Neil Ford test naming

[Howard Lewis Ship <hl...@gmail.com>]

Something I'm trying out in the new code base is to use Neil Ford's
naming convention for test methods.  Names are
words_separated_with_underscores ... the intent is reports generated
about the tests are more readable than wordsInCamelCase.  I think he's
right and I'm liking the result.

http://memeagora.blogspot.com/2006/04/eating-sacred-hamburger.html

In case you haven't noticed, I've been writing a good deal of Tapestry
5 code. Right now I'm in the middle of writing a new IOC container for
Tapestry 5.  Some of the documentation is already on line:

http://tapestry.apache.org/tapestry5/ioc/index.html

It's coming out great. The container will be more powerful and
expressive than HiveMind (or Spring) but will not use any XML.
Instead, it uses a mix of naming conventions and annotations. It's
also very fast.  Because it defers to simple user-written code for a
lot of things, it is not only clearer what's going on, and more
efficient, but also will require far fewer services to accomplish the
same behavior. I'm really happy with it, and expect to back port most
of the features to HiveMind 2.0.

For the moment, there's a dependency on HiveMind for a few utility
classes but I intend to fork that code and specialize it for Tapestry
IOC. Praise open source.

-- 
Howard M. Lewis Ship
Independent J2EE / Open-Source Java Consultant
Creator and PMC Chair, Apache Tapestry
Creator, Jakarta HiveMind

Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tapestry.apache.org
For additional commands, e-mail: dev-help@tapestry.apache.org

Re: Neil Ford test naming

[Howard Lewis Ship <hl...@gmail.com>]

> If I were (hypothetically never ) running things at Sun I'd be knocking on
> Howard's door until a restraining order was created or he decided to help me
> write a good spec. ;)
>

Spec?  That's the thorn in Sun's side. Create useful, productive
tools, not specs. Sun stops one step too short. Create an open source,
high quality tool with plenty of extension points, and let the vendors
differentiate on their integrations and plugins for that framework.
Creating specifications so that vendors can create isolated
implementations lacking in integration points (unless you are the
vendor, with access to the source) is the problem.

We can learn from Rails on this: focus on the tool and developer
productivity (throughout the stack).

Java and the open source community in general, and Tapestry in
particular, has left how to be productive, how to integrate all the
tools and frameworks into something useful, as an exercise to the
reader. That's why there's so much desperate interest in Rails, or the
comfortable suffocating hand of .Net.  People want to be told what to
do.  I think Tapestry 5 will do that ... provide a specific way to do
everything, yet allow the curtains to be pulled back to handle the
exceptional cases.

The IOC container is coming along great.  I'm about to start on
configuration data, which will lead up to contributed lifecycle
models.  From there, I'll be back to the true Tapestry code pretty
quickly.

-- 
Howard M. Lewis Ship
Independent J2EE / Open-Source Java Consultant
Creator and PMC Chair, Apache Tapestry
Creator, Jakarta HiveMind

Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tapestry.apache.org
For additional commands, e-mail: dev-help@tapestry.apache.org

Re: Neil Ford test naming

[Jesse Kuhnert <jk...@gmail.com>]

Sounds good to me. I'll start using the more descriptive form of unit test
method names.

I'm very excited about the potential of T5. Seems to address all manner of
internal/external wants/needs for tapestry in general.

It's still amazing to me what a stark contrast there is in java web
framework development leads. There are a lot of well intentioned/capable
people doing good things out there but none quite so obviously
seasoned/knowledgeable in contrast to what we see being done here. (I'm not
speaking of myself )

If I were (hypothetically never ) running things at Sun I'd be knocking on
Howard's door until a restraining order was created or he decided to help me
write a good spec. ;)

On 7/18/06, Howard Lewis Ship <hl...@gmail.com> wrote:
>
> Something I'm trying out in the new code base is to use Neil Ford's
> naming convention for test methods.  Names are
> words_separated_with_underscores ... the intent is reports generated
> about the tests are more readable than wordsInCamelCase.  I think he's
> right and I'm liking the result.
>
> http://memeagora.blogspot.com/2006/04/eating-sacred-hamburger.html
>
> In case you haven't noticed, I've been writing a good deal of Tapestry
> 5 code. Right now I'm in the middle of writing a new IOC container for
> Tapestry 5.  Some of the documentation is already on line:
>
> http://tapestry.apache.org/tapestry5/ioc/index.html
>
> It's coming out great. The container will be more powerful and
> expressive than HiveMind (or Spring) but will not use any XML.
> Instead, it uses a mix of naming conventions and annotations. It's
> also very fast.  Because it defers to simple user-written code for a
> lot of things, it is not only clearer what's going on, and more
> efficient, but also will require far fewer services to accomplish the
> same behavior. I'm really happy with it, and expect to back port most
> of the features to HiveMind 2.0.
>
> For the moment, there's a dependency on HiveMind for a few utility
> classes but I intend to fork that code and specialize it for Tapestry
> IOC. Praise open source.
>
> --
> Howard M. Lewis Ship
> Independent J2EE / Open-Source Java Consultant
> Creator and PMC Chair, Apache Tapestry
> Creator, Jakarta HiveMind
>
> Professional Tapestry training, mentoring, support
> and project work.  http://howardlewisship.com
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@tapestry.apache.org
> For additional commands, e-mail: dev-help@tapestry.apache.org
>
>


-- 
Jesse Kuhnert
Tacos/Tapestry, team member/developer

Open source based consulting work centered around
dojo/tapestry/tacos/hivemind.