Use the function udm_get_res_field_ex() to fetch "non-standard" sections.
Example:
<?php
$tags = udm_get_res_field_ex($agent, $row, "meta.mytag");
?>
(PHP 4 >= 4.0.5, PHP 5 < 5.1.0, PECL mnogosearch >= 1.0.0)
udm_get_res_field — Fetch a result field
$res
, int $row
, int $field
) : stringFetch a mnoGoSearch result field.
res
res - a link to result identifier, received
after call to udm_find().
row
row - the number of the link on the current page.
May have values from 0 to
UDM_PARAM_NUM_ROWS-1.
field
field - field identifier, may have the following values:
UDM_FIELD_URL - document URL field
UDM_FIELD_CONTENT - document
Content-type field (for example, text/html).
UDM_FIELD_CATEGORY - document category field. Use
udm_cat_path() to get full path to current category
from the categories tree root. (This parameter is available only in PHP
4.0.6 or later).
UDM_FIELD_TITLE - document title field.
UDM_FIELD_KEYWORDS - document keywords field (from META KEYWORDS tag).
UDM_FIELD_DESC - document description field (from META DESCRIPTION tag).
UDM_FIELD_TEXT - document body text (the first couple of lines to give an
idea of what the document is about).
UDM_FIELD_SIZE - document size.
UDM_FIELD_URLID - unique URL ID of the link.
UDM_FIELD_RATING - page rating (as calculated by mnoGoSearch).
UDM_FIELD_MODIFIED - last-modified field in unixtime format.
UDM_FIELD_ORDER - the number of the current document in set of found documents.
UDM_FIELD_CRC - document CRC.
udm_get_res_field() returns result field value on
success, FALSE on error.
Use the function udm_get_res_field_ex() to fetch "non-standard" sections.
Example:
<?php
$tags = udm_get_res_field_ex($agent, $row, "meta.mytag");
?>
It seems the description length is now determined by UDM_MAXDESCSIZE in udm_common.h instead of UDM_MAXTEXTSIZE
To further improve the "description" field quality, there are two additional techniques you can use.
First, you can define this value if you know it, by setting a description meta tag in the head section.
Second, mnogosearch has an <index></noindex> tag pair that appears to make the enclosed content completely invisible to the search engine. I've used this on the sites I maintain to 'vanish' the navigational elements, field names, and other non-content type HTML and page window dressing so that only the important content is indexed and reported on.
This last technique as an added (and more important benefit) of helping to eliminate a lot of false matches in searches.
The returned page description is only 100 characters long.
To store more information, you can change the appropriate database field length to 255 characters (field "description" in the table "url"), and change the following line in udm_common.h.
#define UDM_MAXTEXTSIZE 255