You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Arron <ar...@keyboardmonkey.com> on 2002/01/04 14:39:08 UTC
[PATCH] - PropertyUtils - bug 5639
Attached is the diff for the patch on PropertyUtils.java.
The code is tested and works great.
Or if that don't work (heaven forbid), replace lines 419 to 423
(inclusive) with the code below.
Object value = readMethod.invoke(bean, new Object[0]);
if (!value.getClass().isArray()) {
if (!(value instanceof java.util.List)) {
throw new IllegalArgumentException("Property '" + name
+ "' is not indexed");
} else {
return ((java.util.List)value).get(index);
}
}
return (Array.get(value, index));
Re: [PATCH] - PropertyUtils - bug 5639
Posted by robert burrell donkin <ro...@mac.com>.
On Saturday, January 5, 2002, at 04:21 AM, Arron Bates wrote:
<snip>
> The only things which gets me about this, is when you have something to
> go into struts and you're new to the group, you can't commit. The
> committers are busy doing their own thing (which is fine), and they don't
> want to add it because they either think someone else will get it, or
> they don't want to support it. I'd love to be a committer (votes everyone
> (I'm an excellent coder, really I am :)), but the process is (as stated
> on jakarta) that you submit patches and contribute to the group via the
> above process, then at some point, someone will vote on you becoming a
> committer. The process is fine, and processes have a reason... if that's
> the process. How can I ever move on to contribute real code if I never
> get a patch in there simply because it wasn't committed.
it's interesting that the process seems (to me) to work a little
differently in practice. contributing to the debates on the mailing lists
and answering user questions promptly over a period of time is probably
the best way to get noticed. (i often think that this type of work is just
as great a contribution as creating code.) if you have development ideas,
try to discuss them first on the mailing list. this saves time since
people will already know the design ideas behind the patch when you submit
it. the most welcomed patches are those that save committer's time and
energy. so, volunteer to look into bugs found by other users or to create
patches which scratch the current development itches of committers. of
course, this is only idle speculation...
- robert
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PATCH] - PropertyUtils - bug 5639
Posted by robert burrell donkin <ro...@mac.com>.
On Saturday, January 5, 2002, at 04:21 AM, Arron Bates wrote:
>
>> Is it just that no one is watching out for PropertyUtils, or the people
>> who usually watch on it are busy?
>>
> All it takes is one committer for the commons project and 15 seconds of
> their time.
the problem here is that the commons operates a 'chinese walls' policy.
each committer is a committer for the whole jakarta-commons but there are
only a few components for which they actively develop. committing a patch
for another component means (at the very least) talking to the existing
committers and adding yourself to the list of committers responsible for
that component.
FYI committing a patch takes a lot longer than 15 seconds - especially on
an unfamiliar project. committing a patch means that you're taking
personal responsibility. this means being able and willing to defend the
design of that patch.
in general, though, i'd agree that the process by which patches are
reviewed isn't as smooth as it might be. any ideas about how we could
improve it?
- robert
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PATCH] - PropertyUtils - bug 5639
Posted by Arron Bates <ar...@keyboardmonkey.com>.
>Is it just that no one is watching out for PropertyUtils, or the people
>who usually watch on it are busy?
>
All it takes is one committer for the commons project and 15 seconds of
their time.
>>Since this is the third or fourth person in recent memory to submit a
>>patch for List support, will this ability ever be added?
>>
I think that they're all my attempts to get the patch committed. I sent
a mail with the patch. Then remembered the group's bugzilla, so I put it
in there. The process is then to write an email with the patch to the
bug with the subject written a certain way. So I did that. Still no
action. So I reckon I'll just keep mailing it until it's there.
Reason is... it's a worthy patch which makes life much easier for those
people who want to nest beans. A part of the whole reason for the
BeanUtils package.
Map implementations are supported in a different process, why not List?...
The only things which gets me about this, is when you have something to
go into struts and you're new to the group, you can't commit. The
committers are busy doing their own thing (which is fine), and they
don't want to add it because they either think someone else will get it,
or they don't want to support it. I'd love to be a committer (votes
everyone (I'm an excellent coder, really I am :)), but the process is
(as stated on jakarta) that you submit patches and contribute to the
group via the above process, then at some point, someone will vote on
you becoming a committer. The process is fine, and processes have a
reason... if that's the process. How can I ever move on to contribute
real code if I never get a patch in there simply because it wasn't
committed. And in terms of support... it's for the group to support
(including non committers), not any individual committer. It would take
a committer 15 seconds to make this patch.
