What is the work of pulseIn?

From the docs:

Reads a pulse (either HIGH or LOW) on a pin. For example, if value is HIGH, pulseIn() waits for the pin to go HIGH, starts timing, then waits for the pin to go LOW and stops timing. Returns the length of the pulse in microseconds. _pulsein

So, in this case, pulseIn(echoPin, HIGH) starts counting the number of microseconds until echoPin goes HIGH and stores that in duration.

It starts and ends on that line, it's what is known as a blocking function. It will really sit there until echoPin goes high and tells you how long it took (unless you specify a timeout).

That also means that any delays you have before or after the pulseIn call don't affect it in any way.

The way you get distance from this time is by the following equation:

distance = (duration/2) / 29.1;

You divide by two because it goes out and back so the time would be double that of a one-way travel. The 29.1 is the speed of sound (which is 343.5 m/s => 1 / 0.03435 = 29.1). So note the result is in CM's, not inches. You could probably figure it out by looking at the data sheet of the sensor or just get a lot of samples relating duration to distance (you would measure the distance manually) and get an equation that is very similar.

Indeed it would be a problem if you started to measure the pulse length 1000 microseconds after it starts. However, this is not how the HC-SR04 sensor works:

• the sensor is triggered by the falling edge of TRIG, at digitalWrite(trigPin, LOW);

• the ECHO pulse starts about 0.3 ms after the trigger

That's why a delay of 1 ms doesn't affect the measurement result. pulseIn(echoPin, HIGH) will actually wait for the ECHO pin go HIGH, then start measuring the pulse length until it goes LOW again. As such, the duration of the TRIG pulse can be reduced to 10 microseconds (minimum TRIG duration for HC-SR04), to make the measurements faster.