[jsr-314-open] CLOSED Re: Need guidance: invalid assumptions in design of resource versioning feature

[Ed Burns]

>>>>> On Fri, 11 Sep 2009 11:18:36 -0700, Ryan Lubke <Ryan.Lubke at Sun.COM> said:

On 9/9/09 8:35 AM, Ed Burns wrote:

EB> I'm not fond of reducing the flexibility that we already have.  Ryan has
EB> implemented something that he's satisfied with.  Ryan, can you please
EB> share the details so we can further discuss the options for ammending
EB> the specification?

RL> Here's where I am currently with this issue.

RL> Implementation wise, I changed the logic to support classpath
RL> resources only within the context of the web application.  This
RL> requires the use of ServletContext.getResourcePaths() in order to
RL> obtain all of the included JAR files as well as any potential
RL> content under /WEB-INF/classes/META-INF/resources and storing the
RL> paths that were discovered in some structure for lookup purposes.

RL> This worked fine in GFv2/3 and JBoss 5.1.0.  However, we've run into
RL> an issue where this falls apart in embedded situations (mvn
RL> jetty:run as an example).  We can't be guaranteed that /WEB-INF/lib
RL> or /WEB-INF/classes will exist in embedded scenarios.  Sure, there
RL> are workarounds, but it's going to surprise developers in this
RL> situation and I don't think that's what we want.

RL> I've tried some other potential fixes and they tended to work in some 
RL> cases and then failed in others.

RL> At this point, I'm thinking we should punt on trying to support 
RL> versioned classpath resources and address
RL> the issue proper in 2.1.

Ok, I have added issue C033 to the changelog. [1].

Ed

[1] http://wiki.java.net/bin/view/Projects/Jsf2MR1ChangeLog


-- 
| ed.burns at sun.com  | office: 408 884 9519 OR x31640
| homepage:         | http://ridingthecrest.com/