|
Quick Lists
|
|
Bug ID:
|
6593716
|
|
Votes
|
0
|
|
Synopsis
|
Serialized Form javadoc for LinkedBlockingDeque contradicts implementation
|
|
Category
|
java:classes_util_concurrent
|
|
Reported Against
|
b105
|
|
Release Fixed
|
|
|
State
|
3-Accepted,
bug
|
|
Priority:
|
4-Low
|
|
Related Bugs
|
|
|
Submit Date
|
16-AUG-2007
|
|
Description
|
Problem description:
====================
The Serialized Form javadoc for the java.util.concurrent.LinkedBlockingDeque class at:
http://java.sun.com/javase/6/docs/api/serialized-form.html#java.util.concurrent.LinkedBlockingDeque
states as follows:
---Excerpt-from-spec---
private void writeObject(ObjectOutputStream s) throws IOException
Save the state of this deque to a stream (that is, serialize it).
Serial Data:
The capacity (int), followed by elements (each an Object) in the proper order, followed by a null
---End-of-excerpt-from-spec---
Actually other fields of the class is written using default serialization too, so more reasonable description here should looks like:
...
Serial Data:
Default fields, followed by elements (each an Object) in the proper order, followed by a null
...
Posted Date : 2007-08-16 14:37:10.0
|
|
Work Around
|
N/A
|
|
Evaluation
|
How would a reader of the specification know what the "default fields"
are, without reading the source code?
I do agree that our Serialized Form specifications are
generally incomplete and imprecise,
so much so that implementers of compatible implementations have to do
some reverse-engineering in practice.
Posted Date : 2007-08-16 18:00:06.0
|
|
Comments
|
PLEASE NOTE: JDK6 is formerly known as Project Mustang
|
|
|
 |