After the user acknowledges the dialog box, the VI returns control to the
main VI.
• Continue or stop message—Displays a dialog box with buttons, which
the user can use to either continue or stop. If the user selects Stop, the
VI calls the Stop function to halt execution.
• OK message with warnings—Displays a dialog box with any warnings
and a single OK button. After the user acknowledges the dialog box, the
VI returns control to the main VI.
• Continue or stop message with warnings—Displays a dialog box with
any warnings and buttons, which the user can use to either continue or
stop. If the user selects Stop, the VI calls the Stop function to halt
execution.

D. Ranges of Error Codes

VIs and functions in LabVIEW can return numeric error codes. Each
product or group of VIs defines a range of error codes. Refer to the Ranges
of LabVIEW Error Codes topic of the LabVIEW Help for error code tables
listing the numeric error codes and descriptions for each product and VI
grouping.

In addition to defining error code ranges, LabVIEW reserves some error

code ranges for you to use in your application. You can define custom error
codes in the range of –8999 through –8000, 5000 through 9999, or 500,000
through 599,999. 52

LabVIEW Core 2 Course Manual 4-4 ni.com

 Lesson 4 Error Handling

Some numeric error codes are used by more than one group of VIs and
functions. For example, error 65 is both a serial error code, indicating a
serial port timeout, and a networking error code, indicating that a network
connection is already established.

E. Error Handlers
An error handler is a VI or code that changes the normal flow of the program
when an error occurs. The Simple Error Handler VI is an example of a
built-in error handler that is used in LabVIEW. You can implement other
error handlers that are customized for your application. For example, you
might choose to log error information to a file. Another common error
handler is a VI that redirects code to a cleanup or shutdown routine when an
error occurs so that your application exits gracefully. Figure 4-3 shows a
state machine error handler that sets the next state to be the Shutdown state
when an error in status is TRUE.

Figure 4-3. State Machine Error Handler

53

© National Instruments Corporation 4-5 LabVIEW Core 2 Course Manual

54
 Lesson 4 Error Handling

Self Review: Quiz

1. Merge Errors VI concatenates error information from multiple sources.
a. True
b. False

2. All errors have negative error codes and all warnings have positive error
codes.
a. True
b. False

55

© National Instruments Corporation 4-7 LabVIEW Core 2 Course Manual

56
 Lesson 4 Error Handling

Self Review: Quiz Answers

1. Merge Errors VI concatenates error information from multiple sources.
a. True
b. False

2. All errors have negative error codes and all warnings have positive error
codes.
a. True
b. False

57

© National Instruments Corporation 4-9 LabVIEW Core 2 Course Manual

Lesson 4 Error Handling

Notes

58

LabVIEW Core 2 Course Manual 4-10 ni.com

Controlling the User Interface
5
When writing programs, often you must change the attributes of front panel
objects programmatically. For example, you may want to make an object
invisible until a certain point in the execution of the program. In LabVIEW,
you can use VI Server to access the properties and methods of front panel
objects. This lesson explains the Property Nodes, Invoke Nodes, VI Server,
and control references.

Topics
A. Property Nodes
B. Invoke Nodes
C. VI Server Architecture
D. Control References

59

© National Instruments Corporation 5-1 LabVIEW Core 2 Course Manual

Lesson 5 Controlling the User Interface

A. Property Nodes
Property Nodes access the properties of an object. In some applications, you
might want to programmatically modify the appearance of front panel
objects in response to certain inputs. For example, if a user enters an invalid
password, you might want a red LED to start blinking. Another example is
changing the color of a trace on a chart. When data points are above a certain
value, you might want to show a red trace instead of a green one. Property
Nodes allow you to make these modifications programmatically. You also
can use Property Nodes to resize front panel objects, hide parts of the front
panel, add cursors to graphs, and so on.

Property Nodes in LabVIEW are very powerful and have many uses. Refer
to the LabVIEW Help for more information about Property Nodes.

Creating Property Nodes

When you create a property from a front panel object by right-clicking the
object, selecting Create»Property Node, and selecting a property from the
shortcut menu, LabVIEW creates a Property Node on the block diagram that
is implicitly linked to the front panel object. If the object has a label, the
Property Node has the same label. You can change the label after you create
the node. You can create multiple Property Nodes for the same front panel
object.

Using Property Nodes

When you create a Property Node, it initially has one terminal representing
a property you can modify for the corresponding front panel object. Using
this terminal on the Property Node, you can either set (write) the property
or get (read) the current state of that property.

For example, if you create a Property Node for a digital numeric control
using the Visible property, a small arrow appears on the right side of the
Property Node terminal, indicating that you are reading that property value.
You can change the action to write by right-clicking the terminal and
selecting Change To Write from the shortcut menu. Wiring a False
Boolean value to the Visible property terminal causes the numeric control to
vanish from the front panel when the Property Node receives the data.
Wiring a True Boolean value causes the control to reappear.

Figure 5-1. Using Property Nodes

60

LabVIEW Core 2 Course Manual 5-2 ni.com

 Lesson 5 Controlling the User Interface

To get property information, right-click the node and select Change All to
Read from the shortcut menu. To set property information, right-click the
node and select Change All to Write from the shortcut menu. If a property
is read only, Change to Write is dimmed in the shortcut menu. If the small
direction arrow on the Property Node is on the right, you are getting the
property value. If the small direction arrow on a Property Node is on the left,
you are setting the property value. If the Property Node in Figure 5-1 is set
to Read, when it executes it outputs a True value if the control is visible or
a False value if it is invisible.

Tip Some properties are read-only, such as the Label property, or write only, such as the
Value (Signaling) property.

To add terminals to the node, right-click the white area of the node and select
Add Element from the shortcut menu or use the Positioning tool to resize
the node. Then, you can associate each Property Node terminal with a
different property from its shortcut menu.

Tip Property Nodes execute each terminal in order from top to bottom.

Some properties use clusters. These clusters contain several properties

that you can access using the cluster functions. Writing to these properties
as a group requires the Bundle function and reading from these properties
requires the Unbundle function. To access bundled properties, select All
Elements from the shortcut menu. For example, you can access all the
elements in the Position property by selecting Properties»Position»
All Elements from the shortcut menu.

However, you also can access the elements of the cluster as individual
properties, as shown in Figure 5-2.

Figure 5-2. Properties Using Clusters

61

© National Instruments Corporation 5-3 LabVIEW Core 2 Course Manual

Lesson 5 Controlling the User Interface

B. Invoke Nodes
Invoke Nodes access the methods of an object.

Use the Invoke Node to perform actions, or methods, on an application or

VI. Unlike the Property Node, a single Invoke Node executes only a single
method on an application or VI. Select a method by using the Operating tool
to click the method terminal or by right-clicking the white area of the node
and selecting Methods from the shortcut menu. You also can create an
implicitly linked Invoke Node by right-clicking a front panel object,
selecting Create»Invoke Node, and selecting a method from the shortcut
menu.

The name of the method is always the first terminal in the list of parameters
in the Invoke Node. If the method returns a value, the method terminal
displays the return value. Otherwise, the method terminal has no value.

The Invoke Node lists the parameters from top to bottom with the name of
the method at the top and the optional parameters, which are dimmed, at the
bottom.

Example Methods
An example of a method common to all controls is the Reinitialize to
Default method. Use this method to reinitialize a control to its default value
at some point in your VI. The VI class has a similar method called
Reinitialize All to Default.

Figure 5-3 is an example of a method associated with the Waveform Graph

class. This method exports the waveform graph image to the clipboard or to
a file.

Figure 5-3. Invoke Node for the Export Image Method

62

LabVIEW Core 2 Course Manual 5-4 ni.com

 Lesson 5 Controlling the User Interface

C. VI Server Architecture
The VI Server is an object-oriented, platform-independent technology that
provides programmatic access to LabVIEW and LabVIEW applications. VI
Server performs many functions; however, this lesson concentrates on using
the VI Server to control front panel objects and edit the properties of a VI
and LabVIEW.To understand how to use VI Server, it is useful to understand
the terminology associated with it.

Object-Oriented Terminology
Object-oriented programming is based on objects. An object is a member of
a class. A class defines what an object is able to do, what operations it can
perform (methods), and what properties it has, such as color, size, and so on.

Objects can have methods and properties. Methods perform an operation,

such as reinitializing the object to its default value. Properties are the
attributes of an object. The properties of an object could be its size, color,
visibility, and so on.

Control Classes
LabVIEW front panel objects inherit properties and methods from a class.
When you create a Stop control, it is an object of the Boolean class and has
properties and methods associated with that class, as shown in Figure 5-4.

Control Class
Control
Example Property: Visible
Example Method: Reinitialize to Default

Boolean Sub-Class Array Sub-Class

Boolean Array
Example Property: Boolean Text Example Property: Number of Rows

Stop Object
Stop
Visible: Yes
Reinitialize to Default: No
Boolean Text: Stop

Figure 5-4. Boolean Class Example

VI Class
Controls are not the only objects in LabVIEW to belong to a class. A VI
belongs to the VI Class and has its own properties and methods associated
with it. For instance, you can use VI class methods to abort a VI, to adjust
the position of the front panel window, and to get an image of the block
diagram. You can use VI class properties to change the title of a front panel
63

© National Instruments Corporation 5-5 LabVIEW Core 2 Course Manual

Lesson 5 Controlling the User Interface

window, to retrieve the size of the block diagram, and to hide the Abort
button.

D. Control References
A Property Node created from the front panel object or block diagram
terminal is an implicitly linked Property Node. This means that the Property
Node is linked to the front panel object. What if you must place your
Property Nodes in a subVI? Then the objects are no longer located on the
front panel of the VI that contains the Property Nodes. In this case, you need
an explicitly linked Property Node. You create an explicitly linked Property
Node by wiring a reference to a generic Property Node.

If you are building a VI that contains several Property Nodes or if you are
accessing the same property for several different controls and indicators,
you can place the Property Node in a subVI and use control references to
access that node. A control reference is a reference to a specific front panel
object.

This section shows one way to use control references. Refer to the
Controlling Front Panel Objects topic of the LabVIEW Help for more
information about control references.

Creating a SubVI with Property Nodes

As shown in Figure 5-5, the simplest way to create explicitly linked
Property Nodes is to complete the following steps:
1. Create your VI.
2. Select the portion of the block diagram that is in the subVI, as shown in
the first part of Figure 5-5.
3. Select Edit»Create SubVI. LabVIEW automatically creates the control
references needed for the subVI.
4. Customize and save the subVI. As you can see in the second part of
Figure 5-5, the subVI uses the default icon.

