翻译强力驱动i need to extract and show some words before and after a query word, something like google search results, for example:
$str = "hi user! welcome to new php open source world, we are trying to learn you something!"; $query = "new php"; $result = "... welcome to new php open source ...";i searched google an SO but didn't find a clear answer or maybe my php knowledge was not enough! is there a workable and easy-to-use function to do this job?
function yourFuncName($str, $query, $numOfWordToAdd) { list($before, $after) = explode($query, $str); $before = rtrim($before); $after = ltrim($after); $beforeArray = array_reverse(explode(" ", $before)); $afterArray = explode(" ", $after); $countBeforeArray = count($beforeArray); $countAfterArray = count($afterArray); $beforeString = ""; if($countBeforeArray < $numOfWordToAdd) { $beforeString = implode(' ', $beforeArray); } else { for($i = 0; $i < $numOfWordToAdd; $i++) { $beforeString = $beforeArray[$i] . ' ' . $beforeString; } } $afterString = ""; if($countAfterArray < $numOfWordToAdd) { $afterString = implode(' ', $afterArray); } else { for($i = 0; $i < $numOfWordToAdd; $i++) { $afterString = $afterString . $afterArray[$i] . ' '; } } $string = $beforeString . $query . ' ' . $afterString; return $string; }Output is: user! welcome to new php open source world, ($numOfWordToAdd = 3)
Here is an working example I thing that it is clear what I did and how:
<?php $str = "hi user! welcome to new php open source world, we are trying to learn you something!"; $query = "new php"; $expl = explode($query, $str); // items on the left side of middle string $expl_left = explode(" ", $expl[0]); $left_cnt = count($expl_left); $new_left = $expl_left[$left_cnt-3] . " " . $expl_left[$left_cnt-2]; // items on the right side of middle string $expl_right = explode(" ", $expl[1]); $new_right = $expl_right[1] . " " . $expl_right[2]; // new string formated $new = "... " . $new_left . " " . $query . " " . $new_right . " ..."; print $new; ?>If you have some questions feel free to ask...
$result = preg_replace('/(.+)?([^\s]+.{10}'.$query.'.{10}[^\s]+)(.+)?/', '... $2 ...', $str);This will return the same result from the same string and query you gave. If the before or after length starts or ends (respectively) in the middle of a word, it will continue until it completes the word before it stops.
Assuming a "word" is any series of non-whitespace characters, the following will extract 3 words on either side of new php out of the string $subject, but accept less if necessary:
if (preg_match('/(?:\S+\s+){1,3}new php(?:\s+\S+){1,3}/', $subject, $regs)) { $result = $regs[0]; }Change the 3s to any number you like.
I used the following function with explode:
public static function returnSearch($query, $str, $wordcount) { $explode = explode($query, $str); $result = null; //if explode count is one the query was not found if (count($explode) == 1) { $result = implode(' ', array_slice(str_word_count($explode[0], 2), -$wordcount, $wordcount)) . " "; } //if explode count is more than one the query was found at least one time if (count($explode) > 1) { //check for if the string begins with the query if (!empty($explode[0])) { $result = "..." . implode(' ', array_slice(str_word_count($explode[0], 2), -$wordcount, $wordcount)) . " "; } $result = $result . $query; if (!empty($explode[1])) { $result = $result . " " . implode(' ', array_slice(str_word_count($explode[1], 2), 0, $wordcount)) . "..."; } } //return result return $result; }