Oh sorry, I missed that you had tightened the imports. I took a quick look at StringJoiner. It looks to me like this won't be an optimization, because StringJoiner uses StringBuilder internally, and will actually perform more total operations. It looks to me like StringJoiner itself could be optimized by storing the strings in an array and generating the final String perfectly at the end when all the inputs are known, using the private String constructor that takes a char[].
On Fri, Jul 18, 2014 at 3:57 PM, Ivan Gerasimov <[email protected]> wrote: > In the test I only replaced > > import java.lang.reflect.*; > > with > > import java.lang.reflect.Modifier; > > and added the new bugid to the @bug tag. > > Sincerely yours, > Ivan > > > On 19.07.2014 2:53, Martin Buchholz wrote: > > It looks like the test has a new import, but no actual new code that uses > it. > > > On Fri, Jul 18, 2014 at 3:39 PM, Ivan Gerasimov <[email protected] > > wrote: > >> Hello! >> >> StringBuilder can be replaced with StringJoiner, which will make the code >> shorter and, probably, more efficient. >> The existing regtest (java/lang/reflect/Modifier/toStringTest.java) can >> be used to make sure nothing is broken. >> >> Would you please help review the fix? >> >> BUGURL: https://bugs.openjdk.java.net/browse/JDK-8051382 >> WEBREV: http://cr.openjdk.java.net/~igerasim/8051382/0/webrev/ >> >> Sincerely yours, >> Ivan >> > > >
