#911 False positive in UseStringBufferForStringAppends

[Susan Hoover]

Using PMD 4.2.5, we have encountered a case where PMD seems to be issuing a false positive for UseStringBufferForStringAppends. The case is where we are assigning the result of a method call on a String in a String[] back to the String[]. Something like:

String[] tokens = input.split(COLON_SEPARATOR);
...
tokens = tokens[1].split(COMMA_SEPARATOR);  // UseStringBufferForStringAppends is reported for this line

Full working sample code is in the attached file.

Yes, this is an ugly DD anomaly sort of thing, but regardless, I think it should not be reporting UseStringBufferForStringAppends. In our environment, UseStringBufferForStringAppends is priority 3, but DD anomaly is only priority 5.

[Susan Hoover]

example code for false positive

[Susan Hoover]

Usage of sample code:

$java UseStringBufferForStringAppendsTest "foo: 1, 2, 3, 4, 6"
Left: foo
Right:
1
2
3
4
6
--- end of list

[Raphael Corre]

I have experienced the same issue using the following code :

String[] tab;
tab = java.util.Arrays.copyOf(tab, newLength);