jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefan Guggisberg <stefan.guggisb...@gmail.com>
Subject Re: Jackrabbit dependencies
Date Thu, 03 Feb 2011 10:00:08 GMT
On Wed, Feb 2, 2011 at 7:20 PM, Jukka Zitting <jzitting@adobe.com> wrote:
> Hi,
> On 02/02/2011 02:33 PM, Stefan Guggisberg wrote:
>> +1 for removing it.
> Done, see JCR-2875.


other candidates for exclusion:

- bouncycastle
- rome

IIUC bouncycastle is used for parsing encrypted pdf's,
and rome is used for parsing RSS feeds.

if those are optional dependencies of tika,
i'd like to remove them from jackrabbit core.
if they aren't optional, we should probably
consider alternatives...


> To answer Angela's question, the netcdf dependency came in as a transitive
> dependency from Tika 0.8. I didn't think the extra size was too big a deal
> and wanted to avoid extra excludes. But as said I don't feel too strongly
> about this.
> And to Thomas' point about no parser dependencies: I wouldn't go there.
> Perhaps we could/should do a separate jackrabbit-lite package if there are
> people who really need that, but personally I think that full text indexing
> is such an important part of JCR functionality that we should support it
> right out of the box.
> What we could do is move the tika-parsers dependency from jackrabbit-core to
> the jackrabbit-webapp and jackrabbit-jca modules. Then someone who has a
> direct low-level dependency to jackrabbit-core wouldn't automatically get
> any of the parser libraries, but we'd still have them included in the
> deployable release packages.
> --
> Jukka Zitting

View raw message