[dev] Re: IMP Filters Array Structure.

[Michael M Slusarz]

Quoting Mike Cochrane <mike@graftonhall.co.nz>:

|         * Filter entry (stored in preferences):
|         * Array
|         * (
|         *   [bl] => Array
|         *   (
|         *     *List of blacklisted addresses*
|         *   )
|         *
|         *   [rule] => Array
|         *   (
|         *     [*Filter number*] => Array
|         *     (
|         *       [flt] => Array
|         *       (
|         *         [] => Array
|         *           (
|         *             [fld] => Array(*Field name(s)*)
|         *             [txt] => Array(*Text to match*)
| >                     [typ] => (constant or string) Match Type (is,
| isn't,
| contains, begins with, ends with etc)
|         *           )
|         *       )
| >               [nme] => (string) Filter Name
| >               [cmb] => (constant) *Combine code* (AND OR)
| >               [cas] => (boolean)Case Sesitive
|         *       [act] => *Action code*
|         *       [fol] => *Folder name to move to*
| also [fol] might be better named target as it would make more sense for
| redirects.
|         *     )
|         *   )
|      * )
| 
| Just my thought as then this could be integrated into IMP and a drop in
| replacement for the current filtering....

The only things that are necessary is the [bl] array, the [rule] array, the 
[*filter number*] array, and the [flt] array.  Additional fields should not 
break the current filtering code.  It has been discussed before to have a 
filtering driver system with different drivers (e.g. the basic "IMP 
preferences" filter driver, a siever filter driver, etc.) so if this is 
something you want to do, I'd say go for it :)

michael

______________________________________________
Michael Slusarz [slusarz@bigworm.colorado.edu]
The University of Colorado at Boulder