jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Julian Reschke <julian.resc...@gmx.de>
Subject Re: WebDav Jackrabbit retrieve remote file content by URL
Date Tue, 09 Apr 2013 08:05:19 GMT
On 2013-04-04 10:41, Bertrand Delacretaz wrote:
> Hi,
>
> On Wed, Apr 3, 2013 at 5:04 PM, Antonio Gioia <antonio.gioia82@gmail.com> wrote:
>> ...I'd like to add some "link" to files stored in a remote storage, I don't
>> wont import the files present in this remote storage in my jackrabbit
>> repository. So, I was thinking still to use a subtype of nt:resource node
>> adding a property to store the URL (a simple string) for the remote files,
>> but in this way what I can retrieve the content of the remote files?...
>
> Do you mean retrieving the content via WebDAV, or via HTTP?
>
> If it's WebDAV I think you'd have to proxy the whole file retrieval
> protocol (PROPFIND, PROPGET, GET etc.) which is certainly possible by
> customizing the Jackrabbit WebDAV server, or reimplementing the parts
> that you need on something like http://milton.io/ and using that
> instead of Jackrabbit's WebDAV server, if it's easier to customize.
>
> Retrieving the files via HTTP would probably be easier, you'd just
> need to implement a servlet that maps incoming URLs to JCR nodes,
> checking if the addressed node is a local or remote file, and
> serving/proxying the content accordingly. In that case you might not
> even need a specific node type, you could use nt:unstructured with a
> remoteFileURL property, at least initially.
> ...

But note the potential security hole - if the user can author the link, 
he/she might be able to access other HTTP servers within the LAN the 
server is running on. (Or think "file:").

Best regards, Julian

Mime
View raw message