harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mark Hindess" <mark.hind...@googlemail.com>
Subject Re: [jira] Commented: (HARMONY-141) Constructors of java.nio.charset.CharsetEncoder do not validate arguments
Date Fri, 03 Mar 2006 06:57:39 GMT
Obviously we should encourage submission of patches but the committers
should remember that tthere are non-committers around who are happy to
help get worthwhile contributions in shape!

Perhaps just adding a comment to the JIRI suggesting the submitter
provide a patch or ask for help on the list?

I've been spending a little time fixing up contributions that I think
are worthwhile - mostly patches that have been broken by the recent
shared native code moves that Tim committed.  I'm willing to do this
because I broke them ;-) but mostly because I know it helps reduce the
load on committers and because having your patch broken/delayed can be
discouraging for new contributors.

For me this is a good way to become more familiar with the code base. 
I'm sure that will be true for others joining the project.  I don't
think there is any harm in having plenty of obvious tasks for
potential new contributors.

Our committers shouldn't be slow to ask for help either.  For
instance, a number of the recent large contributions aren't going to
be trivial to integrate in to subversion, but I'm sure there are many
people - at least one anyway! - looking forward to seeing them in
subversion who would be willing to help create patches [0] from
contributions if it helped move them forward.

Regards,
 Mark.

[0] More likely a tar with the new files and a patch to existing
build.xml files.

On 02/03/06, Geir Magnusson Jr <geir@pobox.com> wrote:
> maybe we should reject things not in patch format, as it puts
> unnecessary load on the committers.
>
> Feel free to drop the code into comment or email so that we can discuss,
> but patch is much easier for everyone involved....
>
>
> Tim Ellison wrote:
> > p.s.  I accept test cases in patch format too ;-)
> >
> > Regards,
> > Tim
> >
> > Richard Liang (JIRA) wrote:
> >>     [ http://issues.apache.org/jira/browse/HARMONY-141?page=comments#action_12368097
]
> >>
> >> Richard Liang commented on HARMONY-141:
> >> ---------------------------------------
> >>
> >> Here are the test cases which will pass on RI but fail on Harmony.
> >>
> >>     public void testConstructorIlegalAverageBytesPerChar() {
> >>         try {
> >>             Charset cs = Charset.forName("UTF-8"); //$NON-NLS-1$
> >>             CharsetEncoder encoder = new MockCharsetEncoderForHarmony141(cs,
1.1f, 1);
> >>             fail("Should throw IllegalArgumentException."); //$NON-NLS-1$
> >>         } catch (IllegalArgumentException e) {
> >>             // expected
> >>         }
> >>     }
> >>
> >>     public void testConstructorIlegalAverageBytesPerChar2() {
> >>         try {
> >>             Charset cs = Charset.forName("ISO8859-1"); //$NON-NLS-1$
> >>             CharsetEncoder encoder = new MockCharsetEncoderForHarmony141(cs,
1.1f, 1,
> >>                     new byte[] { 0x1a});
> >>             fail("Should throw IllegalArgumentException."); //$NON-NLS-1$
> >>         } catch (IllegalArgumentException e) {
> >>             // expected
> >>         }
> >>     }
> >>
> >>     public static class MockCharsetEncoderForHarmony141 extends CharsetEncoder
{
> >>
> >>         protected MockCharsetEncoderForHarmony141(Charset cs, float averageBytesPerChar,
> >>                 float maxBytesPerChar) {
> >>             super(cs, averageBytesPerChar, maxBytesPerChar);
> >>         }
> >>
> >>         public MockCharsetEncoderForHarmony141(Charset cs, float averageBytesPerChar,
> >>                 float maxBytesPerChar, byte[] replacement) {
> >>             super(cs, averageBytesPerChar, maxBytesPerChar, replacement);
> >>         }
> >>
> >>         protected CoderResult encodeLoop(CharBuffer in, ByteBuffer out) {
> >>             return null;
> >>         }
> >>
> >>     }
> >>
> >>
> >>> Constructors of java.nio.charset.CharsetEncoder do not validate arguments
> >>> -------------------------------------------------------------------------
> >>>
> >>>          Key: HARMONY-141
> >>>          URL: http://issues.apache.org/jira/browse/HARMONY-141
> >>>      Project: Harmony
> >>>         Type: Bug
> >>>     Reporter: Richard Liang
> >>> Constructors of java.nio.charset.CharsetEncoder should throw IllegalArgumentException
when averageBytesPerChar exceeds maxBytesPerChar.
> >
>


--
Mark Hindess <mark.hindess@googlemail.com>
IBM Java Technology Centre, UK.

Mime
View raw message