Mailing-List: contact dev-help@clerezza.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@clerezza.apache.org
Received-SPF: pass (nike.apache.org: local policy includes SPF record at
 spf.trusted-forwarder.org)
MIME-Version: 1.0
In-Reply-To: <etPan.524c4c42.4516dde9.1c43@dhcp-10-0-1-108.searchbox.lan>
References: <etPan.524c4a50.7c3dbd3d.1c43@dhcp-10-0-1-108.searchbox.lan>
	<etPan.524c4c42.4516dde9.1c43@dhcp-10-0-1-108.searchbox.lan>
Date: Tue, 8 Oct 2013 16:22:17 +0200
Message-ID: 
 <CALvhUEUGuDa5LfH+YH9hg6kawQMeaLG+aW3YvDiu1SXs+aKjtw@mail.gmail.com>
Subject: Re: VirtualProperty Typing
From: =?ISO-8859-1?Q?Reto_Bachmann=2DGm=FCr?= <reto@wymiwyg.com>
To: dev@clerezza.apache.org
Content-Type: multipart/alternative; boundary=001a11c3799cfabd8104e83b7e4f

--001a11c3799cfabd8104e83b7e4f
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable

Hi Stephane

(found this mail in my draft, sorry for the delay)

On Wed, Oct 2, 2013 at 6:39 PM, Stephane Gamard <stephane@gamard.net> wrote=
:

> While we're at it, and since it does not hurt basically, I would propose
> that we can add the following attributes to any VirtualProperties:
> - isSearchable
>
Don't get this. Why should somebody define a non-searcheable virtual
property?


> - isType (high level, ie: name, text, date, quantity, =85)
>

ok, could you give the complete list of them with a short description.

Cheers,
Reto

>
> This could greatly improve search retrieval and quality over the time of
> next implementation. These attributes might not be mandatory at all, but
> can have a great added value.
>
> _Stephane
>
>
> On October 2, 2013 at 6:31:13 PM, Stephane Gamard (stephane@gamard.net)
> wrote:
>
> Hi Team,
>
> As I dive deeper and deeper into my learning and updates on the rdf.cris
> I've come to realise that the Faceting implementation is not optimal for
> Lucene. We're experiencing extremely slow faceting do to the post-search
> facet collector that iterates thru the document list untill hits.length.
>
> It would be fairly trivial now to implement facets as per Lucene
> specifications. The most "drastic" change that I am yet too ignorant to t=
ry
> for myself is to have the ability to know when a VirtualProperty should b=
e
> considered for faceting at indexing time.
>
> This would require a small update of the DefinitionGraph by maybe adding =
a
> property "facetable" for VirtualProperties?
>
> _Stephane
>
>
>

--001a11c3799cfabd8104e83b7e4f--