CSV file handling in MATLAB - Step-by-Step Execution

Concept Flow - CSV file handling

Open CSV file

↓

Read header line

↓

Read each data line

↓

Parse line into values

↓

Store values in array/table

↓

Repeat until end of file

↓

Close file

This flow shows how MATLAB reads a CSV file line by line, parses values, stores them, and closes the file.

Execution Sample

MATLAB

filename = 'data.csv';
fileID = fopen(filename, 'r');
header = fgetl(fileID);
data = [];
while ~feof(fileID)
  line = fgetl(fileID);
  values = sscanf(line, '%f,%f,%f');
  data = [data; values'];
end
fclose(fileID);

This code opens a CSV file, reads the header, then reads and parses each data line into a numeric array.

Execution Table

Step	Action	Variable Values	Output/Effect
1	Open file 'data.csv'	fileID = valid file handle	File opened for reading
2	Read header line	header = 'A,B,C'	Header line stored
3	Check end of file	feof(fileID) = false	Continue loop
4	Read next line	line = '1.0,2.0,3.0'	Line read as string
5	Parse line	values = [1.0; 2.0; 3.0]	Values converted to numbers
6	Append to data	data = [1.0 2.0 3.0]	Data array updated
7	Check end of file	feof(fileID) = false	Continue loop
8	Read next line	line = '4.0,5.0,6.0'	Line read as string
9	Parse line	values = [4.0; 5.0; 6.0]	Values converted to numbers
10	Append to data	data = [1.0 2.0 3.0; 4.0 5.0 6.0]	Data array updated
11	Check end of file	feof(fileID) = true	Exit loop
12	Close file	fileID = closed	File closed

💡 End of file reached, loop stops reading lines

Variable Tracker

Variable	Start	After Step 4	After Step 6	After Step 8	After Step 10	Final
fileID	undefined	valid handle	valid handle	valid handle	valid handle	closed
header	undefined	'A,B,C'	'A,B,C'	'A,B,C'	'A,B,C'	'A,B,C'
line	undefined	'1.0,2.0,3.0'	'1.0,2.0,3.0'	'4.0,5.0,6.0'	'4.0,5.0,6.0'	'4.0,5.0,6.0'
values	undefined	undefined	[1.0;2.0;3.0]	[1.0;2.0;3.0]	[4.0;5.0;6.0]	[4.0;5.0;6.0]
data	[]	[]	[1.0 2.0 3.0]	[1.0 2.0 3.0]	[1.0 2.0 3.0; 4.0 5.0 6.0]	[1.0 2.0 3.0; 4.0 5.0 6.0]

Key Moments - 3 Insights

Why do we use fgetl(fileID) instead of fscanf for reading lines?

fgetl reads one line as a string, allowing us to parse it manually. fscanf reads formatted data but may skip lines or not handle mixed data well. See steps 2,4,8 in execution_table.

Why do we transpose values with values' before appending to data?

sscanf returns a column vector; transposing makes it a row vector to append as a new row in data. See steps 5 and 6.

What stops the while loop from running forever?

feof(fileID) returns true when the file end is reached, causing the loop to exit. See steps 3,7,11.

Visual Quiz - 3 Questions

Look at the execution_table at step 6, what is the content of data?

A []

B [1.0; 2.0; 3.0]

C [1.0 2.0 3.0]

D [4.0 5.0 6.0]

💡 Hint

Check the 'Variable Values' column at step 6 in execution_table.

At which step does the loop stop reading lines from the file?

A Step 10

B Step 11

C Step 12

D Step 9

💡 Hint

Look for feof(fileID) becoming true in execution_table.

If the CSV file had 4 columns instead of 3, how would the sscanf format string change?

A '%f,%f,%f,%f'

B '%f,%f'

C '%f,%f,%f'

D '%f'

💡 Hint

sscanf format string matches the number of columns; see code in execution_sample.

Concept Snapshot

CSV file handling in MATLAB:
- Open file with fopen(filename, 'r')
- Read header line with fgetl(fileID)
- Loop: read each line with fgetl
- Parse line with sscanf(line, '%f,%f,%f')
- Append parsed data to array
- Close file with fclose(fileID)

Full Transcript

This example shows how MATLAB reads a CSV file step-by-step. First, the file is opened for reading. Then the header line is read and stored. The program enters a loop where it checks if the end of file is reached. If not, it reads the next line as a string. This line is parsed into numbers using sscanf with a format matching the CSV columns. The parsed numbers are transposed and appended as a new row to the data array. This repeats until the end of the file is reached, then the file is closed. Variables like fileID, header, line, values, and data change values as the program runs. Key points include using fgetl to read lines, transposing parsed values to append rows, and using feof to stop the loop. The visual quiz tests understanding of data content after steps, loop exit condition, and format string for parsing.

