You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@jackrabbit.apache.org by Angela Schreiber <an...@day.com> on 2007/02/13 10:54:28 UTC

Jcr2Spi refactoring -> Changes to SPI interfaces

hi

with rev. 506927 a couple of modifications to the
jcr2spi subproject were commited, that led to minor
changes in the SPI interfaces.

the reason for the refactoring was, that we wanted to
separate the hierarchical information from the item states
and allow to access items without having to load the ancestors.

changes to the SPI interfaces:

1) o.a.j.spi.ItemInfo
    new method 'getPath' that allows to built the missing
    hierarchy entries in case a item identified by a
    uniqueID is loaded.

2) o.a.j.spi.LockInfo
    new method 'getNodeId', that allows to retrieve the ItemId
    if the lockholding node, that is exposed with javax.jcr.lock.Lock

3) the SPI implementation must now be able to deal with
    ItemIds, that equivocally omit the uniqueID part and
    instead only contain a path. this results from the fact,
    that the jcr2spi may initially ignore the uniqueID defined
    with a parent node, as long as neither the latter nor a child
    entry has been loaded.

The changes have been tested with the test-cases present
in jackrabbit-jcr-tests and some implementation specific
tests (see contrib/spi/client/src/test/j/o/a/j/jcr2spi).
Known issues and todos are listed with the TODO.txt present
in jcr2spi project.

regards
angela