Objects Selected to Create SubVI Edit»Create SubVI used

Figure 5-5. Using Edit»Create SubVI to Create Control References
64

LabVIEW Core 2 Course Manual 5-6 ni.com

 Lesson 5 Controlling the User Interface

Figure 5-6 shows the subVI created. Notice that the front panel Control
Refnum controls have been created and connected to a Property Node on the
block diagram.

Created SubVI Front Panel Created SubVI Block Diagram

Figure 5-6. Sub VI Created Using Edit»Create SubVI

Note A red star on the Control Reference control indicates that the refnum is strictly
typed. Refer to the Strictly Typed and Weakly Typed Control Refnums section of the
Controlling Front Panel Objects topic of the LabVIEW Help for more information about
weakly and strictly typed control references.

Creating Control References

To create a control reference for a front panel object, right-click the object
or its block diagram terminal and select Create»Reference from the
shortcut menu.

You can wire this control reference to a generic Property Node. You can pass
the control reference to a subVI using a control refnum terminal.

Using Control References

Setting properties with a control reference is useful for setting the same
property for multiple controls. Some properties apply to all classes of
controls, such as the Disabled property. Some properties are only applicable
to certain control classes, such as the Lock Boolean Text in Center property.

The following example shows how to construct a VI that uses a control

reference on the subVI to set the Enable/Disable state of a control on the
main VI front panel.

65

© National Instruments Corporation 5-7 LabVIEW Core 2 Course Manual

Lesson 5 Controlling the User Interface

1 2

1 Main VI 2 SubVI
Figure 5-7. Control References

The main VI sends a reference for the digital numeric control to the subVI
along with a value of zero, one, or two from the enumerated control. The
subVI receives the reference by means of the Ctl Refnum on its front panel
window. Then, the reference is passed to the Property Node. Because the
Property Node now links to the numeric control in the main VI, the Property
Node can change the properties of that control. In this case, the Property
Node manipulates the enabled/disabled state.

Notice the appearance of the Property Node in the block diagram. You
cannot select a property in a generic Property Node until the class is chosen.
The class is chosen by wiring a reference to the Property Node. This is an
example of an explicitly linked Property Node. It is not linked to a control
until the VI is running and a reference is passed to the Property Node. The
advantage of this type of Property Node is its generic nature. Because it has
no explicit link to any one control, it may be reused for many different
controls. This generic Property Node is available on the Functions palette.

Selecting the Control Type

When you add a Control Refnum to the front panel of a subVI, you next
need to specify the VI Server Class of the control. This specifies the type of
control references that the subVI will accept. In the previous example,
Control was selected as the VI Server Class type, as shown in Figure 5-7.
This allows the VI to accept a reference to any type of front panel control.

However, you can specify a more specific class for the refnum to make the
subVI more restrictive. For example, you can select Digital as the class, and
the subVI only can accept66references to numeric controls of the class

LabVIEW Core 2 Course Manual 5-8 ni.com

 Lesson 5 Controlling the User Interface

Digital. Selecting a more generic class for a control refnum allows it to

accept a wider range of objects, but limits the available properties to those
that apply to all objects which the Property Node can accept.

To select a specific control class, right-click the control and select Select VI
Server Class»Generic»GObject»Control from the shortcut menu. Then,
select the specific control class.

Creating Properties and Methods with the Class Browser Window

You can use the Class Browser window to select an object library and create
a new property or method.

Complete the following steps to create a new property or method using the
Class Browser window.
1. Select View»Class Browser to display the Class Browser window.

2. From the Object library pull-down menu, select a library.

3. Select a class from the Class pull-down menu. Use the following buttons
to navigate the classes.
• Click the Select View button to toggle between an alphabetical view
and a hierarchical view of the items in the Class pull-down menu and
the Properties and Methods list.
• Click the Search button to launch the Class Browser Search dialog
box.
67

© National Instruments Corporation 5-9 LabVIEW Core 2 Course Manual

Lesson 5 Controlling the User Interface

4. From the Properties and Methods list in the Class Browser window,
select a property or method. The property or method you select appears
in the Selected property or method box.
5. Click the Create button or the Create Write button to attach a node
with the selected property or method to your mouse cursor and add the
node to the block diagram. The Create button creates a property for
reading or a method. This button dims when you select a write-only
property. To create a property for writing, click the Create Write
button. The Create Write button dims when you select a method or
read-only property. You also can drag a property or method from the
Properties and Methods list directly to the block diagram.
6. Repeat steps 2 through 5 for any other properties and methods you want
to create and add to the block diagram.

68

LabVIEW Core 2 Course Manual 5-10 ni.com

 Lesson 5 Controlling the User Interface

Self-Review: Quiz
1. For each of the following items, determine whether they operate on a VI
class or a Control class.
• Format and Precision
• Blinking
• Reinitialize to Default Value
• Show Tool Bar

2. You have a ChartGraph control refnum, shown at left, in a subVI. Which

of the following control references could you wire to the control refnum
terminal of the subVI? (multiple answers)
a. Control reference of an XY graph
b. Control reference of a numeric array
c. Control reference of a waveform chart
d. Control reference of a Boolean control

69

© National Instruments Corporation 5-11 LabVIEW Core 2 Course Manual

70
 Lesson 5 Controlling the User Interface

Self-Review: Quiz Answers

1. For each of the following items, determine whether they operate on a VI
class or a Control class.
• Format and Precision: Control
• Blinking: Control
• Reinitialize to Default Value: Control
• Show Tool Bar: VI

2. You have a GraphChart control refnum, shown at left, in a subVI. Which

control references could you wire to the control refnum terminal of the
subVI?
a. Control reference of an XY graph
b. Control reference of a numeric array
c. Control reference of a waveform chart
d. Control reference of a Boolean control

71

© National Instruments Corporation 5-13 LabVIEW Core 2 Course Manual

Lesson 5 Controlling the User Interface

Notes

72

LabVIEW Core 2 Course Manual 5-14 ni.com

File I/O Techniques
6
Frequently, the decision to separate the production of data and the
consumption of data into separate processes occurs because you must write
the data to a file as it is acquired. In such cases, you must choose a file
format. This lesson explains ASCII, Binary, and Technical Data
Management Streaming (TDMS) file formats and when each is a good
choice for your application.

Topics
A. File Formats
B. Binary Files
C. TDMS Files

73

© National Instruments Corporation 6-1 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

A. File Formats
At their lowest level, all files written to your computer’s hard drive are a
series of binary bits. However, many formats for organizing and
representing data in a file are available. In LabVIEW, three of the most
common techniques for storing data are the ASCII file format, direct binary
storage, and the TDMS file format. Each of these formats has advantages
and some formats work better for storing certain data types than others.

When to Use Text (ASCII) Files

Use text format files for your data to make it available to other users or
applications if disk space and file I/O speed are not crucial, if you do not
need to perform random access reads or writes, and if numeric precision is
not important.

Text files are the easiest format to use and to share. Almost any computer
can read from or write to a text file. A variety of text-based programs can
read text-based files.

Store data in text files when you want to access it from another application,
such as a word processing or spreadsheet application. To store data in text
format, use the String functions to convert all data to text strings. Text files
can contain information of different data types.

Text files typically take up more memory than binary and datalog files if
the data is not originally in text form, such as graph or chart data, because
the ASCII representation of data usually is larger than the data itself.
For example, you can store the number –123.4567 in 4 bytes as a
single-precision, floating-point number. However, its ASCII representation
takes 9 bytes, one for each character.

In addition, it is difficult to randomly access numeric data in text files.

Although each character in a string takes up exactly 1 byte of space, the
space required to express a number as text typically is not fixed. To find the
ninth number in a text file, LabVIEW must first read and convert the
preceding eight numbers.

You might lose precision if you store numeric data in text files. Computers
store numeric data as binary data, and typically you write numeric data to a
text file in decimal notation. Loss of precision is not an issue with binary
files.

74

LabVIEW Core 2 Course Manual 6-2 ni.com

 Lesson 6 File I/O Techniques

When to Use Binary Files

Storing binary data, such as an integer, uses a fixed number of bytes on disk.
For example, storing any number from 0 to 4 billion in binary format, such
as 1, 1,000, or 1,000,000, takes up 4 bytes for each number.

Use binary files to save numeric data and to access specific numbers from a
file or randomly access numbers from a file. Binary files are machine
readable only, unlike text files, which are human readable. Binary files are
the most compact and fastest format for storing data. You can use multiple
data types in binary files, but it is uncommon.

Binary files are more efficient because they use less disk space and because
you do not need to convert data to and from a text representation when you
store and retrieve data. A binary file can represent 256 values in 1 byte of
disk space. Often, binary files contain a byte-for-byte image of the data as it
was stored in memory, except for cases like extended and complex numeric
values. When the file contains a byte-for-byte image of the data as it was
stored in memory, reading the file is faster because conversion is not
necessary.

Datalog Files
A specific type of binary file, known as a datalog file, is the easiest method
for logging cluster data to file. Datalog files store arrays of clusters in a
binary representation. Datalog files provide efficient storage and random
access, however, the storage format for datalog files is complex, and
therefore they are difficult to access in any environment except LabVIEW.
Furthermore, in order to access the contents of a datalog file, you must know
the contents of the cluster type stored in the file. If you lose the definition of
the cluster, the file becomes very difficult to decode. For this reason, datalog
files are not recommended for sharing data with others or for storing data in
large organizations where you could lose or misplace the cluster definition.

When to Use TDMS Files

To reduce the need to design and maintain your own data file format,
National Instruments has created a flexible data model called Technical
Data Management Streaming, which is natively accessible through
LabVIEW, LabWindows™/CVI™, and DIAdem, and is portable to other
applications such as Microsoft Excel. The TDMS data model offers several
unique benefits such as the ability to scale your project requirements and
easily attach descriptive information to your measurements while streaming
your data to disk.

The TDMS file format consists of two files—a .tdms file and a
.tdms_index file. The .tdms file is a binary file that contains data and
stores properties about that data. The .tdms_index file is a binary index
75

© National Instruments Corporation 6-3 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

file that speeds up access while reading and provides consolidated

information on all the attributes and pointers in the TDMS file. All the
information in the .tdms_index file is also contained in the .tdms file.
For this reason, the .tdms_index file can be automatically regenerated
from the .tdms file. Therefore, when you distribute TDMS files, you only
need to distribute the .tdms file. The internal structure of the TDMS file
format is publicly documented, so it is possible to create third-party
programs to write and read TDMS files. In addition, there is a TDM Excel
Add-in Tool available on ni.com that you can install to load .tdms files
into Microsoft Excel.

Use TDMS files to store test or measurement data, especially when the data
consists of one or more arrays. TDMS files are most useful when storing
arrays of simple data types such as numbers, strings, or Boolean data.
TDMS files cannot store arrays of clusters directly. If your data is stored in
arrays of clusters, use another file format, such as binary, or break the cluster
up into channels and use the structure of the TDMS file to organize them
logically.

