This is where ht://Dig's templates are put to work. First, let's count the number of lines returned by the search:

<?php
$rc = count($result);
?>

The first two lines are always going to be the HTTP header

Content-type: text/html \n
\n

so we can skip them (remember, ht://Dig was designed to output HTML).

If the search produced an error, we'd get no results or a one-line error, so:

<?php
if ($rc<3):
     echo "There was an error executing this query.  Please try later.\n";
?>

Let's also check to see that we got some matches. If we didn't, then ht://Dig will just echo out the two-line HTTP header, then the contents of the results-nomatch.html file. We use that information to perform the check:

<?php
elseif ($result[2]=="NOMATCH"):
 echo "There were no matches for <B>$search</B> found on the website.<P>\n";
?>

We know to look for "NOMATCH" because that's the string in results-nomatch.html).

Similarly, we can check for a boolean syntax error:

<?php
elseif ($result[2]=="SYNTAXERROR"):
     echo "There is a syntax error in your search for <B>$search</B>:<BR>";
     echo "<PRE>" . $result[3] . "</PRE>\n";
?>

If none of the above conditions were true, then we have at least one match! In this case, ht://Dig first outputs the variables in the results-header.html template, so:

<?php
else:
     $matches = $result[2];
     $firstdisplayed = $result[3];
     $lastdisplayed = $result[4];
     $words = $result[5];

     echo "Your search for <B>$words</B> returned <B>$matches</B> match";
     echo ($matches==1) ? "" : "es";
?>

I'm echoing out the variable words here, instead of $search. That's because if I have fuzzy searching turned on and I search for "play", then $search will equal "play", but $words will equal "(play or played or playing or player or plays or players)" ... which what you are really searching for.

Now we echo out each of the matches.

<?php
     $i=6;
?>

(6 is the number of variables in results-header plus 2)

<?php
     while($i<$rc) {

          # grab the match information

          $title = $result[$i];
          $url = $result[$i+1];
          $percent = $result[$i+2];
          $excerpt = $result[$i+3];

          # output the match information

          echo "<A HREF=\"" . $url . "\">" . $title . 
		                   "</A><BR>\n";
          echo "(" . $percent . "% match)<BR>\n";
          echo "<blockquote>" . $excerpt . "</blockquote><BR><BR>\n";

          # move to the next match

          $i = $i + 4;
     }

endif;
?>

And that's it!

There are a few quirks to keep in mind. Notably, ht://Dig outputs an additional new-line after the $(STARSLEFT) and $(STARSRIGHT) variables in the template. You need to keep this in mind when figuring out which line of $result corresponds to what piece of information. For example, if I changed $(PERCENT) to $(STARSLEFT) in my .conf file, I would need to make the following changes to the code:

<?php
          $title = $result[$i];
          $url = $result[$i+1];
          $stars = $result[$i+2];
          $excerpt = $result[$i+4];

          ...

          $i=$i+5;
?>

But the basic strategy is always the same: PHP loops through the array $result and outputs the information. If you want to see the contents of $result for yourself (to make sure you're getting the right results), just replace the entire last else- block above with:

<?php
     while (list($k,$v)=each($result)) {
          echo "$k -> $v \n";
     }
?>

You should all be running out now and adding "Search this Site" buttons to your PHP-driven web pages. This tutorial covered the basics, but there are a lot of advanced things you can do with ht://Dig and PHP:

• show matches in groups (e.g. "Matches 1-10", "Matches 11-20")
• "fix" the URL for matches for sites that are in framesets
• index and search not only HTML and PHP pages, but also PDF (Acrobat) and Microsoft Word documents using external parsers
• use the raw results of the indexing process to generate a site-map

Look for more tutorials later on these and other topics.

But until then, happy searching!