frjo/swedishstemmer5/swedishstemmer.module

<?php
// $Id: site_map.module,v 1.30 2007/03/21 08:14:34 frjo Exp $

// 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: site_map.module,v 1.30 2007/03/21 08:14:34 frjo Exp $
3
4	// Algorithm based on http://snowball.tartarus.org/algorithms/swedish/stemmer.html.
5
6	/**
7	* Implementation of hook_search_preprocess.
8	*/
9	function swedishstemmer_search_preprocess(&$text) {
10	// Split words from noise and remove apostrophes
11	$words = preg_split('/([^a-zA-ZéåäöÅÄÖ\']+)/u', $text, -1, PREG_SPLIT_DELIM_CAPTURE);
12
13	// Process each word
14	$odd = true;
15	foreach ($words as $k => $word) {
16	if ($odd) {
17	$words[$k] = swedishstemmer_stem($word);
18	}
19	$odd = !$odd;
20	}
21
22	// Put it all back together
23	return implode('', $words);
24	}
25
26	/**
27	* Stem a Swedish word.
28	*/
29	function swedishstemmer_stem($word) {
30	// Lowercase
31	$word = drupal_strtolower($word);
32
33	/* R1 is the region after the first non-vowel following a vowel, or is the
34	null region at the end of the word if there is no such non-vowel. */
35	if (preg_match('/[aeiouyäåö][^aeiouyäåö]/u', $word, $matches, PREG_OFFSET_CAPTURE)) {
36	$r1 = $matches[0][1] + 2;
37	}
38
39	// Steps 1-3: suffix removal
40	$word = swedishstemmer_step1($word, $r1);
41	$word = swedishstemmer_step2($word, $r1);
42	$word = swedishstemmer_step3($word, $r1);
43
44	return $word;
45	}
46
47	function swedishstemmer_step1($word, $r1) {
48	// Step 1:
49	// Search for the longest among the following suffixes in R1, and perform the action indicated.
50	if ($r1) {
51	$word = preg_replace(array_reverse(array('/a$/', '/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);
52	}
53
54	// Delete 's' if preceded by a valid s-ending
55	$word = preg_replace('/([bcdfghjklmnoprtvy])s$/', '\\1', $word);
56
57	return $word;
58	}
59
60	function swedishstemmer_step2($word, $r1) {
61	// Step 2:
62	// Search for one of the following suffixes in R1, and if found delete the last letter.
63	if ($r1) {
64	$word = preg_match('/(dd\|gd\|nn\|dt\|gt\|kt\|tt)$/', $word) ? substr($word, 0, -1) : $word;
65	}
66
67	return $word;
68	}
69
70	function swedishstemmer_step3($word, $r1) {
71	// Step 3:
72	// Search for the longest among the following suffixes in R1, and perform the action indicated.
73	if ($r1) {
74	$word = preg_replace('/(lig\|ig\|els)$/', '', $word);
75	$word = preg_replace('/löst$/', 'lös', $word);
76	$word = preg_replace('/fullt$/', 'full', $word);
77	}
78
79	return $word;
80	}