Github user trkurc commented on the pull request:
https://github.com/apache/nifi/pull/202#issuecomment-179579922
@taftster - code readability is subjective, I prefer actually prefer what
is in the PR. The String.format makes it an order of magnitude slower based on
my initial tests. I see no appreciable difference between StringBuffer and
StringBuilder
The results of the below code:
``
1 ===== 304 elapsed
2 ===== 1717 elapsed
3 ===== 248 elapsed
```
The StringBuffer is faster, but it is likely due to some of the JVM string
magic.
```java
int iter = 1<<20;
AtomicLong ai = new AtomicLong(0);
long now = System.currentTimeMillis();
for(int i=0; i < iter; i++) {
final String s=Long.toHexString(ai.getAndIncrement());
String x2 = new
StringBuilder("00000000-0000-0000-0000000000000000".substring(0,(35-s.length()))+s).insert(23,
'-').toString();
}
System.out.printf("1 ===== %d elapsed\n",
(System.currentTimeMillis() - now ));
now = System.currentTimeMillis();
for(int i=0; i < iter; i++) {
String x2 = new StringBuilder()
.append("00000000-0000-0000-")
.append(String.format("%016x", ai.getAndIncrement()))
.insert(23, '-')
.toString();
}
System.out.printf("2 ===== %d elapsed\n",
(System.currentTimeMillis() - now ));
now = System.currentTimeMillis();
for(int i=0; i < iter; i++) {
final String s=Long.toHexString(ai.getAndIncrement());
String x2 = new
StringBuffer("00000000-0000-0000-0000000000000000".substring(0,(35-s.length()))+s).insert(23,
'-').toString();
}
System.out.printf("3 ===== %d elapsed\n",
(System.currentTimeMillis() - now ));
```
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
