Hello.
Please review this simple fix for
6990094 "ObjectInputStream cloneArray doesn't handle short[]"
http://cr.openjdk.java.net/~darcy/6990094.0/
The complete patch is
--- old/src/share/classes/java/io/ObjectInputStream.java 2010-12-03
00:31:24.000000000 -0800
+++ new/src/share/classes/java/io/ObjectInputStream.java 2010-12-03
00:31:24.000000000 -0800
@@ -3498,8 +3498,8 @@
return ((int[]) array).clone();
} else if (array instanceof long[]) {
return ((long[]) array).clone();
- } else if (array instanceof double[]) {
- return ((double[]) array).clone();
+ } else if (array instanceof short[]) {
+ return ((short[]) array).clone();
} else {
throw new AssertionError();
}
You'll notice there is no regression test for this change. One
justification is that the fix is in the "obviously no bugs" category.
[1] There is an if-else instanceof chain over Object arrays and arrays
of each primitive type; there are two checks for double and none for
short so changing the one of the double checks to short is "obviously
correct." Also, I've taken a stab a writing an explicit regression test
for this condition, but the problem only manifests in cases beyond my
direct serialization experience where a class has overridden the
readUnshared method.
-Joe
[1] "There are two ways of constructing a software design. One way is to
make it so simple that there are obviously no deficiencies. And the
other way is to make it so complicated that there are no obvious
deficiencies." C.A.R. Hoare]
