7199551: (bf) CharBuffer.append(CharSequence) throws BufferOverflowException for read-only buffer

[Ulf Zibis]

Shouldn't the read-only check also have more priority before checkBounds() ?
If some software would invoke this method with bad bounds, it would take some time, the read-only 
clash becomes discovered.

-Ulf

Am 19.09.2012 19:09, schrieb Alan Bateman:
>
> This one is a behavioral change caused by the 7190219 [1]. If CharBuffer.put(String) or 
> append(String) is invoked on a read-only buffer, and the buffer has fewer elements remaining than 
> the String length, then BufferOverflowException is thrown rather than ReadOnlyBufferException. 
> Technically where there is more than one possible exception then it shouldn't matter which one is 
> thrown but in this case ReadOnlyBufferException is preferable and has been the long standing behavior.
>
> The webrev with the proposed change (trivial) is here:
>
> http://cr.openjdk.java.net/~alanb/7199551/webrev/
>
> In the webrev then the test classes are generated, only BasicChar.java has the new tests, the 
> others can be ignored.
>
> I see that 7190219 has also been pushed to jdk7u so that may need to be looked at too.
>
> Thanks,
> Alan.
>
> [1] http://hg.openjdk.java.net/jdk8/tl/jdk/rev/bf0c6f91bc22

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/nio-dev/attachments/20120920/6264c9f9/attachment.html