Use TDMS files to create a structure for your data. Data within a file is
organized into channels. You can also organize channels into channel
groups. A file can contain multiple channel groups. Well-grouped data
simplifies viewing and analysis and can reduce the time required to search
for a particular piece of data.

Use TDMS files when you want to store additional information about your
data. For example, you might want to record the following information:
• Type of tests or measurements
• Operator or tester name
• Serial numbers
• Unit Under Test (UUT) numbers for the device tested
• Time of the test
• Conditions under which the test or measurement was conducted

B. Binary Files
Although all file I/O methods eventually create binary files, you can directly
interact with a binary file by using the Binary File functions. The following
list describes the common functions that interact with binary files.

Open/Create/Replace File—Opens a reference to a new or existing file for

binary files as it does for ASCII Files.

76

LabVIEW Core 2 Course Manual 6-4 ni.com

 Lesson 6 File I/O Techniques

Write to Binary File—Writes binary data to a file. The function works

much like the Write to Text File function, but can accept most data types.

Read from Binary File—Reads binary data starting at its current file
position. You must specify to the function the data type to read. Use this
function to access a single data element or wire a value to the count input.
This causes the function to return an array of the specified data type.

Get File Size—Returns the size of the file in bytes. Use this function in
combination with the Read from Binary File function when you want to read
all of a binary file. Remember that if you are reading data elements that are
larger than a byte you must adjust the count to read.

Get/Set File Position—These functions get and set the location in the file
where reads and writes occur. Use these functions for random file access.

Close File—Closes an open reference to a file.

Figure 6-1 shows an example that writes an array of doubles to a binary file.
Refer to the Arrays section of this lesson for more information about the
Prepend array or string size? option.

Figure 6-1. Writing a Binary File

Binary Representation
Each LabVIEW data type is represented in a specified way when written to
a binary file. This section discusses the representation of each type and
important issues when dealing with the binary representation of each type.

Tip A bit is a single binary value. Represented by a 1 or a 0, each bit is either on or off.
A byte is a series of 8 bits.

Boolean Values
LabVIEW represents Boolean values as 8-bit values in a binary file. A value
of all zeroes represents False. Any other value represents True. This divides
files into byte-sized chunks and simplifies reading and processing files. To
77

© National Instruments Corporation 6-5 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

efficiently store Boolean values, convert a series of Boolean values into an

integer using the Boolean Array To Number function. Figure 6-2 shows two
methods for writing six Boolean values to a binary file.

Figure 6-2. Writing Boolean Values to a Binary File

Table 6-1 displays a binary representation of the file contents resulting from
running the programs in Figure 6-2. Notice that Method B is a more efficient
storage method.

Table 6-1. Results of Figure 6-2

Method A 00000001 00000001 00000000 00000001

00000000 00000001
Method B 00101011

8-bit Integers
Unsigned 8-bit integers (U8) directly correspond to bytes written to the file.
When you must write values of various types to a binary file, convert each
type into an array of U8s using the Boolean Array To Number, String to
Byte Array, Split Number, and Type Cast functions. Then, you can
concatenate the various arrays of U8s and write the resulting array to a file.
This process is unnecessary when you write a binary file that contains only
one type of data.
78

LabVIEW Core 2 Course Manual 6-6 ni.com

 Lesson 6 File I/O Techniques

Table 6-2. U8 Representation

Binary Value U8 Value

00000000 0
00000001 1
00000010 2
11111111 255

Other Integers
Multi-byte integers are broken into separate bytes and are stored in files in
either little-endian or big-endian byte order. Using the Write to Binary File
function, you can choose whether you store your data in little-endian or
big-endian format.

Little-endian byte order stores the least significant byte first, and the most
significant byte last. Big-endian order stores the most significant byte first,
and the least significant byte last.

From a hardware point of view, Intel x86 processors use the little-endian
byte order while Motorola, PowerPC and most RISC processors use the
big-endian byte order. From a software point of view, LabVIEW uses the
big-Endian byte order when handling and storing data to disk, regardless of
the platform. However, the operating system usually reflects the byte order
format of the platform it's running on. For example, Windows running on an
Intel platform usually stores data to file using the little-endian byte order. Be
aware of this when storing binary data to disk. The binary file functions of
LabVIEW have a byte order input that sets the endian form of the data.

Table 6-3. Integer Representations

Little-endian Big-endian
U32 Value Value Value
0 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000000
1 00000001 00000000 00000000 00000000
00000000 00000000 00000000 00000001
255 11111111 00000000 00000000 00000000
00000000 00000000 00000000 11111111

79

© National Instruments Corporation 6-7 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

Table 6-3. Integer Representations (Continued)

Little-endian Big-endian
U32 Value Value Value
65535 11111111 11111111 00000000 00000000
00000000 00000000 11111111 11111111
4,294,967,295 11111111 11111111 11111111 11111111
11111111 11111111 11111111 11111111

Floating-Point Numbers
Floating point numbers are stored as described by the IEEE 754 Standard
for Binary Floating-Point Arithmetic. Single-precision numerics use 32-bits
each and double-precision numerics use 64-bits each. The length of
extended-precision numerics depends on the operating system.

Strings
Strings are stored as a series of unsigned 8-bit integers, each of which is a
value in the ASCII Character Code Equivalents Table. This means that there
is no difference between writing strings with the Binary File functions and
writing them with the Text File functions.

Arrays
Arrays are represented as a sequential list of each of their elements. The
actual representation of each element depends on the element type. When
you store an array to a file you have the option of preceding the array with
a header. A header contains a 4-byte integer representing the size of each
dimension. Therefore, a 2D array with a header contains two integers,
followed by the data for the array. Figure 6-3 shows an example of writing
a 2D array of 8-bit integers to a file with a header. The prepend array or
string size? input of the Write to Binary File function enables the header.
Notice that the default value of this terminal is True. Therefore, headers are
added to all binary files by default.

80

LabVIEW Core 2 Course Manual 6-8 ni.com

 Lesson 6 File I/O Techniques

Figure 6-3. Writing a 2D Array of Unsigned Integers to a File with a Header

Table 6-4 shows the layout of the file that the code in Figure 6-3 generates.
Notice that the headers are represented as 32-bit integers even though the
data is 8-bit integers.

Table 6-4. Example Array Representation In Binary File

4 3 1 2 3 4 5 6 7 8 9 10 11 12

Clusters
Datalog files best represent clusters in binary files. Refer to the Datalog
Files section for more information.

Sequential vs. Random Access

When reading a binary file, there are two methods of accessing data. The
first is to read each item in order, starting at the beginning of a file. This is
called sequential access and works similar to reading an ASCII file. The
second is to access data at an arbitrary point within the file for random
access. For example, if you know that a binary file contains a 1D array of
32-bit integers that was written with a header and you want to access the
tenth item in the array, you could calculate the offset in bytes of that element
in the file and then read only that element. In this example, the element has
an offset of 4 (the header) + 10 (the array index) × 4 (the number of bytes in
an I32) = 44.

Sequential Access
To sequentially access all the data in a file, you can call the Get File Size
function and use the result to calculate the number of items in the file, based
on the size of each item and the layout of the file. You can then wire the
number of items to the count terminal of the Read Binary function.

81

© National Instruments Corporation 6-9 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

Figure 6-4 shows an example of this method.

Figure 6-4. Sequentially Reading an Entire File

Alternately, you can sequentially access the file one item at a time by
repeatedly calling the Read Binary function with the default count of 1.
Each read operation updates the position within the file so that you read a
new item each time read is called. When using this technique to access data
you can check for the End of File error after calling the Read Binary
function or calculate the number of reads necessary to reach the end of the
file by using the Get File Size function.

Random Access
To randomly access a binary file, use the Set File Position function to set the
read offset to the point in the file you want to begin reading. Notice that the
offset is in bytes. Therefore, you must calculate the offset based on the
layout of the file. In Figure 6-5, the VI returns the array item with the index
specified, assuming that the file was written as a binary array of
double-precision numerics with no header, like the one written by the
example in Figure 6-1.

Figure 6-5. Randomly Accessing a Binary File

Datalog Files
Datalog files are designed for storing a list of records to a file. Each record
is represented by a cluster, and can contain multiple pieces of data with any
data type. Datalog files are binary files, however, they use a different API
82

LabVIEW Core 2 Course Manual 6-10 ni.com

 Lesson 6 File I/O Techniques

than other binary files. The Datalog functions allow you to read and write
arrays of clusters to and from datalog files.

When you open a datalog file for either reading or writing, you must specify
the record type used by the file. To do this, wire a cluster of the appropriate
type to the Open/Create/Replace Datalog function. After the file is open,
you program datalog files like any other binary file. Random access is
available, although offsets are specified in records instead of bytes.

Figure 6-6 shows an example of writing a datalog file. Notice that the cluster
bundles the data and opens the datalog file.

Figure 6-6. Writing a Datalog File

Figure 6-7 shows an example of randomly accessing a datalog file. Notice

that the Record Definition cluster matches the cluster used to write the file.
If the record type wired to the Open/Create/Replace Datalog function does
not match the records in the file being opened, an error occurs.

Figure 6-7. Reading a Datalog File

Instead of using random access, you can read an entire datalog file by wiring
the output of the Get Number of Records function to the count input of the
Read Datalog function. 83

© National Instruments Corporation 6-11 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

C. TDMS Files
Creating TDMS Files
In LabVIEW, you can create TDMS Files in two ways. Use the Write to
Measurement File Express VI and Read from Measurement File Express VI
or the TDM Streaming API.

With the Express VIs you can quickly save and retrieve data from the TDMS
format. Figure 6-8 shows the configuration dialog box for the Write to
Measurement File Express VI. Notice that you can choose to create a
LabVIEW measurement data file (LVM) or TDMS file type. However, these
Express VIs give you little control over your data grouping and properties
and do not allow you to use some of the features that make TDMS files
useful, such as defining channel names and channel group names.

Figure 6-8. Creating a TDMS with Write to Measurement File Express VI

To gain access to the full capabilities of TDMS files, use the TDM
Streaming functions. Use the TDM Streaming functions to attach
descriptive information to84your data and quickly save and retrieve data.

LabVIEW Core 2 Course Manual 6-12 ni.com

 Lesson 6 File I/O Techniques

Some of the commonly used TDM Streaming functions are described in the
TDMS API section of this lesson.

Data Hierarchy
Use TDMS files to organize your data in channels and in channel groups.

A channel stores measurement signals or raw data in a TDMS file. The

signal is an array of measurement data. Each channel also can have
properties that describe the data. The data stored in the signal is stored as
binary data on disk to conserve disk space and efficiency.

