Re: [greenstone-users] Newbie seeking some guidance

From graeme
DateFri, 21 Jul 2006 01:30:32 -0400
Subject Re: [greenstone-users] Newbie seeking some guidance
In-Reply-To (44C013B4-2010904-cs-waikato-ac-nz)


On 7/20/06, Michael Dewsnip <mdewsnip@cs.waikato.ac.nz> wrote:
Hi Graeme,

> The first is concerning the browser classifier. I was following the
> conversation on GenericList classifier but I'm not certain how to
> approach this in my case. I have to support two scripts, essentially
> Latin and Arabic. How can I use the GenericList to support both, I
> tried but ended up with just Latin script entries.

I can't quite envisage what you mean here -- do you want two versions of
the browser, one in Latin and one in Arabic? (And only see one at a time?)


Sorry I'll try and explain the situation better. We have one collection that consists of texts that are written in the Latin script and texts that are written in the Arabic script. I want to be able to present these as a single collection. At the moment we have 4,600 titles (and growing). I have a number of different ways to browse these texts, by language, by author, by title etc.
What I want to be able to do is to go to the browse by title and have them appear in some kind of sub groups. At the moment I have all 4,600 on the same page. When I tried to group them by the first letter I tried the AtoZ classifier which worked but it only included those texts that had some Latin script in the title. I want to be able to generate the sub groups that include both the Latin and the Arabic script. (sorry I don't have greenstone up at the moment so I may of got the terminology wrong). I envisage that it would take the first letter of the title and if it was from the Latin script classify it accordingly, if t was from the Arabic script classify it according to the Arabic alphabet, I don't mind what it does if if finds a character that doesn't fall into either alphabet, such as a number, so long as it doesn't lose it.
When I tried the AtoZ it looked for a character from the Latin alphabet when it found one it classified it according to that letter, if it didn't find a Latin character then it just ignored the text.

Sorry for the long message, but I hope that explains what it is I wish to achieve and the problem that I have been facing.

many thanks,

graeme.