[
https://issues.apache.org/jira/browse/JCR-1847?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12674559#action_12674559
]
Stefan Guggisberg commented on JCR-1847:
----------------------------------------
> the base exception was used solely inside jackrabbit-core just exactly to
> deal with invalid (malformed) path, unknown names, invalid names, item-state
> exceptions etc. and it was caught again within jackrabbit-core and rethrown
> as RepositoryException (in very exceptional cases a subclass of it) all over
> the place.
FWIW:
the original (i.e. my;) idea behind BaseException was that it allows to clearly
distinguish/detect internal exceptions from api exceptions (i.e.
RepositiryExcception). exceptions thrown by internal code might not always be
correct /useful in the context of an JCR api method. by using a separate base
class for internal exceptions prevents potentially incorrect/meaningless
internal exceptions being thrown by public JCR api methods.
just as a background information, i don't wanna warm up previous discussions...
> Path methods should throw a specific exception instead of RepositoryEx
> ----------------------------------------------------------------------
>
> Key: JCR-1847
> URL: https://issues.apache.org/jira/browse/JCR-1847
> Project: Jackrabbit Content Repository
> Issue Type: Improvement
> Components: jackrabbit-spi
> Reporter: Alexander Klimetschek
> Assignee: angela
> Priority: Minor
>
> The o.a.j.spi.Path interface declares most methods to throw a RepositoryEx,
> whereas a o.a.j.spi.commons.conversion.NameEx (or another specific exception)
> would make more sense, since the RepositoryEx is too generic for what seems
> to be a "malformed names/path" nature for all things that happen in a Path
> implementation.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
