Hi,
This might be slightly off-topic as its more related to the development
process and less to struts itself.
I've been lurking on the (heavily trafficked) mailing list for a while
now and haven't seen this subject mentioned in a great amount of detail. I
did search the archives as well before posting.
My question relates to doing Test Driven Development (TDD) - red light,
green light, re-factor, so on...
Here's what I had in (my murky) mind:
1. Start off by creating a test using StrutsTestCase (MockTestCase
version)
2. Use ant to run the test and watch it fail (red light)
3. Create the form and action classes including javadoc tags for xdoclet
to pick up
4. Run ant webdoclet task which will generate the relevant
struts-config.xml, web.xml and validat*.xml files
5. Run ant test again to watch it pass (green light)
6. Refactor and repeat...
Is this on target?
So, what about the ant target dependencies? Seems to me that the "test"
target should be the default and the dependencies should be something
like:
"test-compile" depends="compile,webdoclet,test-init"
"test" depends="test-compile"
so that I can keep running ant test (or ant since test is the default) no
matter what I'm doing and the "right" thing will be done.
All of the editing is assumed to be done with vim. If I was to throw
IntelliJ IDEA into the mix then does IDEA/Struts Console integration
really buy me anything over xdoclet? In other words, is Struts Console
addressing the same problem space as xdoclet (maintainability of config
files)?
Be gentle in your replies as this is my first foray into TDD :-) How about
a tutorial on TDD development with struts? Rick? Really enjoyed your
tutorial, just to let you know.
--
Haroon
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
