[Nepomuk] Re: Indexing

[Vishesh Handa]

On Mon, Sep 27, 2010 at 11:01 PM, Bèrto ëd Sèra <berto.d.sera at gmail.com>wrote:

> Hi!
>
>> Internally Nepomuk uses a recursive call on each directory which is to be
>> indexed, filters out all the files/folders and then passes the files it
>> should index to Strigi. The amount of data read ( from the file ) depends on
>> the Strigi Analyzer. They don't usually (?) read the contents of the entire
>> file, but I'm not too sure about this.
>>
>> What I thought is that this stuff IS already indexed (both on the shared
> server and on another login, same laptop). Wouldn't it be nice if one could
> import it and save loads of process/network traffic?
>
>
Ahh. I see what you mean. Well, there is no simple way to do it right now,
but I'm working on it. It's one of the things which I want to do once I
complete my Sync Library (which I'm in the process of rewriting) The problem
is very similar to the remove-able media problem - we want stuff indexed on
a removable medium, but we don't want to re-index it every time it is
plugged into a different Nepomuk enabled system.

The main problem with importing stuff is identifying the resources
correctly. Nepomuk assigns a random uri to each new resource. This is what I
worked on in the summer. So, it's just a matter of improving the Sync
Library's API, and moving it to kdelibs.

- Vishesh Handa


> Bèrto
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.kde.org/pipermail/nepomuk/attachments/20100928/df6d8445/attachment.htm