[dna-issues] [JBoss JIRA] Commented: (DNA-101) potential user confusion between org.jboss.dna.spi.graph ValueFactory and ValueFormatException interfaces and similar javax.jcr interfaces

Tue Jun 3 13:30:20 EDT 2008

    [ http://jira.jboss.com/jira/browse/DNA-101?page=comments#action_12415437 ] 

Randall Hauch commented on DNA-101:
-----------------------------------

This is good feedback, and we should consider whether it is a problem.  The DNA SPI will not be exposed to users of DNA; they will instead use the JCR API.  The DNA SPI is really for implementers of custom connectors, sequencers, etc., and one of the goals of the SPI is to prevent a dependency on JCR and to provide a more simplistic graph model.

The SPI ValueFactory interface is named the same thing, and we may want to consider changing the name to prevent confusion.  Yes, it operates in a similar manner to JCR's ValueFactory.  But JCR's API in this regard is more difficult to use and it creates an interim Value object that serves no purpose (and that recently has been discussed as causing some memory issues: see http://www.nabble.com/Memory-use-of-item-states-to17275118.html#a17339203).  The SPI ValueFactory interfaces can also be used for conversion of values, which cannot be done with JCR's.

> potential user confusion between org.jboss.dna.spi.graph ValueFactory and ValueFormatException interfaces and similar javax.jcr interfaces
> ------------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: DNA-101
>                 URL: http://jira.jboss.com/jira/browse/DNA-101
>             Project: DNA
>          Issue Type: Feature Request
>          Components: API
>    Affects Versions: 0.2
>            Reporter: Greg Haber
>            Priority: Trivial
>
> The new org.jboss.dna.spi.graph.ValueFactory and package org.jboss.dna.spi.graph.ValueFormatException interfaces may be potentially confusing to users that are using both DNA and the underlying JCR APIs in the same code.  This is because there are interfaces with exactly the same names, and with very similar purposes (the JCR interfaces deal with instances of the javax.jcr.Value class, where the DNA interfaces are for standard Java objects), in the javax.jcr package.  We should think about changing the names of the DNA interfaces to avoid such potential confusion.  This is a very minor issue, I would not object if we decided to leave the interface names unchanged.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira