The following comment has been added to this issue:
Author: Marcus Brito
Created: Wed, 12 Mar 2003 6:49 AM
Body:
Every class used by every xdoclet class should be fully imported. That's a design
decision, and probably won't change in 1.x development cycle (mayhaps 2.x, but I don't
think so).
As some background info, VO generation is one of the many problems why we did this.
Some classes may not be generated when you run xdoclet, like you pointed out: if you
generate interfaces and value objects in the same ant task, the interfaces won't be
available by the time value objects are generated, so xjavadoc couldn't search for
them. Forcing all classes to be imported using their fully qualified name solves this
problems since xjavadoc doesn't need to search for classes anymore.
---------------------------------------------------------------------
View the issue:
http://opensource.atlassian.com/projects/xdoclet/secure/ViewIssue.jspa?key=XDT-372
Here is an overview of the issue:
---------------------------------------------------------------------
Key: XDT-372
Summary: Value object generator ignores destdir for class searches
Type: Bug
Status: Assigned
Priority: Major
Time Spent: Unknown
Estimate: 0 minutes
Project: XDoclet
Component: None
Versions:
1.2 Beta 2
Assignee: xdoclet-devel (Use for new issues)
Reporter: Joël Larose
Created: Tue, 11 Mar 2003 3:47 PM
Updated: Tue, 11 Mar 2003 3:54 PM
Environment: JSDK 1.4
Windows XP Pro (native dos and cygwin)
RedHat 8.0
Description:
There is a problem with the way the value-object generator looks for classes when
trying to make composite (and I suppose aggregate) value-objects. It should try to
find the source files in the destdir that was provided to the ejbdoclet task.
Specifically, it has a problem finding the previously generated interface (Local or
Remote) for the EJB for which it is trying to compose the VO into the VO it is trying
to generate. (Okay, that was convoluted. ) This is a problem even when I haven't
used the packageSubstitution, which means the local and remote interfaces might be in
the same package as the current class. When I try to use packageSubstitution,
importing newpackage.* doesn't work. The only way I have found to solve the problem
is explicitely importing the required interfaces individually.
If, in the build file, I apply <remoteinterface/> and <localinterface/> before
<valueobject/>, then I should not have to explicitely import interfaces
*individually*. The valueobject substask ought to be able to find the necessary
classes in the destdir.
---------------------------------------------------------------------
JIRA INFORMATION:
This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
http://opensource.atlassian.com/projects/xdoclet/Administrators.jspa
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
-------------------------------------------------------
This SF.net email is sponsored by:Crypto Challenge is now open!
Get cracking and register here for some mind boggling fun and
the chance of winning an Apple iPod:
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en
_______________________________________________
xdoclet-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/xdoclet-devel
