Name

search_excerpt — Returns excerpts with hit words from text

Synopsis

search_excerpt ( in hit_words any ,
  in text varchar ,
  in within_first int ,
  in max_excerpt int ,
  in total int ,
  in html_hit_tag varchar );
 

Description

This function produces representative samples for use in displaying a query hit. There are two modes: html mode and text mode. In html mode everything looks like html tags is ignored and searching of excerpt begins from <body> tag. All found hit words highlighted by html_hit_tag.

In text mode text is treated as plain text, html tag detection is disabled and hit words is not highlighted.

Parameters

hit_words

array of hit words to be found in text. Number of hit words can not be more than 10.

text

original text where hit words are searched

within_first

number of chars in text from the start to consider. Default value is 20000.

max_excerpt

maximum length of single excerpt phrase. Default value is 90.

total

maximum length of whole excerpt. Default value is 200.

html_hit_tag

HTML tag which is used to highlight hit words in excerpt. Default value is "b". If it is NULL text mode is used

Return Types

varchar

Examples

Example 24.365. Creating search excerpt

creates a search excerpt from found DAV resource

for select RES_CONTENT from WS.WS.SYS_DAV_RES 
 where contains (RES_CONTENT, 'place and knowledge')
do {
  http_value (search_excerpt (vector ('place', 'knowledge'), 
                              cast (RES_CONTENT as varchar), 
                              200000, 90, 200, 'b', 1),
              'P');
}