A channel group is a segment of a TDMS file that contains properties to

store information as well as one or more channels. You can use channel
groups to organize your data and to store information that applies to multiple
channels.

TDMS files each contain as many channel group and channel objects as you
want. Each of the objects in a file has properties associated with it, which
creates three levels of properties you can use to store data. For example, test
conditions are stored at the file level. UUT information is stored at the
channel or channel group level. Storing plenty of information about your
tests or measurements can make analysis easier.

TDMS API
The following describes some of the most commonly used TDM Streaming
VIs and functions.
• TDMS Open—Opens a reference to a TDMS file for reading or writing.

• TDMS Write—Streams data to the specified TDMS file. It also allows

you to create channels and channel groups within your file.
• TDMS Read—Reads the specified TDMS file and returns data from the
specified channel and/or channel group.
• TDMS Set Properties—Sets the properties of the specified TDMS file,
channel group, or channel.
• TDMS Get Properties—Returns the properties of the specified TDMS
file, channel group, or channel.
• TDMS Close—Closes a reference to a TDMS File. Notice that you only
must close the file reference, any references that you acquire to channels
and channel groups close automatically when you close the file
reference.
• TDMS File Viewer—Opens the specified TDMS file and presents the
file data in the TDMS File Viewer dialog box.

85

© National Instruments Corporation 6-13 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

• TDMS List Contents—Provides a list of group and channel names

contained within the specified TDMS file.
• TDMS Defragment—Defragments the file data in the specified TDMS
data. Use this function to clean up your TDMS data when it becomes
cluttered and to increase performance.
• TDMS Flush—Flushes the system memory of all TDMS data to
maintain data security.

TDMS Programming
Writing a TDMS File
Figure 6-9 shows the simplest form of writing measurement data with the
TDMS API. This example writes data to the channel Main Channel in the
channel group Main Group.

Figure 6-9. Write Data to a TDMS File at the Channel Level

Reading a TDMS File

Figure 6-10 shows the simplest form of reading data using the TDMS API.
This example reads all the data in the channel Main Channel from channel
group Main Group and displays it in the Channel Data waveform graph.
Next, the example reads data from all the channels in the channel group
Main Group and displays it in the Group Data waveform graph.

86

LabVIEW Core 2 Course Manual 6-14 ni.com

 Lesson 6 File I/O Techniques

Figure 6-10. Read Data Back from the TDMS File

Writing TDMS Custom Properties

The TDMS data model automatically creates certain properties when some
data types are written. However, in many cases you may want to create a
property specific to your particular needs, such as UUT, serial number, and
test temperature. This task can be accomplished using the TDMS Set
Properties function with which you can write properties at the file, group, or
channel level.

The file level of a property determines which input terminals need to be

wired. For example, to write a group property, only the group name input
must be wired. To write a channel property, both the group name and
channel name inputs must be wired. But, to write a file property, the group
name and channel names inputs should be left unwired. Figure 6-11
illustrates these examples.

Figure 6-11. Write Custom Properties at Three Different Levels

before Writing Data to the File
87

© National Instruments Corporation 6-15 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

With the TDMS Set Properties function, you can specify an individual
property by itself or specify many properties by using arrays. Figure 6-11
shows two properties specified at the file level (TestID and Plant). You could
expand this to specify many more properties by increasing the size of the
array. Arrays are not necessary if only a single property, such as a serial
number, is written.

Property values can also be different data types. In Figure 6-11, string
properties are written at the file and group level. But at the channel level,
two numeric properties are written to specify the minimum and maximum
temperature.

Reading TDMS Custom Properties

When a TDMS file has been written, the properties can be read back into
LabVIEW using TDMS Get Properties function. Properties are returned
only for the level specified by the wiring of the group name and channel
name inputs. This process is similar to writing the properties, as it is shown
in Figure 6-12.

Figure 6-12. Read TDMS Properties from Three Different Levels

In this configuration, the property values are returned as an array of Variant

data because the data could be a string, double, Boolean, or another data
type. The data can be displayed on the front panel as a Variant or it can be
converted in LabVIEW to the appropriate data type. If a property name and
its data type are known, they can be wired as inputs to TDMS Get Properties
function and read directly with the correct data type.

88

LabVIEW Core 2 Course Manual 6-16 ni.com

 Lesson 6 File I/O Techniques

TDMS File Viewer

Use the TDMS File Viewer VI when developing a TDMS application to
automatically see everything that has been written to a TDMS file by
making a single VI call. The TDMS File Viewer VI is flexible and can read
complex TDMS files. The TDMS File Viewer VI is included with the
TDMS API, so it can be easily placed in a program. Place the TDMS File
Viewer VI after the file is closed to use it. The TDMS File Viewer launches
another window in which you can view the data and properties inside the
TDMS file.

Figure 6-13. TDMS File Viewer

89

© National Instruments Corporation 6-17 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

Grouping TDMS Data

Carefully consider the best way to group your data because the data
grouping can have a significant impact on the execution speed and
implementation complexity of writes and reads. Consider the original
format of your data and how you want to process or view the data when
choosing a grouping scheme.

One technique is to group data by the type of data. For example, you might
put numeric data in one channel group and string data in another, or you
might put time domain data in one group and frequency domain data in
another. This makes it easy to compare the channels in a group, but can
make it difficult to find two channels that are related to each other.
Figure 6-14 shows an example of grouping by the type of data. In this
example, the temperature data is placed in one group and the wind data is
placed in another. Each group contains multiple channels of data. Notice
that when grouping by data type you typically have a fixed number of
groups, two in this case, and a dynamically determined number of channels.

Figure 6-14. Grouping Data by Type

Another technique is to group related data. For example, you might put all
the data that applies to a single UUT in one group. Grouping related data
allows you to easily locate all the related data about a particular subject, but
makes it harder to compare individual pieces of data among subjects.
90

LabVIEW Core 2 Course Manual 6-18 ni.com

 Lesson 6 File I/O Techniques

Relational grouping helps convert cluster-based storage to a TDMS format.

You can store all the information from a given cluster in a channel group,
with arrays in the cluster representing channels within the group, and scalar
items in the cluster representing properties of the channel group.
Figure 6-15 shows an example of relational grouping.

Figure 6-15. Grouping Related Data

Notice that the input data is an array of clusters, each of which contains
multiple pieces of information about a test. Each test is stored as a separate
channel group. Information that applies to the entire test, such as Test Status,
is stored as properties of the channel group. Arrays of data, such as the time
data and power spectrum, are stored in channels, and information which
relates to the arrays of data, such as the RMS Value and Fundamental
Frequency, are stored as properties of the channels. Relational data typically
uses a fixed number of channels in a group, but the number of groups is
dynamic.

91

© National Instruments Corporation 6-19 LabVIEW Core 2 Course Manual

92
 Lesson 6 File I/O Techniques

Self-Review: Quiz
1. You need to store test results and organize the data into descriptive
groups. In the future, you need to efficiently view the test results by
group. Which file storage format should you use?
a. Tab-delimited ASCII
b. Custom binary format
c. TDMS
d. Datalog

2. You must write a program which saves Portable Network Graphics

(PNG) image files. Which file storage method should you use?
a. Storage file VIs
b. Binary file functions
c. ASCII file VIs
d. Datalog file VIs

3. You must store data that other engineers will later analyze with
Microsoft Excel. Which file storage format should you use?
a. Tab-delimited ASCII
b. Custom binary format
c. TDMS
d. Datalog

4. Which of the following is a little-endian representation of an unsigned

32-bit integer (U32) with a value of 10?
a. 00001010 00000000 00000000 00000000
b. 00000000 00000000 00000000 00001010
c. 00001010
d. 01010000 00000000 00000000 00000000

93

© National Instruments Corporation 6-21 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

5. You can use the Binary File functions to read ASCII files.
a. True
b. False

6. TDMS files store properties only at the channel or channel group level.
a. True
b. False

94

LabVIEW Core 2 Course Manual 6-22 ni.com

 Lesson 6 File I/O Techniques

Self-Review: Quiz Answers

1. You must store the results of tests to a file. In the future, you need to
efficiently search for the tests which meet specific criteria. Which file
storage format makes it easiest to query the data?
a. Tab-delimited ASCII
b. Custom binary format
c. TDMS
d. Datalog

2. You must write a program which saves Portable Network Graphics

(PNG) image files. Which file storage method should you use?
a. Storage file VIs
b. Binary file functions
c. ASCII file VIs
d. Datalog file VIs

3. You need to store data which other engineers will later analyze with
Microsoft Excel. Which file storage format should you use?
a. Tab-delimited ASCII
b. Custom binary format
c. TDMS
d. Datalog

4. Which of the following is a little endian representation of an unsigned

32-bit integer (U32) with a value of 10?
a. 00001010 00000000 00000000 00000000
b. 00000000 00000000 00000000 00001010
c. 00001010
d. 01010000 00000000 00000000 00000000

95

© National Instruments Corporation 6-23 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

5. You can use the Binary File functions to read ASCII files.
a. True
b. False

6. TDMS files store properties only at the channel or channel group level.
a. True
b. False

96

LabVIEW Core 2 Course Manual 6-24 ni.com

 Lesson 6 File I/O Techniques

Notes

97

© National Instruments Corporation 6-25 LabVIEW Core 2 Course Manual

Lesson 6 File I/O Techniques

Notes

98

LabVIEW Core 2 Course Manual 6-26 ni.com

Improving an Existing VI
7
A common problem when you inherit VIs from other developers is that
features may have been added without attention to design, thus making it
progressively more difficult to add features later in the life of the VI. This is
known as software decay. One solution to software decay is to refactor the
software. Refactoring is the process of redesigning software to make it more
readable and maintainable so that the cost of change does not increase over
time. Refactoring changes the internal structure of a VI to make it more
readable and maintainable, without changing its observable behavior.

In this lesson, you will learn methods to refactor inherited code and
experiment with typical issues that appear in inherited code.

Topics
A. Refactoring Inherited Code
B. Typical Refactoring Issues
C. Comparing VIs

99

© National Instruments Corporation 7-1 LabVIEW Core 2 Course Manual

Lesson 7 Improving an Existing VI

A. Refactoring Inherited Code

Write large and/or long-term software applications with readability in mind
because the cost of reading and modifying the software is likely to outweigh
the cost of executing the software. It costs more for a developer to read and
understand poorly designed code than it does to read code that was created
to be readable. In general, more resources are allocated to reading and
modifying software than to the initial implementation. Therefore VIs that
are easy to read and modify are more valuable than those that are not.

Creating well-designed software facilitates rapid development and

decreases possible decay. If a system starts to decay, you can spend large
amounts of time tracking down regression failures, which is not productive.
Changes also can take longer to implement because it is harder to
understand the system if it is poorly designed.

