[thelist] Architecture for arbitrary data management

Daniel Burke dan.p.burke at gmail.com
Mon Mar 3 04:02:04 CST 2008


As it's arbitrary user data, and your targeting linux, I'd consider
using xml files, and searching them with beagle. At least that way
you'll get semi fast searching from lots of text data. If your
genuinely stuck with LAMP and nothing else, well, tough call.

On Sat, Mar 1, 2008 at 1:18 AM, Joe Flintham <list at menticulture.com> wrote:
> kasimir-k wrote:
>  > I'd approach it with something like this:
>  >
>  > table items
>  > - itemId
>  > - userId
>  > - timeAdded
>  >
>  > table properties
>  > - propertyId
>  > - propertyName
>  >
>  > table items_x_properties
>  > - itemId
>  > - propertyId
>  > - propertyValue
>  >
>  > .k
>  >
>  Thanks kasimir - one question:
>
>  presumably you'd make items_x_properties.propertyValue a longtext field,
>  even though the values could be anything from an integer to a float to a
>  short string to a journal length article?
>
>  Joe
>
>
> --
>
>  * * Please support the community that supports you.  * *
>  http://evolt.org/help_support_evolt/
>
>  For unsubscribe and other options, including the Tip Harvester
>  and archives of thelist go to: http://lists.evolt.org
>  Workers of the Web, evolt !
>



More information about the thelist mailing list