[PATCH]

[Hyrum Wright <hy...@hyrumwright.org>]

In going through the codebase, I noticed a few areas among the
guessers that seemed a little verbose.  The attached patch cleans them
up a bit.

Also, I'm a bit curious why we still require a source version of Java 1.4.

Best,
-Hyrum

Re: [PATCH]

[Stefan Bodewig <bo...@apache.org>]

On 2010-08-05, Jochen Wiedmann wrote:

> On Wed, Aug 4, 2010 at 10:40 PM, Stefan Bodewig <bo...@apache.org> wrote:

>> Even though Ant still lives in Java 1.4 land, there wouldn't be any
>> problem with an Antlib that required Java 5 - Maven 2.2 requires Java5
>> by now.  I don't really think we are sticking with Java 1.4 for any real
>> reason other than "nobody has started the migration".

> RAT is an important tool for Ant based builds.

Sure, but if the Antlib requires Java5 that doesn't mean you have to do
the rest of the build with it.

> As Maven now depends on Java 1.5, I'd prefer to wait for Ant.

Ant 1.8.0 released early this year was the first release to require Java
1.4 - I don't expect Ant to move to Java5 before mid-2011.

Stefan

Re: [PATCH]

[Jochen Wiedmann <jo...@gmail.com>]

On Wed, Aug 4, 2010 at 10:40 PM, Stefan Bodewig <bo...@apache.org> wrote:

> Even though Ant still lives in Java 1.4 land, there wouldn't be any
> problem with an Antlib that required Java 5 - Maven 2.2 requires Java5
> by now.  I don't really think we are sticking with Java 1.4 for any real
> reason other than "nobody has started the migration".

RAT is an important tool for Ant based builds. As Maven now depends on
Java 1.5, I'd prefer to wait for Ant. OTOH, I'd also have no problems
with Retroweaver, or something similar.

Jochen


-- 
Germanys national anthem is the most boring in the world - how telling!

Re: [PATCH]

["Hyrum K. Wright" <hy...@mail.utexas.edu>]

On Thu, Aug 5, 2010 at 9:16 AM, Stefan Bodewig <bo...@apache.org> wrote:
> On 2010-08-04, Hyrum K. Wright wrote:
>>
>> (This coming from the guy is is also subverting RAT by writing a
>> competitor. :)
>
> Sorry I haven't found time to look into Mouse, yet. And it looks as if
> Gump and a Commons Compress release will be eating up my OSS time for
> the next week or so.

No worries, I'm only gradually fleshing out functionality.  I'll post
to this list when Mouse is "done" (but folks are of course willing to
play around with it before then: it's currently being housed in Apache
Labs).

-Hyrum

Re: [PATCH]

[Stefan Bodewig <bo...@apache.org>]

On 2010-08-04, Hyrum K. Wright wrote:

> Well, we can at least change the source and target Java versions,
> which removes the 1.4 restriction, and then just change stuff as it
> comes along.

This is true.

> (This coming from the guy is is also subverting RAT by writing a
> competitor. :)

Sorry I haven't found time to look into Mouse, yet. And it looks as if
Gump and a Commons Compress release will be eating up my OSS time for
the next week or so.

Stefan

Re: [PATCH]

["Hyrum K. Wright" <hy...@mail.utexas.edu>]

On Wed, Aug 4, 2010 at 3:40 PM, Stefan Bodewig <bo...@apache.org> wrote:
> On 2010-08-04, Hyrum K. Wright wrote:
>
>> Thanks, and sorry about the rather abrupt subject line (I forgot to
>> fill in the rest before sending the mail!)
>
> No problem.
>
>> On Wed, Aug 4, 2010 at 3:04 PM, Stefan Bodewig <bo...@apache.org> wrote:
>>> On 2010-08-04, Hyrum Wright wrote:
>
>>>> Also, I'm a bit curious why we still require a source version of Java
>>>> 1.4.
>
>>> Not sure, my guess is that we didn't have a compelling reason to use
>>> Java5 features so far.  There is a JIRA issue open that want to see us
>>> move to Java5.
>
>> I'm interested mainly to clean up the various loop iterations.  I
>> don't know how much the use of generics would benefit us.
>
> Agreed.  There may be places where we could benefit from enums as well.
> But even then most changes would be cosmetic and so far nobody wanted to
> invest time into it.  You're welcome 8-)

Well, we can at least change the source and target Java versions,
which removes the 1.4 restriction, and then just change stuff as it
comes along.  There isn't any requirement (that I know of) that a 1.5
conversion happen en masse.  I'd be happy to help as a diversion from
$REAL_WORK.