Consider the inherited VI shown in Figure 7-1.

Figure 7-1. Inherited VI

100

LabVIEW Core 2 Course Manual 7-2 ni.com

 Lesson 7 Improving an Existing VI

You can refactor the code as shown in Figure 7-2.

Figure 7-2. Refactored Inherited Code

The refactored code performs the same function as the inherited code, but
the refactored code is more readable. The inherited code violates many of
the block diagram guidelines you have learned.

When you make a VI easier to understand and maintain, you make it more
valuable because it is easier to add features to or debug the VI. The
refactoring process does not change observable behavior. Changing the way
a VI interacts with clients (users or other VIs) introduces risks that are not
present when you limit changes to those visible only to developers. The
benefit of keeping the two kinds of changes separate is that you can better
manage risks.

Refactoring versus Performance Optimization

Although you can make changes that optimize the performance of a VI, this
is not the same as refactoring. Refactoring specifically changes the internal
structure of a VI to make it easier to read, understand, and maintain.
A performance optimization is not refactoring because the goal of
optimization is not to make the VI easier to understand and modify. In fact,
performance optimization can make VIs more difficult to read and
understand, which might be an acceptable trade-off. Sometimes you must
sacrifice readability for improved performance, however, readability usually
takes priority over speed of performance.

When to Refactor
The right time to refactor is when you are adding a feature to a VI or
debugging it. Although you might be tempted to rewrite the VI from scratch,
there is value in a VI that works, even if the block diagram is not readable.
Good candidates for complete rewrites are VIs that do not work or VIs that
satisfy only a small portion of your needs. You also can rewrite simple VIs
101

© National Instruments Corporation 7-3 LabVIEW Core 2 Course Manual

Lesson 7 Improving an Existing VI

that you understand well. Consider what works well in an existing VI before
you decide to refactor.

B. Typical Refactoring Issues

When you refactor a VI, manage the risk of introducing bugs by making
small, incremental changes to the VI and testing the VI after each
change. The flowchart shown in Figure 7-3 indicates the process for
refactoring a VI.

Improve Block Diagram

Test VI

No
Is VI Acceptable?

Yes

Add Features or Use VI

Figure 7-3. Refactoring Flowchart

When you refactor to improve the block diagram, make small cosmetic
changes before tackling larger issues. For example, it is easier to find
duplicated code if the block diagram is well organized and the terminals are
well labeled.

There are several issues that can complicate working with an inherited VI.
The following sections describe typical problems and the refactoring
solutions you can use to make inherited VIs more readable.

Disorganized or Poorly Designed Block Diagram

Improve the readability of a disorganized VI by relocating objects within the
block diagram. You also can create subVIs for sections of the VI that are
disorganized. Place comments on areas of a VI that are disorganized to
improve the readability of the VI.

Overly Large Block Diagram

A VI that has a block diagram that is larger than the screen size is difficult
to read. You should refactor the VI to make it smaller. The act of scrolling
complicates reading a block diagram and understanding the code. Improve
a large block diagram by moving objects around. Another technique to
102

LabVIEW Core 2 Course Manual 7-4 ni.com

 Lesson 7 Improving an Existing VI

reduce the screen space a block diagram occupies is to create subVIs for
sections of code within the block diagram. If you cannot reduce the block
diagram to fit on the screen, limit the scrolling to one direction.

Poorly Named Objects and Poorly Designed Icons

Inherited VIs often contain controls and indicators that do not have
meaningful names. For example, the name of Control 1, shown in
Figure 7-4, does not indicate its purpose. Control 2 is the same control,
renamed to make the block diagram more readable and understandable.

1 2

1 Poorly Named Control 2 Meaningfully Named Control

Figure 7-4. Naming Controls

VI names and icons also are important for improving the readability of a VI.
For example, the name My Acq.vi, shown on the left in Figure 7-5, does
not provide any information about the purpose of the VI. You can give the
VI a more meaningful name by saving a copy of the VI with a new name and
replacing all instances of the VI with the renamed VI. A simpler method is
to open all callers of the VI you want to rename, then save the VI with a new
name. When you use this method, LabVIEW automatically relinks all open
callers of the VI to the new name. Acq Window Temperature.vi reflects
a more meaningful name for the VI.

1 2 3

1 Poorly Named VI 3 Meaningful VI Name and VI Icon

2 Meaningfully Named VI

Figure 7-5. Poorly Named SubVI

The VI icon also should clarify the purpose of the VI. The default icons used
for VI 1 and VI 2 in Figure 7-5 do not represent the purpose of the VI. You
can improve the readability of the VI by providing a meaningful icon, as
shown for VI 3.

By renaming controls and VIs and creating meaningful VI icons, you can
improve the readability of an inherited VI.
103

© National Instruments Corporation 7-5 LabVIEW Core 2 Course Manual

Lesson 7 Improving an Existing VI

Unnecessary Logic
When you read the block diagram in Figure 7-6, notice that it contains
unnecessary logic. If a portion of the block diagram does not execute, delete
it. Understanding code that executes is difficult, but trying to understand
code that never executes is inefficient and complicates the block diagram.

Figure 7-6. Unnecessary Logic

Duplicated Logic
If a VI contains duplicated logic, you always should refactor the VI by
creating a subVI for the duplicated logic. This can improve the readability
and testability of the VI.

Lack of Dataflow Programming

If there are Sequence structures and local variables on the block diagram,
the VI probably does not use data flow to determine the programming flow.

You should replace most Sequence structures with the state machine design
pattern. Delete local variables and wire the controls and indicators directly.

Complicated Algorithms
Complicated algorithms can make a VI difficult to read. Complicated
algorithms can be more difficult to refactor because there is a higher
probability that the changes introduce errors. When you refactor a
complicated algorithm, make minor changes and test the code frequently.
In some cases you can refactor a complicated algorithm by using built-in
LabVIEW functions. For example, the VI in Figure 7-7 checks a user name
and password against a database.

104

LabVIEW Core 2 Course Manual 7-6 ni.com

 Lesson 7 Improving an Existing VI

Figure 7-7. Complicated Algorithm VI

You could refactor this VI using the built-in functions for searching strings,
as shown in Figure 7-8.

Figure 7-8. Refactored VI

C. Comparing VIs
The LabVIEW Professional Development System includes a utility to
determine the differences between two VIs loaded into the memory. Select
Tools»Compare»Compare VIs to display the Compare VIs dialog box.

From this dialog box, you can select the VIs you want to compare, as well
as the characteristics of the VIs to check. When you compare the VIs, both
VIs display a Differences window that lists all differences between the
two VIs. In this window, you can select various differences and details that
you can circle for clarity.

105

© National Instruments Corporation 7-7 LabVIEW Core 2 Course Manual

Lesson 7 Improving an Existing VI

Refactoring Checklist
Use the following refactoring checklist to help determine if you should
refactor a VI. If you answer yes to any of the items in the checklist, refer to
the guidelines in the When to Refactor section of this lesson to refactor
the VI.

❑ Disorganized block diagram

❑ Overly large block diagram

❑ Poorly named objects and poorly designed icons

❑ Unnecessary logic

❑ Duplicated logic

❑ Lack of dataflow programming

❑ Complicated algorithms

106

LabVIEW Core 2 Course Manual 7-8 ni.com

 Lesson 7 Improving an Existing VI

Notes

107

© National Instruments Corporation 7-9 LabVIEW Core 2 Course Manual

Lesson 7 Improving an Existing VI

Notes

108

LabVIEW Core 2 Course Manual 7-10 ni.com

Creating and Distributing Applications
8
This lesson describes the process of creating a stand-alone application and
installer for your LabVIEW projects.

Topics
A. Preparing the Files
B. Build Specifications
C. Building the Application and Installer

109

© National Instruments Corporation 8-1 LabVIEW Core 2 Course Manual

Lesson 8 Creating and Distributing Applications

A. Preparing the Files

A stand-alone application allows the user to run your VIs without installing
the LabVIEW development system. Installers distribute the stand-alone
application. Installers can include the LabVIEW Run-Time Engine, which
is necessary for running stand-alone applications. However, you can also
download the LabVIEW Run-Time Engine at ni.com/downloads.

Before you can create a stand-alone application with your VIs, you must
first prepare your files for distribution. The following topics describe a few
of the issues you need to consider as part of your preparation. Refer to the
Preparing Files section of the Building Applications Checklist topic in the
LabVIEW Help for more information.

VI Properties
Use the VI Properties dialog box to customize the window appearance
and size. you might want to configure a VI to hide scroll bars, or you
might want to hide the buttons on the toolbar.

Path Names
Consider the path names you use in the VI. Assume you read data from a
file during the application, and the path to the file is hard-coded on the block
diagram. Once an application is built, the file is embedded in the executable,
changing the path of the file. Being aware of these issues will help you to
build more robust applications in the future.

Quit LabVIEW
In a stand-alone application, the top-level VI must quit LabVIEW or close
the front panel when it is finished executing. To completely quit and close
the top-level VI, you must call the Quit LabVIEW function on the block
diagram of the top-level VI.

External Code
Know what external code your applications uses. For example, do you call
any system or custom DLLs or shared libraries? Are you going to process
command line arguments? These are advanced examples that are beyond the
scope of this course, but you must consider them for the application. Refer
to the Using External Code in LabVIEW topic in the LabVIEW Help.

VI Server Properties and Methods

Not all VI Server properties and methods are supported in the LabVIEW
Run-Time Engine. Unsupported VI server properties and methods return
errors when called from the Run-Time Engine. Use proper error handling to
detect any of these unsupported properties and methods. You can also
110

LabVIEW Core 2 Course Manual 8-2 ni.com

 Lesson 8 Creating and Distributing Applications

review the VI Server Properties and Methods Not Supported in the

LabVIEW Run-Time Engine topic in the LabVIEW Help.

Providing Online Help in Your LabVIEW Applications

As you put the finishing touches on your application, you should provide
online help to the user. To create effective documentation for VIs, create VI
and object descriptions that describe the purpose of the VI or object and give
users instructions for using the VI or object.

Use the following functions, located on the Help palette, to

programmatically show or hide the Context Help window and link from
VIs to HTML files or compiled help files:
• Use the Get Help Window Status function to return the status and
position of the Context Help window.
• Use the Control Help Window function to show, hide, or reposition
the Context Help window.
• Use the Control Online Help function to display the table of contents,
jump to a specific topic in the file, or close the online help.
• Use the Open URL in Default Browser VI to display a URL or HTML
file in the default Web browser.

B. Build Specifications
After you have prepared your files for distribution, you need to create a build
specification for your application. The Build Specifications node in the
Project Explorer window allows you to create and configure build
specifications for LabVIEW builds. A build specification contains all the
settings for the build, such as files to include, directories to create, and
settings for VIs.

