Profiling with timestamp
30 secondes minimum execution
Calculating the differential between code implementing bad code and code implementing good code practice
Validation
The rule is validated by the measurement.
The code used to validate:
public class Main {
public static void main(String[] args) {
new Main().runTest();
}
private void executeBad() {
int index = 1;
int nb = 2;
if (nb == 0) {
nb = index;
} else if (nb == 1) {
nb = index * 2;
} else if (nb == 2) {
nb = index * 3;
} else {
nb = -1;
}
}
private void executeGood() {
int index = 1;
int nb = 2;
switch (nb) {
case 0:
nb = index;
break;
case 1:
nb = index * 2;
break;
case 2:
nb = index * 3;
break;
default:
nb = -1;
}
}
public void runTest() {
// Initialize
long totalGood = 0;
long totalBad = 0;
long nb = 10;
long nbIteration = 0;
long beginTest = System.currentTimeMillis();
// Run test
do {
// Test good
long begin = System.currentTimeMillis();
for(int i = 0; i < nb; i++) {
executeGood();
}
long end = System.currentTimeMillis();
totalGood += end - begin;
// Test bad
begin = System.currentTimeMillis();
for(int i = 0; i < nb; i++) {
executeBad();
}
end = System.currentTimeMillis();
totalBad += end - begin;
// Update counters
nbIteration += nb;
// minimum 30 seconds of test
} while((System.currentTimeMillis() - beginTest) < 60_000);
// Print results
printResults(totalGood, totalBad, nbIteration);
}
private void printResults(long totalGood, long totalBad, long nbIteration) {
int difference = (int) (((totalBad - totalGood) * 100) / totalGood);
System.out.println("Total good: " + totalGood + " ms");
System.out.println("Total bad: " + totalBad + " ms");
System.out.println("Total iteration: " + nbIteration);
System.out.println("\nAverage good: " + (totalGood / nbIteration) + " ms");
System.out.println("Average bad: " + (totalBad / nbIteration) + " ms");
System.out.println("\n" + (difference > 0 ? "Gain" : "Loss") + " of " + difference + "%");
}
}
Validation result
Total good: 18461 ms
Total bad: 16778 ms
Total iteration: 16843082980
Associated rule title
[JAVA] Java (multiple If/Else/Elseif)
Associated rule link
https://github.com/green-code-initiative/ecoCode/blob/main/ecocode-rules-specifications/src/main/rules/EC2/java/EC2.asciidoc
Language and platform
Example: Java Itel proc 12th Gen Core i7-1255U 1,7 GHz Docker container : python:3.9-slim memory = 4go swap = 6go cpu=2
Measure justification
Profiling with timestamp 30 secondes minimum execution Calculating the differential between code implementing bad code and code implementing good code practice
Validation
The rule is validated by the measurement. The code used to validate: public class Main {
} Validation result Total good: 18461 ms Total bad: 16778 ms Total iteration: 16843082980
Average good: 0 ms Average bad: 0 ms
Loss of -9%