Topic: highlighting search terms

[michaellunsford]

well, I started this in the regular PHP section, but it no longer fits there. Suffice it to say, I'm trying to take individual search terms that are being $_POSTed and highlighting them in the search results.

The Original Post talked about using str_replace to handle this. New problem, though, when the same search terms show up inside a HTML tag (like <img src="search_term">).

I'm trying "/\b(?!<.+?>)search_term\b/" -- but it's still finding "search_term" inside <img src="search_term">.

Thanks!

[effigy]

Either separate the tags from the content and process, or just analyze the non-tagged content:

Code: [Select]

<pre>
<?php
$tests = array( 
'<img src="search_term">',
'<a>search_term</a>',
'<a>Xsearch_termX</a>',
);

$term = 'search_term';
echo "Searching for <b>$term</b>...<br>";
foreach ($tests as $test) {
echo htmlspecialchars($test), ' => ';
$test = preg_replace_callback(
'/>(.+?)</',
create_function(
'$matches',
'return preg_replace("/\b(' . preg_quote($term) . ')\b/", "<b>\\\1</b>", $matches[0]);'
),
$test
); 
echo htmlspecialchars($test), '<br>';
}
?>
</pre>



[michaellunsford]

Wow, that's intense. All kinds of functions I've never heard of. In fact after reading the manual page on create_function(), I still don't understand it. $matches doesn't exist anywhere outside of create_function()???

Anyway, it's doing the same thing my original preg_replace() was doing (finding matches inside tags). Here's the new function, minus the array stuff:

Code: [Select]

<?php
$test="<table>\r\n<tr>\r\n\t<td><img src=\"somestring.jpg\" alt=\"\"></td>\r\n</tr>\r\n<tr>\r\n\t<td>somestring</td>\r\n</tr>\r\n</table>\r\n";
$term = 'somestring';
$test = preg_replace_callback(
'/>(.+?)</',
create_function(
'$matches',
'return preg_replace("/\b(' . preg_quote($term) . ')\b/", "<span style=\"background:#FF0;\">\\\1</span>", $matches[0]);'
),
$test
); 
echo $test;
?>
and the output

Code: [Select]

<table>
<tr>
        <td><img src="<span style="background:#FF0;">somestring</span>.jpg" alt=""></td>
</tr>
<tr>
        <td><span style="background:#FF0;">somestring</span></td>
</tr>
</table>

[michaellunsford]

hey, check this out...

Code: [Select]

<?php
$test="<table>\r\n<tr>\r\n\t<td><img src=\"somestring.jpg\" alt=\"\"></td>\r\n</tr>\r\n<tr>\r\n\t<td>somestring</td>\r\n</tr>\r\n</table>\r\n";
preg_match("/>(.+?)</",$test,$new_array);
print_r($new_array);
?>
result:

Code: [Select]

Array
(
    [0] => ><img src="somestring.jpg" alt=""><
    [1] => <img src="somestring.jpg" alt="">
)


[effigy]

Ah, of course. Change />(.+?)</ to />(.*?)</.

[michaellunsford]

First, let me say it works great!

However, I don't understand why />(.*?)</ is returning ">somestring<" instead of just "somestring" ???

[effigy]

Where? The entire match includes the angle brackets, but the 1st capture does not. It depends on which part you're using.

[michaellunsford]

here's what />(.*?)</ matches in the example:

Code: [Select]

    [0] => Array
        (
            [0] => ><
            [1] => ><
            [2] => > <
            [3] => >somestring<
            [4] => > <
        )

    [1] => Array
        (
            [0] =>
            [1] =>
            [2] => 
            [3] => somestring
            [4] => 
        )


[effigy]

A better alternative might be /(?<=>)([^<]+)/.

[michaellunsford]

Dude, it's magic. One more thing that is not quite working right: MySQL is returning case insensitive results. Is there any way to make '/\b(' . preg_quote($newstring) . ')\b/' case insensitive?

Also, if you're willing to educate, I'm struggling to follow this part of the code. (?<=>) is a lookbehind for ">"? It's working perfectly to locate the text I'm searching for -- even text that isn't preceeded by ">". I don't get it, is lookbehind optional?

[effigy]

Yes, put an "i" after the closing delimiter: /pattern/i

You are correct about the lookbehind; however, they are not optional (by default). Why do you think it is matching text that isn't preceded by ">"?

[michaellunsford]

It's finding the search term, even after newlines. So, I guess (?<=>) evades newlines, too?

[effigy]

([^<]+) is capturing the CRs and NLs, and (?<=>) is still anchoring at ">". Observe:

Code: [Select]

<pre>
<?php
$test = "<table>\r\n<tr>\r\n\t<td><img src=\"somestring.jpg\" alt=\"\"></td>\r\n</tr>\r\n<tr>\r\n\t<td>somestring</td>\r\n</tr>\r\n</table>\r\n";
preg_match_all('/(?<=>)([^<]+)/', $test, $matches);
$replace = array(
"\n" => '\n',
"\r" => '\r',
);
foreach ($matches as &$array) {
foreach ($array as &$match) {
$match = preg_replace('/([\r\n])/e', '$replace["\1"]', $match);
}
}
print_r($matches);
?>
</pre>



[michaellunsford]

ahhh... I should have known that all this late night programming must be killing brain cells

[fataqui]

I do like this...

<?php

$arrayofwords = array ();

$arrayofwords[0] = "This";

$arrayofwords[1] = "text";

$arrayofwords[2] = "need";

$arrayofwords[3] = "words";

$str = 'This is my <img src="" title="This image text"> long text <a href="#">words</a> where I need to highlight words in the HTML text.';

$str = preg_replace ( "/(?!(?:[^<]+>|[^>]+<\/a>))\b(" . implode ( '|', $arrayofwords ) . ")\b/is", "<strong>\\1</strong>", $str );

echo $str;

?>

[lwc]

Of course, your code only works in sites with strictly Latin words. Otherwise, see this.

[ddrudik]

I would suggest:
search_term(?!(?=[^<>]*>))

[ddrudik]

Also, if you want to exclude search_term from within head/script/a blocks as well as from within tags:

Code: [Select]

$html=preg_replace_callback('~(<head>.*?</head>|<script\s[^>]*>.*?</script>|<a\s[^>]*>.*?</a>)|search_term(?!(?=[^<>]*>))~is',create_function('$matches','return isset($matches[1]) ? $matches[1] : "<strong>$matches[0]</strong>" ;'),$html);



[lwc]

I do like this...

I like this too. But can you deal with "&lt;" and "&gt;" too?