Note If you previously hid Build Specifications in the Project Explorer window, you
must display the item again to access it in the Project Explorer window.

You can create and configure the following types of build specifications:
• Stand-alone applications—Use stand-alone applications to provide
other users with executable versions of VIs. Applications are useful
when you want users to run VIs without installing the LabVIEW
development system. Stand-alone applications require the LabVIEW
Run-Time Engine. (Windows) Applications have a .exe extension.
(Mac OS) Applications have a .app extension.
• Installers—(Windows) Use installers to distribute stand-alone
applications, shared libraries, and source distributions that you create
with the Application Builder. Installers that include the LabVIEW
111

© National Instruments Corporation 8-3 LabVIEW Core 2 Course Manual

Lesson 8 Creating and Distributing Applications

Run-Time Engine are useful if you want users to be able to run

applications or use shared libraries without installing LabVIEW.
• .NET Interop Assemblies—(Windows) Use .NET interop assemblies to
package VIs for the Microsoft .NET Framework. You must install the
Microsoft .NET Framework 2.0 or higher to build a .NET interop
assembly using the Application Builder.
• Shared libraries—Use shared libraries if you want to call VIs using
text-based programming languages, such as LabWindows/CVI,
Microsoft Visual C++, and Microsoft Visual Basic. Using shared
libraries provides a way for programming languages other than
LabVIEW to access code developed with LabVIEW. Shared libraries
are useful when you want to share the functionality of the VIs you build
with other developers. Other developers can use the shared libraries but
cannot edit or view the block diagrams unless you enable debugging.
(Windows) Shared libraries have a .dll extension. (Mac OS) Shared
libraries have a .framework extension. (Linux) Shared libraries have a
.so extension. You can use .so or you can begin with lib and end with
.so, optionally followed by the version number. This allows other
applications to use the library.
• Source distributions—Use source distributions to package a collection
of source files. Source distributions are useful if you want to send code
to other developers to use in LabVIEW. You can configure settings for
specified VIs to add passwords, remove block diagrams, or apply other
settings. You also can select different destination directories for VIs in a
source distribution without breaking the links between VIs and subVIs.
• Web services (RESTful)—(Windows) Publish VIs within LabVIEW
Web services to provide a standardized method for the LabVIEW Web
Server to deploy applications that any HTTP client can access. Web
services support clients across most major platforms and programming
languages and allow you to easily implement and deploy Web
applications over a network using LabVIEW.
• Zip files—Use zip files when you want to distribute files or an entire
LabVIEW project as a single, portable file. A zip file contains
compressed files, which you can send to users. Zip files are useful if you
want to distribute selected source files to other LabVIEW users. You
also can use the Zip VIs to create zip files programmatically.

Refer to the Configuring Build Specifications section of the Building

Applications Checklist topic in the LabVIEW Help for more information.

112

LabVIEW Core 2 Course Manual 8-4 ni.com

 Lesson 8 Creating and Distributing Applications

C. Building the Application and Installer

System Requirements
Applications that you create with Build Specifications generally have the
same system requirements as the LabVIEW development system. Memory
requirements vary depending on the size of the application created.

You can distribute these files without the LabVIEW development system;
however, to run stand-alone applications and shared libraries, users must
have the LabVIEW Run-Time Engine installed.

Configuring Build Specifications

You must create build specifications in the Project Explorer window.
Expand My Computer, right-click Build Specifications, select New and
the type of build you want to configure from the shortcut menu. Use the
pages in the Source Distribution Properties, Application Properties,
Shared Library Properties, Installer Properties, or Zip File Properties
dialog boxes to configure settings for the build specification. After you
define these settings, click the OK button to close the dialog box and update
the build specification in the project. The build specification appears under
Build Specifications. Right-click a specification and select Build from the
shortcut menu to complete the build. You also can select Build All from the
shortcut menu to build all specifications under Build Specifications. If you
rebuild a given specification, LabVIEW overwrites the existing files from
the previous build that are part of the current build.

Refer to the Caveats and Recommendations for Building Installers topic in

the LabVIEW Help for more information.

113

© National Instruments Corporation 8-5 LabVIEW Core 2 Course Manual

Lesson 8 Creating and Distributing Applications

Summary
• LabVIEW features the Application Builder, which enables you to create
stand-alone executables and installers. The Application Builder is
available in the Professional Development Systems or as an add-on
package.
• Creating a professional, stand-alone application with your VIs involves
understanding the following:
– The architecture of your VI
– The programming issues particular to the VI
– The application building process
– The installer building process

114

LabVIEW Core 2 Course Manual 8-6 ni.com

 Lesson 8 Creating and Distributing Applications

Notes

115

© National Instruments Corporation 8-7 LabVIEW Core 2 Course Manual

Lesson 8 Creating and Distributing Applications

Notes

116

LabVIEW Core 2 Course Manual 8-8 ni.com

Using Variables
A
In this appendix, you learn to use variables to transfer data among multiple
loops and VIs. You also learn about the programming issues involved when
using variables and how to overcome these challenges.

Topics
A. Parallelism
B. Variables
C. Functional Global Variables
D. Race Conditions

117

© National Instruments Corporation A-1 LabVIEW Core 2 Course Manual

Appendix A Using Variables

A. Parallelism
In this course, parallelism refers to executing multiple tasks at the same
time. Consider the example of creating and displaying two sine waves at a
different frequencies. Using parallelism, you place one sine wave in a loop,
and the second sine wave in a different loop.

A challenge in programming parallel tasks is passing data among multiple

loops without creating a data dependency. For example, if you pass the data
using a wire, the loops are no longer parallel. In the multiple sine wave
example, you may want to share a single stop mechanism between the loops,
as shown in Figure A-1.

Figure A-1. Parallel Loops Front Panel

Examine what happens when you try to share data among parallel loops
with a wire using those different methods.

118

LabVIEW Core 2 Course Manual A-2 ni.com

 Appendix A Using Variables

Method 1 (Incorrect)
Place the Loop Control terminal outside of both loops and wire it to each
conditional terminal, as shown in Figure A-2. The Loop control is a data
input to both loops, therefore the Loop Control terminal is read only once,
before either While Loop begins executing. If False is passed to the loops,
the While Loops run indefinitely. Turning off the switch does not stop the
VI because the switch is not read during the iteration of either loop.

Figure A-2. Parallel Loops Method 1 Example

Method 2 (Incorrect)
Move the Loop Control terminal inside Loop 1 so that it is read in each
iteration of Loop 1, as shown in the following block diagram. Although
Loop 1 terminates properly, Loop 2 does not execute until it receives all its
data inputs. Loop 1 does not pass data out of the loop until the loop stops,
so Loop 2 must wait for the final value of the Loop Control, available only
after Loop 1 finishes. Therefore, the loops do not execute in parallel. Also,
Loop 2 executes for only one iteration because its conditional terminal
receives a True value from the Loop Control switch in Loop 1.

Figure A-3. Parallel Loops Method 2 Example

119

© National Instruments Corporation A-3 LabVIEW Core 2 Course Manual

Appendix A Using Variables

Method 3 (Solution)
If you could read the value of the loop control from a file, you would no
longer have a dataflow dependency between the loops, as each loop can
independently access the file. However, reading and writing to files can be
time consuming, at least in processor time. Another way to accomplish this
task is to find the location where the loop control data is stored in memory
and read that memory location directly. The rest of this lesson will provide
information on methods for solving this problem.

B. Variables
In LabVIEW, the flow of data rather than the sequential order of commands
determines the execution order of block diagram elements. Therefore, you
can create block diagrams that have simultaneous operations. For example,
you can run two For Loops simultaneously and display the results on the
front panel, as shown in the following block diagram.

However, if you use wires to pass data between parallel block diagrams, they
no longer operate in parallel. Parallel block diagrams can be two parallel
loops on the same block diagram without any data flow dependency, or
two separate VIs that are called at the same time.

120

LabVIEW Core 2 Course Manual A-4 ni.com

 Appendix A Using Variables

The block diagram in Figure A-4 does not run the two loops in parallel
because of the wire between the two subVIs.

Figure A-4. Data Dependency Imposed by Wire

The wire creates a data dependency, because the second loop does not start
until the first loop finishes and passes the data through its tunnel. To make
the two loops run concurrently, remove the wire. To pass data between the
subVIs, use another technique, such as a variable.

In LabVIEW, variables are block diagram elements that allow you to access
or store data in another location. The actual location of the data varies
depending on the type of the variable. Local variables store data in front
panel controls and indicators. Global variables and single-process shared
variables store data in special repositories that you can access from multiple
VIs. Functional global variables store data in While Loop shift registers.
Regardless of where the variable stores data, all variables allow you to
circumvent normal data flow by passing data from one place to another
without connecting the two places with a wire. For this reason, variables are
useful in parallel architectures, but also have certain drawbacks, such as race
conditions.

Using Variables in a Single VI

Local variables transfer data within a single VI.

In LabVIEW, you read data from or write data to a front panel object using
its block diagram terminal. However, a front panel object has only one block
diagram terminal, and your application might need to access the data in that
terminal from more than one location.

Local and global variables pass information between locations in the

application that you cannot connect with a wire. Use local variables to
access front panel objects from more than one location in a single VI. Use
global variables to access and pass data among several VIs.

Use a Feedback Node to store data from a previous VI or loop execution.

121

© National Instruments Corporation A-5 LabVIEW Core 2 Course Manual

Appendix A Using Variables

Creating Local Variables

Right-click an existing front panel object or block diagram terminal and
select Create»Local Variable from the shortcut menu to create a local
variable. A local variable icon for the object appears on the block diagram.

You also can select a local variable from the Functions palette and place it
on the block diagram. The local variable node is not yet associated with a
control or indicator.

To associate a local variable with a control or indicator, right-click the local

variable node and select Select Item from the shortcut menu. The expanded
shortcut menu lists all the front panel objects that have owned labels.

LabVIEW uses owned labels to associate local variables with front panel
objects, so label the front panel controls and indicators with descriptive
owned labels.

Reading and Writing to Variables

After you create a variable, you can read data from a variable or write data
to it. By default, a new variable receives data. This kind of variable works as
an indicator and is a write local or global. When you write new data to the
local or global variable, the associated front panel control or indicator
updates to the new data.

You also can configure a variable to behave as a data source, or a variable.

Right-click the variable and select Change To Read from the shortcut menu
to configure the variable to behave as a control. When this node executes,
the VI reads the data in the associated front panel control or indicator.

To change the variable to receive data from the block diagram rather than
provide data, right-click the variable and select Change To Write from the
shortcut menu.

On the block diagram, you can distinguish read variables from write
variables the same way you distinguish controls from indicators. A read
variable has a thick border similar to a control. A write variable has a thin
border similar to an indicator.

