|
Posted by Jeff R. on August 29, 2006, 6:38 pm
Please log in for more thread options
To index a database you will need a protocol handler to allow the indexer to
crawl the database. For the items in the database you will need an IFilter
if the items are not of an already known type, meaning if there isn't
already an IFilter in place that is capable of parsing/indexing the items.
hope that helps.
>I want to use WDS to index my application.
>
> The application (simply) consists of a database containing meta-data about
> files. The files are (mainly but not exclusively) Office documents
> contained
> somewhere in the filesystem.
>
> I want to be able to enter a word in the Desktop Search search-box, the
> results should include files where the word is found
> i) in the file itself.
> ii) in the meta-data or
>
> i) I guess is simple - I just ensure the location is included in the
> search
> locations.
>
> but
> ii) - Where do I start?
>
> The database is a single .db file. I have an API which will return me
> key\value pairs from the metadata.
>
> I'm not sure what the process should be here.
> If I write an IFilter what will the IFilter filter?
> Do I need to include the path containing the .db file in the indexed
> locations or can I index a specific file?
> How does Windows know to send this type of .db file to this IFilter. Will
> my
> IFilter receive every file? Or every file with a .db extension? Or just
> this
> .db file?
> When will the .db file be sent to the IFilter (& if so how).
>
> When the IFilter processes the .db what should it do to say:
> "This piece of metadata is accociated with this file"
> rather than:
> "This piece of metadata is from the .db file"
>
>
>
> I think I need a dummies guide
| 
XML Sitemap