On 10/6/10 8:55 PM, Jan Ohlsen wrote:
In the constructor
public SimpleIoProcessorPool(Class<? extends IoProcessor<T>>
processorType, Executor executor, int size)
1) There's the following code:
// We create at least one processor
try {
try {
processorConstructor =
processorType.getConstructor(ExecutorService.class);
pool[0] =
processorConstructor.newInstance(this.executor);
} catch (NoSuchMethodException e) {
// To the next step...
}
try {
processorConstructor =
processorType.getConstructor(Executor.class);
pool[0] =
processorConstructor.newInstance(this.executor);
} catch (NoSuchMethodException e) {
// To the next step...
}
try {
processorConstructor = processorType.getConstructor();
usesExecutorArg = false;
pool[0] = processorConstructor.newInstance();
} catch (NoSuchMethodException e) {
// To the next step...
}
Should only attempt the second& third processorConstructor assignments
if the first one fails etc.?
This piece of code sounds a bit weird to say the least...
We should probably incorporate the code in the following attempts into
the catch() part, like :
try {
try {
processorConstructor =
processorType.getConstructor(ExecutorService.class);
pool[0] = processorConstructor.newInstance(this.executor);
} catch (NoSuchMethodException e) {
// To the next step...
try {
processorConstructor = processorType.getConstructor(Executor.class);
pool[0] = processorConstructor.newInstance(this.executor);
} catch (NoSuchMethodException e) {
// To the next step...
try {
processorConstructor = processorType.getConstructor();
usesExecutorArg = false;
pool[0] = processorConstructor.newInstance();
} catch (NoSuchMethodException e) {
// To the next step...
}
}
}
}
2) The JavaDoc is missing the size parameter
Fixed
Thanks for the report !
PS : It's better to fill a JIRA when you discover such errors, as we probably
will process it asynchronously. Mail are likely to get buried forever :)
PS2 : It seems that the modified code pass the tests successfully. great !
--
Regards,
Cordialement,
Emmanuel Lécharny
www.iktek.com
