frjo/swedishstemmer6/swedishstemmer.module

<?php
// $Id: swedishstemmer.module,v 1.1 2008/10/15 19:23:54 frjo Exp $

/**
 * @file
 * "Improve Swedish language searching by simplifying related words to their root (verbs, plurals, ...).
 * Algorithm based on http://snowball.tartarus.org/algorithms/swedish/stemmer.html.
 */

/**
 * Implementation of hook_search_preprocess.
 */
function swedishstemmer_search_preprocess($text) {
  // Split words from noise and remove apostrophes
  $words = preg_split('/([^a-zA-ZéåäöÅÄÖ\']+)/u', $text, -1, PREG_SPLIT_DELIM_CAPTURE);

  // Process each word
  $odd = true;
  foreach ($words as $k => $word) {
    if ($odd) {
      $words[$k] = swedishstemmer_stem($word);
    }
    $odd = !$odd;
  }

  // Put it all back together
  return implode('', $words);
}

/**
 * Stem a Swedish word.
 */
function swedishstemmer_stem($word) {
  // Lowercase
  $word = drupal_strtolower($word);

  /* R1 is the region after the first non-vowel following a vowel, or is the
     null region at the end of the word if there is no such non-vowel. */
  if (preg_match('/[aeiouyäåö][^aeiouyäåö]/u', $word, $matches, PREG_OFFSET_CAPTURE)) {
    $r1 = $matches[0][1] + 2;
  }

  // Steps 1-3: suffix removal
  $word = swedishstemmer_step1($word, $r1);
  $word = swedishstemmer_step2($word, $r1);
  $word = swedishstemmer_step3($word, $r1);

  return $word;
}

function swedishstemmer_step1($word, $r1) {
  // Step 1:
  // Search for the longest among the following suffixes in R1, and perform the action indicated.
  if ($r1) {
    $word = preg_replace(array_reverse(array('/a$/', '/arna$/', '/erna$/', '/heterna$/', '/orna$/', '/ad$/', '/e$/', '/ade$/', '/ande$/', '/arne$/', '/are$/', '/aste$/', '/en$/', '/anden$/', '/aren$/', '/heten$/', '/ern$/', '/ar$/', '/er$/', '/heter$/', '/or$/', '/as$/', '/arnas$/', '/ernas$/', '/ornas$/', '/es$/', '/ades$/', '/andes$/', '/ens$/', '/arens$/', '/hetens$/', '/erns$/', '/at$/', '/andet$/', '/het$/', '/ast$/')), '', $word, 1);
  }

  // Delete 's' if preceded by a valid s-ending
  $word = preg_replace('/([bcdfghjklmnoprtvy])s$/', '\\1', $word);

  return $word;
}

function swedishstemmer_step2($word, $r1) {
  // Step 2:
  // Search for one of the following suffixes in R1, and if found delete the last letter.
  if ($r1) {
    $word = preg_match('/(dd|gd|nn|dt|gt|kt|tt)$/', $word) ? substr($word, 0, -1) : $word;
  }

  return $word;
}

function swedishstemmer_step3($word, $r1) {
  // Step 3:
  // Search for the longest among the following suffixes in R1, and perform the action indicated.
  if ($r1) {
    $word = preg_replace('/(lig|ig|els)$/', '', $word);
    $word = preg_replace('/löst$/', 'lös', $word);
    $word = preg_replace('/fullt$/', 'full', $word);
  }

  return $word;
}
1	<?php
2	// $Id: swedishstemmer.module,v 1.1 2008/10/15 19:23:54 frjo Exp $
3
4	/**
5	* @file
6	* "Improve Swedish language searching by simplifying related words to their root (verbs, plurals, ...).
7	* Algorithm based on http://snowball.tartarus.org/algorithms/swedish/stemmer.html.
8	*/
9
10	/**
11	* Implementation of hook_search_preprocess.
12	*/
13	function swedishstemmer_search_preprocess($text) {
14	// Split words from noise and remove apostrophes
15	$words = preg_split('/([^a-zA-ZéåäöÅÄÖ\']+)/u', $text, -1, PREG_SPLIT_DELIM_CAPTURE);
16
17	// Process each word
18	$odd = true;
19	foreach ($words as $k => $word) {
20	if ($odd) {
21	$words[$k] = swedishstemmer_stem($word);
22	}
23	$odd = !$odd;
24	}
25
26	// Put it all back together
27	return implode('', $words);
28	}
29
30	/**
31	* Stem a Swedish word.
32	*/
33	function swedishstemmer_stem($word) {
34	// Lowercase
35	$word = drupal_strtolower($word);
36
37	/* R1 is the region after the first non-vowel following a vowel, or is the
38	null region at the end of the word if there is no such non-vowel. */
39	if (preg_match('/[aeiouyäåö][^aeiouyäåö]/u', $word, $matches, PREG_OFFSET_CAPTURE)) {
40	$r1 = $matches[0][1] + 2;
41	}
42
43	// Steps 1-3: suffix removal
44	$word = swedishstemmer_step1($word, $r1);
45	$word = swedishstemmer_step2($word, $r1);
46	$word = swedishstemmer_step3($word, $r1);
47
48	return $word;
49	}
50
51	function swedishstemmer_step1($word, $r1) {
52	// Step 1:
53	// Search for the longest among the following suffixes in R1, and perform the action indicated.
54	if ($r1) {
55	$word = preg_replace(array_reverse(array('/a$/', '/arna$/', '/erna$/', '/heterna$/', '/orna$/', '/ad$/', '/e$/', '/ade$/', '/ande$/', '/arne$/', '/are$/', '/aste$/', '/en$/', '/anden$/', '/aren$/', '/heten$/', '/ern$/', '/ar$/', '/er$/', '/heter$/', '/or$/', '/as$/', '/arnas$/', '/ernas$/', '/ornas$/', '/es$/', '/ades$/', '/andes$/', '/ens$/', '/arens$/', '/hetens$/', '/erns$/', '/at$/', '/andet$/', '/het$/', '/ast$/')), '', $word, 1);
56	}
57
58	// Delete 's' if preceded by a valid s-ending
59	$word = preg_replace('/([bcdfghjklmnoprtvy])s$/', '\\1', $word);
60
61	return $word;
62	}
63
64	function swedishstemmer_step2($word, $r1) {
65	// Step 2:
66	// Search for one of the following suffixes in R1, and if found delete the last letter.
67	if ($r1) {
68	$word = preg_match('/(dd\|gd\|nn\|dt\|gt\|kt\|tt)$/', $word) ? substr($word, 0, -1) : $word;
69	}
70
71	return $word;
72	}
73
74	function swedishstemmer_step3($word, $r1) {
75	// Step 3:
76	// Search for the longest among the following suffixes in R1, and perform the action indicated.
77	if ($r1) {
78	$word = preg_replace('/(lig\|ig\|els)$/', '', $word);
79	$word = preg_replace('/löst$/', 'lös', $word);
80	$word = preg_replace('/fullt$/', 'full', $word);
81	}
82
83	return $word;
84	}