unusual string manipulation

[pjfanning]

I spotted some issues when reviewing the v2.2.5 RC4.

This is an example in your code

ozhera/ozhera-prometheus-agent/ozhera-prometheus-agent-common/src/main/java/org/apache/ozhera/prometheus/agent/util/Http.java

Lines 55 to 57 in 66402b2

	while ((str = br.readLine()) != null) {
	finalStr = new String(str.getBytes(), "UTF-8");
	}

I don't understand what this achieves.

• String .getBytes() uses the JVM default which depends on the machine/OS - UTF-8 on many machines but Windows OS machines can have region specific defaults.
• why not just use the string from the readLine instead of converting it to bytes in a way that can differ based on the JVM setup and then randomly taking these bytes that can be any char encoding depending on the JVM and then pretending that they are UTF-8 - when this is not guaranteed. You already have the string - why mess with it?
• I counted at least 4 other places with the same while loop. Wouldn't it be better to have a shared method for this?