How to perform an equivalent partitioning without using boundary values?

Partitioning into equivalence classes without using boundary values involves identifying sets of input data that are expected to be processed similarly by the system. Instead of focusing on the range boundaries, typical representatives of each class are determined.

The algorithm is as follows:

1. Analyze requirements: understand the functionality and identify which input data influence the system's behavior.
2. Identify classes: divide input data into groups representing similar situations or processing conditions. Data within each class should behave identically.
3. Select representatives: choose one or more typical representatives for each class, without focusing on extreme values.
4. Create test cases: use the selected representatives to form test cases.

Example: A function accepts a number from 1 to 100.

Instead of boundaries (1, 100), classes can be distinguished:

• Numbers in the range (1, 100): for example, 5, 50, 95.
• Numbers outside the range (< 1): for example, -10, 0.
• Numbers outside the range (> 100): for example, 110, 200.

For each class, a representative is chosen, for example:

• Class "Numbers in the range": 50
• Class "Numbers outside the range (< 1)": 0
• Class "Numbers outside the range (> 100)": 150

Thus, the test cases will include these representatives, without explicit emphasis on boundary values 1 and 100.