MATLAB Intro 5 Third plot

Step 5: Plot σ_x vs. r_i, Take 3: File Input/Output

Read in [A] from a File

Instead of specifying [A] within the program, let's give the user the option of reading it in from a text file called, say, input_data.txt.

Create a text file called input_data.txt in your working directory with the elements of [A], each row being on a separate line:
Keep in mind that we have used whitespace to delimit (i.e. differentiate) the different numbers. We'll read in this data using the dlmread command. Check the documentation by searching the help for dlmread.

Scroll down to look at the example under Remarks.

Consider the following statement:

A=dlmread('input_data.txt')

This will read in [A] from the specified file. Let's test this at the command line before we add it to our program:

Check A in the Workspace. Note that each line in the text file is interpreted as a matrix row.

We'll create a new program called beam4.m starting from beam3.m. Select beam3.m in the Editor. Select

Editor > File > Save As

Enter beam4.m for filename. Confirm that you are editing beam4.m. Comment out the original A specification statement and replace it with the above statement using dlmread. Add a semi-colon at the end of the statement.

Save beam4.m. Run it from the command line.

Does your result look right?

Read in [B] from a File

Hey, if [A] can be read in from a file, why not read [B] from the same file?

Append the elements of [B] to input_data.txt and save the file:

Now we need to tell MATLAB to:

1. Read in the first two rows and assign it to A.
2. Read in the last two rows and assign it to B.

An additional complication is that A and B have different number of columns. To handle this mess, we turn to the following nugget from the dlmread documentation:

Key point: MATLAB starts counting the rows and colums at 0 rather than 1. So the first row is 0, second row is 1 and so on. The same thing applies to columns also. See below for an input with n rows and m columns. (i, j) are the row, column coordinates.

[ (0,0) (0,1) (0,2)...(0,m-1) ]

[ (1,0) _____________________ ]

[ (2,0) _____________________ ]

[ ... _______________________ ]

[(n-1,0) ___________ (n-1,m-1)]

Once again, edit beam4.m. Now, use the <span style="color: #660099"><strong><em>dlmread</em></strong></span> command with the delimiter and range inputs to extract both <span style="color: #660099"><strong><em>A</em></strong></span> and <span style="color: #660099"><strong><em>B</em></strong></span> at the same time. The delimiter should be '', the range should be [0 0 1 1] for <span style="color: #660099"><strong><em>A</em></strong></span> and [2 0 3 0] for <span style="color: #660099"><strong><em>B</em></strong></span>. Remember to add your semi-colons.

Save and run beam4.m again. You should still get the same results. We can now move on to the next part of the tutorial.

Save the contents of a matrix to a text file

Just as you can read in values from a text file to a matrix, you can also write a matrix to a text file. The function for this is call dlmwrite. Consider the following line of code:

Output Important Values in Command Window

The next step is to print out data in the command window. Suppose we are asked to report the numerical values of the bending stress for inner radius =0.5, 1.0, and 1.5 cm. These values are already contained in the vectors "ri" and "sigma_x".

The "fprintf" command is useful for outputting information to the command window. You can include both text and numbers in your output and choose the precision of the numbers to match your needs. The documentation for this command can be found by searching help for fprintf

The basic form is as follows:

fprintf('String with references to variables\n', variables)

The "\n" starts a new line at that point in the statement. Let's take a simple example. Suppose you want to output a number and a percentage with text in between. The variable "sp" is found to be 0.962 and the variable "year" is 2005. Since 2005 is a whole number, this is easy to output using "fprintf" by using "%d". If you are using a decimal number or one that may be inexact, like the percentage, then use "%f". If you put numbers and a decimal between the % and the f, you can control how many number spaces there are before and after the decimal. Let's try it with our example.

>> sp = 0.962;

>> year = 2006;

>> fprintf('In %d, Dave McKee had a %2.2f save percentage playing for Cornell.\n\n', year, sp*100)

This will return the following in the command window:

In 2006, Dave McKee had a 96.20 save percentage playing for Cornell.

Notice that we can perform arthimetic operations in the output section and specify the number of decimals to be 2. Also notice that "\n" switches to a new line and "\t" will produce an indent.

The same concept can be used to output the stress at 0.5, 1.0, and 1.5 cms. We know that since sigma_x increments by 0.1 each time, we can pull out the values of sigma_x corresponding to the correct row numbers (1, 6, and 11). Try this command out and see if it makes sense.

fprintf('The values of the stress are:\n \t %3.2f MPa \t %3.2f MPa \t %3.2f MPa.\n\n', sigma_x(1), sigma_x(6), sigma_x(11));

Go to Step 6: Plot σ_x vs. r_i, Take 4: Functions

See and rate the complete Learning Module

Go to all MATLAB Learning Modules