(This coming from the guy is is also subverting RAT by writing a competitor. :)

> Even though Ant still lives in Java 1.4 land, there wouldn't be any
> problem with an Antlib that required Java 5 - Maven 2.2 requires Java5
> by now.  I don't really think we are sticking with Java 1.4 for any real
> reason other than "nobody has started the migration".
>
> I'd be fine with Java5 - once the 0.7 version has been released, that
> is.

Agreed.

-Hyrum

Re: [PATCH]

[Stefan Bodewig <bo...@apache.org>]

On 2010-08-04, Hyrum K. Wright wrote:

> Thanks, and sorry about the rather abrupt subject line (I forgot to
> fill in the rest before sending the mail!)

No problem.

> On Wed, Aug 4, 2010 at 3:04 PM, Stefan Bodewig <bo...@apache.org> wrote:
>> On 2010-08-04, Hyrum Wright wrote:

>>> Also, I'm a bit curious why we still require a source version of Java
>>> 1.4.

>> Not sure, my guess is that we didn't have a compelling reason to use
>> Java5 features so far.  There is a JIRA issue open that want to see us
>> move to Java5.

> I'm interested mainly to clean up the various loop iterations.  I
> don't know how much the use of generics would benefit us.

Agreed.  There may be places where we could benefit from enums as well.
But even then most changes would be cosmetic and so far nobody wanted to
invest time into it.  You're welcome 8-)

Even though Ant still lives in Java 1.4 land, there wouldn't be any
problem with an Antlib that required Java 5 - Maven 2.2 requires Java5
by now.  I don't really think we are sticking with Java 1.4 for any real
reason other than "nobody has started the migration".

I'd be fine with Java5 - once the 0.7 version has been released, that
is.

Stefan

Re: [PATCH]

["Hyrum K. Wright" <hy...@mail.utexas.edu>]

On Wed, Aug 4, 2010 at 3:04 PM, Stefan Bodewig <bo...@apache.org> wrote:
> On 2010-08-04, Hyrum Wright wrote:
>
>> In going through the codebase, I noticed a few areas among the
>> guessers that seemed a little verbose.
>
> Agreed (I've never been looked at that code myself so far), maybe the
> code wanted to avid multiple exits from the methods?
>
>> The attached patch cleans them up a bit.
>
> Slightly modified version (added braces for one line ifs and I couldn't
> hold myself from simplifying one more line) went in as svn revision
> 982385.

Thanks, and sorry about the rather abrupt subject line (I forgot to
fill in the rest before sending the mail!)

>> Also, I'm a bit curious why we still require a source version of Java
>> 1.4.
>
> Not sure, my guess is that we didn't have a compelling reason to use
> Java5 features so far.  There is a JIRA issue open that want to see us
> move to Java5.

I'm interested mainly to clean up the various loop iterations.  I
don't know how much the use of generics would benefit us.

-Hyrum

Re: [PATCH]

[Stefan Bodewig <bo...@apache.org>]

On 2010-08-04, Hyrum Wright wrote:

> In going through the codebase, I noticed a few areas among the
> guessers that seemed a little verbose.

Agreed (I've never been looked at that code myself so far), maybe the
code wanted to avid multiple exits from the methods?

> The attached patch cleans them up a bit.

Slightly modified version (added braces for one line ifs and I couldn't
hold myself from simplifying one more line) went in as svn revision
982385.

> Also, I'm a bit curious why we still require a source version of Java
> 1.4.

Not sure, my guess is that we didn't have a compelling reason to use
Java5 features so far.  There is a JIRA issue open that want to see us
move to Java5.

Stefan