Practice

1. Which MATLAB function is best suited to read numeric data from a CSV file into a matrix?

easy

A. csvread

B. readtable

C. readmatrix

D. load

Solution

1. Step 1: Understand the data type

   Numeric data is best handled as a matrix in MATLAB.

2. Step 2: Choose the appropriate function

   readmatrix is designed to read numeric CSV data directly into a matrix.

3. Final Answer:

   readmatrix -> Option C

4. Quick Check:

   Numeric CSV data = readmatrix [OK]

Hint: Use readmatrix for numeric CSV files quickly [OK]

Common Mistakes:

• Using readtable for purely numeric data
• Using load which is for MAT files
• Confusing csvread with readmatrix (csvread is older)

2. Which of the following is the correct syntax to write a matrix A to a CSV file named data.csv in MATLAB?

easy

A. writetable('data.csv', A);

B. writecsv(A, 'data.csv');

C. csvwrite(A, 'data.csv');

D. writematrix(A, 'data.csv');

Solution

1. Step 1: Identify the correct function for matrix writing

   writematrix is the modern function to write numeric matrices to CSV files.

2. Step 2: Check syntax correctness

   The syntax writematrix(A, 'data.csv'); correctly specifies the matrix and filename.

3. Final Answer:

   writematrix(A, 'data.csv'); -> Option D

4. Quick Check:

   Write matrix to CSV = writematrix [OK]

Hint: Use writematrix(matrix, filename) to save numeric CSV [OK]

Common Mistakes:

• Using non-existent function writecsv
• Using csvwrite which is outdated
• Using writetable for numeric matrix without conversion

3. What will be the output of the following MATLAB code?

A = [1, 2; 3, 4];
writematrix(A, 'test.csv');
B = readmatrix('test.csv');
disp(B);

medium

A. [1 2; 3 4]

B. [1 2 3 4]

C. [1; 2; 3; 4]

D. Error: File not found

Solution

1. Step 1: Write matrix to CSV

   The matrix A is saved as a 2x2 CSV file with rows [1,2] and [3,4].

2. Step 2: Read matrix from CSV

   readmatrix reads the CSV back into the same 2x2 numeric matrix.

3. Final Answer:

   [1 2; 3 4] -> Option A

4. Quick Check:

   Write and read matrix preserves shape [OK]

Hint: writematrix and readmatrix keep matrix shape intact [OK]

Common Mistakes:

• Expecting a single row vector output
• Confusing disp output format with array shape
• Assuming file read error without saving first

4. Consider this MATLAB code snippet:

T = readtable('data.csv');
writetable(T, 'output.csv');

If data.csv contains mixed numeric and text data, what is a common error to watch for?

medium

A. Using csvwrite to write tables causes incorrect output

B. Using writetable on a numeric matrix causes syntax error

C. Forgetting to close the file after reading causes runtime error

D. Using readmatrix instead of readtable causes syntax error

Solution

1. Step 1: Understand data type in CSV

   Mixed numeric and text data require table format to preserve all data correctly.

2. Step 2: Identify correct write function

   csvwrite only supports numeric data and will fail or produce incorrect output for tables.

3. Final Answer:

   Using csvwrite to write tables causes incorrect output -> Option A

4. Quick Check:

   csvwrite not for tables = error [OK]

Hint: Use writetable for mixed data, not csvwrite [OK]

Common Mistakes:

• Using readmatrix for mixed data losing text
• Assuming file must be closed manually in MATLAB
• Using writetable on numeric matrix without conversion

5. You have a CSV file with columns: Name (text), Age (numeric), and Score (numeric). You want to read it, increase all Score values by 10%, and save the updated data back to a CSV file. Which MATLAB code snippet correctly does this?

hard

A. data = csvread('file.csv');
data(:,3) = data(:,3) * 1.1;
csvwrite('file.csv', data);

B. T = readtable('file.csv');
T.Score = T.Score * 1.1;
writetable(T, 'file.csv');

C. T = readtable('file.csv');
T(:,3) = T(:,3) * 1.1;
writetable(T, 'file.csv');

D. A = readmatrix('file.csv');
A(:,3) = A(:,3) * 1.1;
writematrix(A, 'file.csv');

Solution

1. Step 1: Read mixed data correctly

   Use readtable to read CSV with text and numeric columns preserving all data types.

2. Step 2: Update the numeric Score column

   Access the Score column by name and multiply by 1.1 to increase by 10%.

3. Step 3: Save updated table back to CSV

   Use writetable to save the modified table preserving text and numeric data.

4. Final Answer:

   T = readtable('file.csv'); T.Score = T.Score * 1.1; writetable(T, 'file.csv'); -> Option B

5. Quick Check:

   Mixed data update = readtable + modify column + writetable [OK]

Hint: Use readtable and writetable for mixed data updates [OK]

Common Mistakes:

• Using readmatrix loses text column
• Trying to multiply table columns by index incorrectly
• Using csvread/csvwrite which don't support text