122

LabVIEW Core 2 Course Manual A-6 ni.com

 Appendix A Using Variables

Local Variable Example

In the Parallelism section of this lesson, you saw an example of a VI that
used parallel loops. The front panel contained a single switch that stopped
the data generation displayed on two graphs. On the block diagram, the data
for each chart is generated within an individual While Loop to allow for
separate timing of each loop. The Loop Control terminal stopped both
While Loops. In this example, the two loops must share the switch to stop
both loops at the same time.

For both charts to update as expected, the While Loops must operate in
parallel. Connecting a wire between While Loops to pass the switch data
makes the While Loops execute serially, rather than in parallel. Figure A-5
shows a block diagram of this VI using a local variable to pass the switch
data.

Loop 2 reads a local variable associated with the switch. When you set the
switch to False on the front panel, the switch terminal in Loop 1 writes a
False value to the conditional terminal in Loop 1. Loop 2 reads the Loop
Control local variable and writes a False to the Loop 2 conditional terminal.
Thus, the loops run in parallel and terminate simultaneously when you turn
off the single front panel switch.

Figure A-5. Local Variable Used to Stop Parallel Loops

With a local variable, you can write to or read from a control or indicator on
the front panel. Writing to a local variable is similar to passing data to any
other terminal. However, with a local variable you can write to it even if it
is a control or read from it even if it is an indicator. In effect, with a local
variable, you can access a front panel object as both an input and an output.

For example, if the user interface requires users to log in, you can clear the
Login and Password prompts each time a new user logs in. Use a local
variable to read from the Login and Password string controls when a user
logs in and to write empty strings to these controls when the user logs out.

123

© National Instruments Corporation A-7 LabVIEW Core 2 Course Manual

Appendix A Using Variables

Using Variables Among VIs

You also can use variables to access and pass data among several VIs that
run simultaneously. A local variable shares data within a VI. A global
variable also shares data, but it shares data among multiple VIs. For
example, suppose you have two VIs running simultaneously. Each VI
contains a While Loop and writes data points to a waveform chart. The first
VI contains a Boolean control to terminate both VIs. You can use a global
variable to terminate both loops with a single Boolean control. If both loops
were on a single block diagram within the same VI, you could use a local
variable to terminate the loops.

You also can use a single-process shared variable in the same way you use
a global variable. A shared variable is similar to a local variable or a global
variable, but allows you to share data across a network. A shared variable
can be single-process or network-published. Although network-published
shared variables are beyond the scope of this course, by using the
single-process shared variable, you can later change to a network-published
shared variable.

Use a global variable to share data among VIs on the same computer,
especially if you do not use a project file. Use a single-process shared
variable if you may need to share the variable information among VIs on
multiple computers in the future.

Creating Global Variables

Use global variables to access and pass data among several VIs that run
simultaneously. Global variables are built-in LabVIEW objects. When you
create a global variable, LabVIEW automatically creates a special global
VI, which has a front panel but no block diagram. Add controls and
indicators to the front panel of the global VI to define the data types of the
global variables it contains. In effect, this front panel is a container from
which several VIs can access data.

For example, suppose you have two VIs running simultaneously. Each VI
contains a While Loop and writes data points to a waveform chart. The first
VI contains a Boolean control to terminate both VIs. You must use a global
variable to terminate both loops with a single Boolean control. If both loops
were on a single block diagram within the same VI, you could use a local
variable to terminate the loops.

Select a global variable from the Functions palette and place it on the block
diagram.

Double-click the global variable node to display the front panel of the global
VI. Place controls and indicators on this front panel the same way you do on
a standard front panel.
124

LabVIEW Core 2 Course Manual A-8 ni.com

 Appendix A Using Variables

LabVIEW uses owned labels to identify global variables, so label the front
panel controls and indicators with descriptive owned labels.

You can create several single global VIs, each with one front panel object,
or if you want to group similar variables together, you can create one global
VI with multiple front panel objects.

You can create several single global variables, each with one front panel
object, or you can create one global variable with multiple front panel
objects.

A global variable with multiple objects is more efficient because you can
group related variables together. The block diagram of a VI can include
several global variable nodes that are associated with controls and indicators
on the front panel of a global variable. These global variable nodes are either
copies of the first global variable node that you placed on the block diagram
of the global VI, or they are the global variable nodes of global VIs that you
placed on the current VI. You place global VIs on other VIs the same way
you place subVIs on other VIs. Each time you place a new global variable
node on a block diagram, LabVIEW creates a new VI associated only with
that global variable node and copies of it.

Figure A-6 shows a global variable front panel window with a numeric, a
string, and a cluster containing a numeric and a Boolean control. The toolbar
does not show the Run, Stop, or related buttons as a normal front panel
window.

Figure A-6. Global Variable Front Panel Window

After you finish placing objects on the global VI front panel, save it and
return to the block diagram of the original VI. You must then select the
object in the global VI that you want to access. Click the global variable
node and select a front panel object from the shortcut menu. The shortcut
125

© National Instruments Corporation A-9 LabVIEW Core 2 Course Manual

Appendix A Using Variables

menu lists all the front panel objects in the global VI that have owned labels.
You also can right-click the global variable node and select a front panel
object from the Select Item shortcut menu.

You also can use the Operating tool or Labeling tool to click the global
variable node and select the front panel object from the shortcut menu.

If you want to use this global variable in other VIs, select the Select a VI
option on the Functions palette. By default, the global variable is associated
with the first front panel object with an owned label that you placed in the
global VI. Right-click the global variable node you placed on the block
diagram and select a front panel object from the Select Item shortcut menu
to associate the global variable with the data from another front panel object.

Creating Single-Process Shared Variables

You must use a project file to use a shared variable. To create a
single-process shared variable, right-click My Computer in the Project
Explorer window and select New»Variable. The Shared Variable
Properties dialog box appears, as shown in Figure A-7.

Figure A-7. Shared Variable Properties Dialog Box

Under Variable Type, select Single Process. Give the variable a name and
a data type. After you create the shared variable, it automatically appears in
a new library in your project file. Save the library. You can add additional
shared variables to this library as needed. You can drag and drop the variable
from the listing in the Project Explorer window directly to the block
126

LabVIEW Core 2 Course Manual A-10 ni.com

 Appendix A Using Variables

diagram. Use the short-cut menu to switch between writing or reading. Use
the error clusters on the variable to impose data flow.

Using Variables Carefully

Local and global variables are advanced LabVIEW concepts. They are
inherently not part of the LabVIEW dataflow execution model. Block
diagrams can become difficult to read when you use local and global
variables, so you should use them carefully. Misusing local and global
variables, such as using them instead of a connector pane or using them to
access values in each frame of a sequence structure, can lead to unexpected
behavior in VIs. Overusing local and global variables, such as using them to
avoid long wires across the block diagram or using them instead of data
flow, slows performance.

Variables often are used unnecessarily. The example in Figure A-8 shows a
traffic light application implemented as a state machine. Each state updates
the lights for the next stage of the light sequence. In the state shown, the east
and west traffic has a green light, while the north and south traffic has a red
light. This stage waits for 4 seconds, as shown by the Wait (ms) function.

Figure A-8. Too Many Variables Used

127

© National Instruments Corporation A-11 LabVIEW Core 2 Course Manual

Appendix A Using Variables

The example shown in Figure A-9 accomplishes the same task, but more
efficiently and using a better design. Notice that this example is much easier
to read and understand than the previous example, mostly by reducing
variable use. By placing the indicators in the While Loop outside the Case
structure, the indicators can update after every state without using a variable.
This example is less difficult to modify for further functionality, such as
adding left turn signals, than the previous example.

Figure A-9. Reduced Variables

Initializing Variables
To initialize a local or global variable, verify that the variable contains
known data values before the VI runs. Otherwise, the variables might
contain data that causes the VI to behave incorrectly. If the variable relies on
a computation result for the initial value, make sure LabVIEW writes the
value to the variable before it attempts to access the variable for any other
action. Wiring the write action in parallel with the rest of the VI can cause a
race condition.

To make sure it executes first, you can isolate the code that writes the initial
value for the variable to the first frame of a sequence structure or to a subVI
and wire the subVI to execute first in the data flow of the block diagram.

If you do not initialize the variable before the VI reads the variable for the
first time, the variable contains the default value of the associated front
panel object.

128

LabVIEW Core 2 Course Manual A-12 ni.com

 Appendix A Using Variables

Figure A-10 shows a common mistake when using variables. A shared

variable synchronizes the stop conditions for two loops. This example
operates the first time it runs, because the default value of a Boolean is False.
However, each time this VI runs, the Stop control writes a True value to the
variable. Therefore, the second and subsequent times that this VI runs, the
lower loop stops after only a single iteration unless the first loop updates the
variable quickly enough.

Figure A-10. Failing to Initialize a Shared Variable

Figure A-11 shows the VI with code added to initialize the shared variable.
Initialize the variable before the loops begin to insure that the second loop
does not immediately stop.

Figure A-11. Initializing a Shared Variable Properly

129

© National Instruments Corporation A-13 LabVIEW Core 2 Course Manual

Appendix A Using Variables

C. Functional Global Variables

You can use uninitialized shift registers in For Loops or While Loops to
store data as long as the VI is in memory. The shift register holds the last
state of the shift register. Place a While Loop within a subVI and use the
shift registers to store data that can be read from or written to. Using this
technique is similar to using a global variable. This method is often called a
functional global variable. The advantage to this method over a global
variable is that you can control access to the data in the shift register. The
general form of a functional global variable includes an uninitialized shift
register with a single iteration For Loop or While Loop, as shown in
Figure A-12.

1 Uninitialized Shift Register

Figure A-12. Functional Global Variable Format

A functional global variable usually has an action input parameter that

specifies which task the VI performs. The VI uses an uninitialized shift
register in a While Loop to hold the result of the operation.

130

LabVIEW Core 2 Course Manual A-14 ni.com

 Appendix A Using Variables

Figure A-13 shows a simple functional global variable with set and get
functionality.

Figure A-13. Functional Global Variable with Set and Get Functionality

In this example, data passes into the VI and the shift register stores the data
if you configure the enumerated data type to Set. Data is retrieved from the
shift register if the enumerated data type is configured to Get.

Although you can use functional global variables to implement simple

global variables, as shown in the previous example, they are especially
useful when implementing more complex data structures, such as a stack or
a queue buffer. You also can use functional global variables to protect access
to global resources, such as files, instruments, and data acquisition devices,
that you cannot represent with a global variable.

Note A functional global variable is a subVI that is not reentrant. This means that when
the subVI is called from multiple locations, the same copy of the subVI is used.
Therefore, only one call to the subVI can occur at a time.

