Hi

I just noticed this post. I have been giving the same issue some
thought.

I do actually not like the file protocol for this use.

First, there are the security risks. You could, as a remote operator add
a file that is local to fedora as a datastream, even if you could not
read this file. All because file:// is resolved on the Fedora server.
To add datastreams require some admin rights, but if you have some GUI
where users can create new objects the users have these rights.
I do not know if this is a big issue, but I do not like to worry about 
the implications of it.

Second, fedora does NOT require that you use the client to upload the
file beforehand. It requires a url to the file. Why not just put a
webserver up to present the file directory. Then you will have urls of
the form http://localhost:8080/Tetabytedisc/dir/dir/file for each of the
files.

Regards




On Fri, 2009-05-29 at 09:57 +0200, henk van den berg wrote:
> Hi,
> 
> Thanks for the fast reply and sure I voted for the issue.
> 
> A few moments after I posted the original message to the mailing list I
> noticed that the DatastreamManagedContent.TEMP_SCHEME is really just mend
> for Fedora internal use. Oeps, my original file had passed away after being
> ingested by Fedora.
> 
> Still I need to get this batch of data files into Fedora. Did some
> investigation on speed of various FedoraClient and FedoraAPIM methods. After
> a thousand iterations the average time it took to upload a 10Mb file was
> about 500 msec. Than adding this file to a digital object with FedoraAPIM
> addDatastream method took on average nearly a second. Mind that in this
> sequence of operations the file is moved two times (if I'm not missing out
> anything): from the original location to the temporary 'uploaded' and from
> there to its final destination as managed content in the repository. Looks
> like when we cut out the upload-operation we might gain an improvement of
> 33% on speed of file ingestion.
> 
> If urls with the file:///-protocol were allowed in the addDatastream method
> than ingesting data from mounted Tera-byte discs would become significantly
> faster. So sure, 'Allow the retrieval of content via the file URI scheme' (
> https://fedora-commons.org/jira/browse/FCREPO-453) has my full support.
> 
> greetings,
> henk van den berg,
> software developer at DANS - Data Archives and Networked Services
> http://www.dans.knaw.nl
> 
> 
> On 28-05-09 17:01, "Benjamin Armintor" <[email protected]> wrote:
> 
> > There are some open issues related to your problem:
> > https://fedora-commons.org/jira/browse/FCREPO-453
> > https://fedora-commons.org/jira/browse/FCREPO-485
> > 
> > If you have an account on the issue tracker, please vote for them!
> > 
> > With regard to your questions:
> >> Is there a legal way to let Fedora read managed content from disc? (other
> >> than the one described above)
> > 
> > I don't believe there's an out-of-the-box way to specify file:// URLs
> > (at least, there wasn't last time I looked at the release code).  You
> > could probably sort out a way to copy your files into the temp
> > directory and provide the uploaded:// urls yourself, but that seems
> > like a frail solution.
> > 
> >> Is this ValidationUtility class out of sync with the rest of the code body?
> >> Does it accidentally prevent something that elsewhere is legal routine?
> >> And if so, when is it going to be repaired?
> > 
> > The temp and uploaded protocols are supposed to be internal, and the
> > ValidationUtility is supposed to deal with externally supplied
> > content.  In that sense, things aren't broken, but the aforementioned
> > issues reflect a desire for this to change.
> > 
> >> Are there any reasons why not to use the hacked ValidationUtility class
> >> (besides from being a hack); will using it corrupt something somewhere 
> >> else?
> > 
> > Once your objects/updates are ingested, Fedora will be tracking the
> > location of the managed content internally, so you should be safe.
> > 
> > Regards,
> > Benjamin
> > 
> 
> 
> ------------------------------------------------------------------------------
> Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT 
> is a gathering of tech-side developers & brand creativity professionals. Meet
> the minds behind Google Creative Lab, Visual Complexity, Processing, & 
> iPhoneDevCamp as they present alongside digital heavyweights like Barbarian 
> Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com 
> _______________________________________________
> Fedora-commons-developers mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/fedora-commons-developers


------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises 
looking to deploy the next generation of Solaris that includes the latest 
innovations from Sun and the OpenSource community. Download a copy and 
enjoy capabilities such as Networking, Storage and Virtualization. 
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
Fedora-commons-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/fedora-commons-developers

Reply via email to