This patch is the smallest of patches. It's also tested. If this can't
get through, what's the chance on me fixing a larger bug which involves
a bunch of refactoring, or supporting any classes I write and are
added?... The group site is all about joining in, and submitting code...
help us make it better. From where I sit, it's just not the case.
Arron.
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PATCH] - PropertyUtils - bug 5639
Posted by "Craig R. McClanahan" <cr...@apache.org>.
On Fri, 4 Jan 2002 bayard@generationjava.com wrote:
> Date: Fri, 4 Jan 2002 09:50:30 -0500 (EST)
> From: bayard@generationjava.com
> Reply-To: Jakarta Commons Developers List <co...@jakarta.apache.org>
> To: Jakarta Commons Developers List <co...@jakarta.apache.org>
> Subject: Re: [PATCH] - PropertyUtils - bug 5639
>
> Is it just that no one is watching out for PropertyUtils, or the people
> who usually watch on it are busy?
>
The latter.
There is also an important philosophical issue here -- it breaks
conformance with standard JavaBeans definition of the semantics for
indexed properties, which does *not* include List based support. That's
worth some reflective discussion first, and (so far) I remain somewhat
unconvinced that we should deviate here.
> Bay
>
Craig
> On Fri, 4 Jan 2002, Lavandowska wrote:
>
> > Since this is the third or fourth person in recent memory to submit a
> > patch for List support, will this ability ever be added?
> >
> > --- Arron <ar...@keyboardmonkey.com> wrote:
> > > Object value = readMethod.invoke(bean, new Object[0]);
> > > if (!value.getClass().isArray()) {
> > > if (!(value instanceof java.util.List)) {
> > > throw new IllegalArgumentException("Property '" + name
> > > + "' is not indexed");
> > > } else {
> > > return ((java.util.List)value).get(index);
> > > }
> > > }
> > > return (Array.get(value, index));
> >
> >
> > =====
> > Lance Lavandowska
> > Http://www.brainopolis.com/
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Send your FREE holiday greetings online!
> > http://greetings.yahoo.com
> >
> > --
> > To unsubscribe, e-mail: <ma...@jakarta.apache.org>
> > For additional commands, e-mail: <ma...@jakarta.apache.org>
> >
> >
>
>
> --
> To unsubscribe, e-mail: <ma...@jakarta.apache.org>
> For additional commands, e-mail: <ma...@jakarta.apache.org>
>
>
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PATCH] - PropertyUtils - bug 5639
Posted by ba...@generationjava.com.
Is it just that no one is watching out for PropertyUtils, or the people
who usually watch on it are busy?
Bay
On Fri, 4 Jan 2002, Lavandowska wrote:
> Since this is the third or fourth person in recent memory to submit a
> patch for List support, will this ability ever be added?
>
> --- Arron <ar...@keyboardmonkey.com> wrote:
> > Object value = readMethod.invoke(bean, new Object[0]);
> > if (!value.getClass().isArray()) {
> > if (!(value instanceof java.util.List)) {
> > throw new IllegalArgumentException("Property '" + name
> > + "' is not indexed");
> > } else {
> > return ((java.util.List)value).get(index);
> > }
> > }
> > return (Array.get(value, index));
>
>
> =====
> Lance Lavandowska
> Http://www.brainopolis.com/
>
> __________________________________________________
> Do You Yahoo!?
> Send your FREE holiday greetings online!
> http://greetings.yahoo.com
>
> --
> To unsubscribe, e-mail: <ma...@jakarta.apache.org>
> For additional commands, e-mail: <ma...@jakarta.apache.org>
>
>
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PATCH] - PropertyUtils - bug 5639
Posted by Lavandowska <fl...@yahoo.com>.
Since this is the third or fourth person in recent memory to submit a
patch for List support, will this ability ever be added?
--- Arron <ar...@keyboardmonkey.com> wrote:
> Object value = readMethod.invoke(bean, new Object[0]);
> if (!value.getClass().isArray()) {
> if (!(value instanceof java.util.List)) {
> throw new IllegalArgumentException("Property '" + name
> + "' is not indexed");
> } else {
> return ((java.util.List)value).get(index);
> }
> }
> return (Array.get(value, index));
=====
Lance Lavandowska
Http://www.brainopolis.com/
__________________________________________________
Do You Yahoo!?
Send your FREE holiday greetings online!
http://greetings.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>