131

© National Instruments Corporation A-15 LabVIEW Core 2 Course Manual

Appendix A Using Variables

Using Functional Global Variables for Timing

One powerful application of functional global variables is to perform timing
in your VI. Many VIs that perform measurement and automation require
some form of timing. Often an instrument or hardware device needs time to
initialize, and you must build explicit timing into your VI to take into
account the physical time required to initialize a system. You can create a
functional global variable that measures the elapsed time between each time
the VI is called, as shown in Figure A-14.

Figure A-14. Elapsed Time Functional Global Variable

The Elapsed Time case gets the current date and time in seconds and
subtracts it from the time that is stored in the shift register. The Reset Time
case initializes the functional global variable with a known time value.

The Elapsed Time Express VI implements the same functionality as this

functional global variable. The benefit of using the functional global
variable is that you can customize the implementation easily, such as adding
a pause option.

132

LabVIEW Core 2 Course Manual A-16 ni.com

 Appendix A Using Variables

D. Race Conditions
A race condition occurs when the timing of events or the scheduling of tasks
unintentionally affects an output or data value. Race conditions are a
common problem for programs that execute multiple tasks in parallel and
share data between them. Consider the following example in Figure A-15
and Figure A-16.

Figure A-15. Race Condition Example: Loop 1

Figure A-16. Race Condition Example: Loop 2

The two loops both increment a shared variable during each iteration. If you
run this VI, the expected result after clicking the Stop button is that the
Total Count is equal to the sum of Count 1 and Count 2. If you run the VI
for a short period of time, you generally see the expected result. However, if
you run the VI for a longer period of time, the Total Count is less than the
sum of Count 1 and Count 2, because this VI contains a race condition.

133

© National Instruments Corporation A-17 LabVIEW Core 2 Course Manual

Appendix A Using Variables

On a single processor computer, actions in a multi-tasking program like this

example actually happen sequentially, but LabVIEW and the operating
system rapidly switch tasks so that the tasks effectively execute at the same
time. The race condition in this example occurs when the switch from one
task to the other occurs at a certain time. Notice that both of the loops
perform the following operations:
• Read the shared variable.
• Increment the value read.
• Write the incremented value to the shared variable.

Now consider what happens if the loop operations happen to be scheduled

in the following order:
1. Loop 1 reads the shared variable.
2. Loop 2 reads the shared variable.
3. Loop 1 increments the value it read.
4. Loop 2 increments the value it read.
5. Loop 1 writes the incremented value to the shared variable.
6. Loop 2 writes the incremented value to the shared variable.

In this example, both loops write the same value to the variable, and the
increment of the first loop is effectively overwritten by Loop 2. This
generates a race condition, which can cause serious problems if you intend
the program to calculate an exact count.

In this particular example, there are few instructions between when the
shared variable is read and when it is written. Therefore, the VI is less likely
to switch between the loops at the wrong time. This explains why this VI
runs accurately for short periods and only loses a few counts for longer
periods.

Race conditions are difficult to identify and debug, because the outcome
depends upon the order in which the operating system executes scheduled
tasks and the timing of external events. The way tasks interact with each
other and the operating system, as well as the arbitrary timing of external
events, make this order essentially random. Often, code with a race
condition can return the same result thousands of times in testing, but still
can return a different result, which can appear when the code is in use.

The best way to avoid race conditions is by using the following techniques:
• Controlling and limiting shared resources.
• Identifying and protecting critical sections within your code.
• Specifying execution order.
134

LabVIEW Core 2 Course Manual A-18 ni.com

 Appendix A Using Variables

Controlling and Limiting Shared Resources

Race conditions are most common when two tasks have both read and write
access to a resource, as is the case in the previous example. A resource is any
entity that is shared between the processes. When dealing with race
conditions, the most common shared resources are data storage, such as
variables. Other examples of resources include files and references to
hardware resources.

Allowing a resource to be altered from multiple locations often introduces

the possibility for a race condition. Therefore, an ideal way to avoid race
conditions is to minimize shared resources and the number of writers to the
remaining shared resources. In general, it is not harmful to have multiple
readers or monitors for a shared resource. However, try to use only one
writer or controller for a shared resource. Most race conditions only occur
when a resource has multiple writers.

In the previous example, you can reduce the dependency upon shared
resources by having each loop maintain its count locally. Then, share the
final counts after clicking the Stop button. This involves only a single read
and a single write to a shared resource and eliminates the possibility of a
race condition. If all shared resources have only a single writer or controller,
and the VI has a well sequenced instruction order, then race conditions do
not occur.

Protecting Critical Sections

A critical section of code is code that must behave consistently in all
circumstances. When you use multi-tasking programs, one task may
interrupt another task as it is running. In nearly all modern operating
systems, this happens constantly. Normally, this does not have any effect
upon running code, however, when the interrupting task alters a shared
resource that the interrupted task assumes is constant, then a race condition
occurs.

Figure A-15 and Figure A-16 contain critical code sections. If one of the
loops interrupts the other loop while it is executing the code in its critical
section, then a race condition can occur. One way to eliminate race
conditions is to identify and protect the critical sections in your code. There
are many techniques for protecting critical sections. Two of the most
effective are functional global variables and semaphores.

135

© National Instruments Corporation A-19 LabVIEW Core 2 Course Manual

Appendix A Using Variables

Functional Global Variables

One way to protect critical sections is to place them in subVIs. You can only
call a non-reentrant subVI from one location at a time. Therefore, placing
critical code in a non-reentrant subVI keeps the code from being interrupted
by other processes calling the subVI. Using the functional global variable
architecture to protect critical sections is particularly effective, because
shift registers can replace less protected storage methods like global or
single-process shared variables. Functional global variables also encourage
the creation of multi-functional subVIs that handle all tasks associated with
a particular resource.

After you identify each section of critical code in your VI, group the sections
by the resources they access, and create one functional global variable for
each resource. Critical sections performing different operations each can
become a command for the functional global variable, and you can group
critical sections that perform the same operation into one command, thereby
re-using code.

You can use functional global variables to protect critical sections of code in
Figure A-15 and Figure A-16. To remove the race condition, replace the
shared variables with a functional global variable and place the code to
increment the counter within the functional global variable, as shown in
Figure A-17, Figure A-18, and Figure A-19.

Figure A-17. Using Functional Global Variables to Protect the

Critical Section in Loop 1

136

LabVIEW Core 2 Course Manual A-20 ni.com

 Appendix A Using Variables

Figure A-18. Using Functional Global Variables to Protect the

Critical Section in Loop 2

Figure A-19. Functional Global Variable Eliminates the Race Condition

Semaphores
Semaphores are synchronization mechanisms specifically designed to
protect resources and critical sections of code. You can prevent critical
sections of code from interrupting each other by enclosing each between an
Acquire Semaphore and Release Semaphore VI. By default, a semaphore
only allows one task to acquire it at a time. Therefore, after one of the tasks
enters a critical section, the other tasks cannot enter their critical sections
until the first task completes. When done properly, this eliminates the
possibility of a race condition.
137

© National Instruments Corporation A-21 LabVIEW Core 2 Course Manual

Appendix A Using Variables

You can use semaphores to protect the critical sections of the VIs, as shown
in Figure A-15 and Figure A-16. A named semaphore allows you to share
the semaphore between VIs. You must open the semaphore in each VI, then
acquire it just before the critical section and release it after the critical
section. Figure A-20 and Figure A-21 show a solution to the race condition
using semaphores.

Figure A-20. Protecting the Critical Section with a Semaphore in Loop 1

Figure A-21. Protecting the Critical Section with a Semaphore in Loop 2

138

LabVIEW Core 2 Course Manual A-22 ni.com

 Appendix A Using Variables

Specifying Execution Order

Code in which data flow is not properly used to control the execution order
can cause some race conditions. When a data dependency is not established,
LabVIEW can schedule tasks in any order, which creates the possibility for
race conditions if the tasks depend upon each other. Consider the example
in Figure A-22.

Figure A-22. Simple Race Condition

The code in this example has four possible outcomes, depending on the
order in which the operations execute.

Outcome 1: Value = (Value × 5) + 2

1. Terminal reads Value.
2. Value × 5 is stored in Value.
3. Local variable reads Value × 5.
4. (Value × 5) + 2 is stored in Value.

Outcome 2: Value = (Value + 2) × 5

1. Local variable reads Value.
2. Value + 2 is stored in Value.
3. Terminal reads Value+2.
4. (Value + 2) × 5 is stored in Value.

Outcome 3: Value = Value × 5

1. Terminal reads Value.
2. Local variable reads Value.
3. Value + 2 is stored in Value.
4. Value × 5 is stored in Value.

139

© National Instruments Corporation A-23 LabVIEW Core 2 Course Manual

Appendix A Using Variables

Outcome 4: Value = Value + 2

1. Terminal reads Value.
2. Local variable reads Value.
3. Value × 5 is stored in Value.
4. Value + 2 is stored in Value.

Although this code is considered a race condition, the code generally

behaves less randomly than the first race condition example because
LabVIEW usually assigns a consistent order to the operations. However,
you should avoid situations such as this one because the order and the
behavior of the VI can vary. For example, the order could change when
running the VI under different conditions or when upgrading the VI to a
newer version of LabVIEW. Fortunately, race conditions of this nature are
easily remedied by controlling the data flow.

140

LabVIEW Core 2 Course Manual A-24 ni.com

 Appendix A Using Variables

Self-Review: Quiz
1. You should use variables frequently in your VIs.
a. True
b. False

2. Which of the following cannot transfer data?

a. Semaphores
b. Functional global variables
c. Local variables
d. Single process shared variables

3. Which of the following must be used within a project?

a. Local variable
b. Global variable
c. Functional global variable
d. Single-process shared variable

4. Which of the following cannot be used to pass data between

multiple VIs?
a. Local variable
b. Global variable
c. Functional global variable
d. Single-process shared variable

141

© National Instruments Corporation A-25 LabVIEW Core 2 Course Manual

142
 Appendix A Using Variables

Self-Review: Quiz Answers

1. You should use variables frequently in your VI.
a. True
b. False
You should use variables only when necessary. Use wires to transfer
data whenever possible.

2. Which of the following cannot transfer data?

a. Semaphores
b. Functional global variables
c. Local variables
d. Single process shared variables

3. Which of the following must be used within a project?

a. Local variable
b. Global variable
c. Functional global variable
d. Single-process shared variable

4. Which of the following cannot be used to pass data between multiple

VIs?
a. Local variable
b. Global variable
c. Functional global variable
d. Single-process shared variable

143

© National Instruments Corporation A-27 LabVIEW Core 2 Course Manual

Appendix A Using Variables

Notes

144

LabVIEW Core 2 Course Manual A-28 ni.com