On Mon, Nov 16, 2009 at 11:48 PM, Henri Yandell wrote:
> On Mon, Nov 16, 2009 at 10:16 AM, Paul Benedict wrote:
>> * Is there any benefit to making NumberUtils.min/max accept a variable
>> arguments over a single array?
>
> Seems good.
>From LANG-396:
NumberUtils.min, NumberUtils.max varargs ru
On Tue, Nov 17, 2009 at 10:04 PM, Paul Benedict wrote:
> I would like to make the following changes. Looking for some feedback:
>
> * Append var-args to the method signatures to accept message format
> parameters where a message is provided.
This issue https://issues.apache.org/jira/browse/LANG-5
Agreed. I'd never read the javadoc which suggests collections usage
and had assumed we had grander schemes.
Is ObjectUtils.NULL worth keeping?
Hen
On Tue, Nov 17, 2009 at 3:01 AM, Paul Benedict wrote:
> The use of ObjectUtils.NULL seems esoteric. If you didn't write this
> email, I wouldn't hav
I would like to make the following changes. Looking for some feedback:
* Append var-args to the method signatures to accept message format
parameters where a message is provided.
* A flag to throw NPE rather IAE on notNull and potentially notEmpty checks
Paul
For those interested, Niallp was kind enough to upload a version of the site
http://people.apache.org/~niallp/jexl-rc3/ here .
Comments welcome.
Henrib
--
View this message in context:
http://old.nabble.com/-JEXL--2.0-RC3-stage-tp26400085p26400085.html
Sent from the Commons - Dev mailing list a
Paul Benedict-2 wrote:
>
> ...
> Do you think this is beneficial to your users?
> ...
>
Who are JEXL users today? Is it better for them to be one compilation away
with no code change (based on known cases) or would they rather continue
using an obsolete version?
I'm a biased user and see Jexl
On Tue, Nov 17, 2009 at 6:41 PM, Henrib wrote:
>
> Tried an RC3 after cleansing the xdoc related elements.
> From a clean svn checkout, the 'mvn site' works as expected.
>
> Using the manual procedure, generated
> http://people.apache.org/~henrib/jexl-2.0rc3/site/staging/ which is styled
> correct
I don't think there is a silver-bullet answer, but o.a.c.jexl2
sensible if the architecture radically changes.
Struts 2 and Tapestry 5 have the # in their package structure. The
reason for this is so old and new versions can run together within the
same web app class loader. This makes sense for f
One (non Maven related) question for the JEXL community:
The (soon-to-be published) JEXL-2.0 RCx version comes with a lot of changes.
The upper level API deprecates ExpressionFactory & ScriptFactory - replaced
by JexlEngine -, Expression methods that used to throw exceptions
(ParseException, Exce
Henrib wrote:
>
> Using the manual procedure, generated
> http://people.apache.org/~henrib/jexl-2.0rc3/site/staging/ which is styled
> correctly but has the site links wrong like:
> http://people.apache.org/~henrib/jexl-2.0rc3/Users/people.apache.org/home/henrib/public_html/jexl-2.0rc3/site
>
Tried an RC3 after cleansing the xdoc related elements.
>From a clean svn checkout, the 'mvn site' works as expected.
Using the manual procedure, generated
http://people.apache.org/~henrib/jexl-2.0rc3/site/staging/ which is styled
correctly but has the site links are wrong:
http://people.apache.o
The use of ObjectUtils.NULL seems esoteric. If you didn't write this
email, I wouldn't have known of its support.
Paul
On Tue, Nov 17, 2009 at 3:33 AM, Stephen Colebourne
wrote:
> The new ObjectUtils.firstNonNull() method checked for both null and the
> ObjectUtils.NULL constant object. Since th
>> * Validate throws IAE on every failure -- including null arguments.
>> Because NPE is not thrown, it will make this class less and less
>> palatable in the future since that is the favored approach nowadays --
>> and implemented by the JDK and Google. One solution is to expose a
>> pluggable fac
The new ObjectUtils.firstNonNull() method checked for both null and the
ObjectUtils.NULL constant object. Since that was the only method with
that behaviour I removed the check for the NULL object.
The question is broader than that... do we want all null checks in
ObjectUtils or [lang] to also
On Mon, Nov 16, 2009 at 11:54 PM, Henri Yandell wrote:
> On Mon, Nov 16, 2009 at 11:48 PM, Henri Yandell wrote:
>> kill EscapeUtils/UnescapeUtils and make StringEscapeUtils undeprecated
>> and the intended front door.
>
> I'll do that. Delete EscapeUtils and UnescapeUtils; replace
> StringEscape
the navigation.xml...
This might be one of the causes as explained in
http://maven.apache.org/guides/mini/guide-site.html .
I'll dig into this.
--
View this message in context:
http://old.nabble.com/Help-cutting-JEXL-2.0-RC1-needed-tp26339499p26385994.html
Sent from the Commons - Dev m
sebb-2-2 wrote:
>
> ...
> I don't know if these are due to site build problems or data problems;
> most likely the latter.
> ...
>
I'll correct the data related ones. Albeit I wish it would be a data problem
only; the navigation.xml for instance does not generate the expected html.
The only wa
17 matches
Mail list logo
