RE: Bean Reloading problem

[Jared Pollister <polliste@securecommerce.net>]

On Sun, 21 Nov 1999, Noel J. Bergman wrote:

> Jared,
> 
> No ... in order to compile the page, the java compiler needs to see the bean
> class, too.  That means that your initArg compiler value needs to tell it
> where the bean classes are located.  It is a good idea to have a directory,
> e.g., /myHost/Servlets/beans which acts as the root directory for your bean
> class packages.

I'm using GnuJSP 0.9.10.  The INSTALL file for that release says that the
"init parameter `repository' is obligatory...".  I'm not sure what
parameter to use to specify only the location of the bean classes.

Anyway, I solved the problem--it was really a stupid mistake on my part.
I didn't make the bean class, in the .java file, a "public" class.  Once I
did that, it worked.

However, I'm still wondering, why do I have to "import" my bean in the JSP
file? The examles at

http://archive.javareport.com/9903/html/from_pages/ftp_feature.shtml

...don't use that directive, and I've seen other examples that don't.
Do you know if there is anyway to use JSP and beans without "importing"?



jared

> 
> As for the repository value in your initArgs, are you using GNUJSP 1.0?  Or
> are you using pre-1.0 code?  There isn't a repository value in the current
> package.  There is a scratchdir value, which is the page cache.
> 
> 	--- Noel
> 
>