Add a token to get the top Taxonomy Term Synonym

Oops, posted the code without the ending php tags

/**
 * Implementation of hook_token_list().
 */
function token_synonym_token_list($type = 'all') {
  $tokens = array();

  if ($type == 'node' || $type == 'all') {
    $tokens['node']['term-synonym']  = t("The top taxonomy term synonym");
  }

  return $tokens;
}

/**
 * Implementation of hook_token_values().
 */
function token_synonym_token_values($type, $object = NULL) {
  $values = array();
  switch ($type) {
    case 'node':
    	$node = $object;
    	foreach($node->taxonomy as $term){$term_array = taxonomy_get_synonyms($term->tid); }
		$values['term-synonym'] = $term_array[0];
		break;
  }
  return $values;
} 

Now I'm really lost. I figured I was doing something wrong in my code (even though it worked correctly when you'd update the path) so I copied the code from the token_node.inc file.

This does not work when the node is created or when the path is updated from the content list:

/**
 * @file
 * Token Synonym - Returns the top synonym term for a node
 *
 * @ingroup token
 */

/**
 * Implementation of hook_token_list().
 */
function token_synonym_token_list($type = 'all') {
  $tokens = array();

  if ($type == 'node' || $type == 'all') {
    $tokens['node']['term-synonym']  = t("The top taxonomy term synonym");
  }

  return $tokens;
}

/**
 * Implementation of hook_token_values().
 */
function token_synonym_token_values($type, $object = NULL) {
  $values = array();
  switch ($type) {
    case 'node':
      $node = $object;

      // And now taxonomy, which is a bit more work. This code is adapted from
      // pathauto's handling code; it's intended for compatibility with it.
      if (module_exists('taxonomy') && !empty($node->taxonomy) && is_array($node->taxonomy)) {
        foreach ($node->taxonomy as $term) {
          $original_term = $term;
          if ((object)$term) {
            // With free-tagging it's somewhat hard to get the tid, vid, name values
            // Rather than duplicating taxonomy.module code here you should make sure your calling module
            // has a weight of at least 1 which will run after taxonomy has saved the data which allows us to
            // pull it out of the db here.
            if (!isset($term->name) || !isset($term->tid)) {
              $vid = db_result(db_query_range("SELECT t.vid FROM {term_node} r INNER JOIN {term_data} t ON r.tid = t.tid INNER JOIN {vocabulary} v ON t.vid = v.vid WHERE r.nid = %d ORDER BY v.weight, t.weight, t.name", $object->nid, 0, 1));
              if (!$vid) {
                continue;
              }
              $term = db_fetch_object(db_query_range("SELECT t.tid, t.name FROM {term_data} t INNER JOIN {term_node} r ON r.tid = t.tid WHERE t.vid = %d AND r.vid = %d ORDER BY t.weight", $vid, $object->vid, 0, 1));
              $term->vid = $vid;
            }

            // Ok, if we still don't have a term name maybe this is a pre-taxonomy submit node
            // So if it's a number we can get data from it
            if (!isset($term->name) && is_array($original_term)) {
              $tid = array_shift($original_term);
              if (is_numeric($tid)) {
                $term = taxonomy_get_term($tid);
              }
            }
            
            $term_array = taxonomy_get_synonyms($term->tid);
            $values['term-synonym'] = $term_array[0];

            break;
          }
        }
      }
      // It's possible to leave that block and still not have good data.
      // So, we test for these and if not set, set them.
      if (!isset($values['term'])) {
        $values['term-synonym'] = '';
      }

      break;
  }

  return $values;
} 

Any ideas?

I figured it out. Here my code was right but I was clearing it out with this

if (!isset($values['term'])) {
        $values['term-synonym'] = '';
      }

Since I wasn't setting "term" in my custom module, it was just clearing my new token. Oops! Here's the final code.

/**
 * @file
 * Token Synonym - Returns the top synonym term for a node
 *
 * @ingroup token
 */

/**
 * Implementation of hook_token_list().
 */
function tokenSYNONYM_token_list($type = 'all') {
  $tokens = array();

  if ($type == 'node' || $type == 'all') {
  	if (module_exists('taxonomy')) {
    	$tokens['node']['term-synonym']  = t("The top taxonomy term synonym");
    }
  }

  return $tokens;
}

/**
 * Implementation of hook_token_values().
 */
function tokenSYNONYM_token_values($type, $object = NULL) {
  $values = array();
    switch ($type) {
      case 'node':
        $node = $object;
  
        // And now taxonomy, which is a bit more work. This code is adapted from
        // pathauto's handling code; it's intended for compatibility with it.
        if (module_exists('taxonomy') && !empty($node->taxonomy) && is_array($node->taxonomy)) {
          foreach ($node->taxonomy as $term) {
            $original_term = $term;
            if ((object)$term) {
              // With free-tagging it's somewhat hard to get the tid, vid, name values
              // Rather than duplicating taxonomy.module code here you should make sure your calling module
              // has a weight of at least 1 which will run after taxonomy has saved the data which allows us to
              // pull it out of the db here.
              if (!isset($term->name) || !isset($term->tid)) {
                $vid = db_result(db_query_range("SELECT t.vid FROM {term_node} r INNER JOIN {term_data} t ON r.tid = t.tid INNER JOIN {vocabulary} v ON t.vid = v.vid WHERE r.nid = %d ORDER BY v.weight, t.weight, t.name", $object->nid, 0, 1));
                if (!$vid) {
                  continue;
                }
                $term = db_fetch_object(db_query_range("SELECT t.tid, t.name FROM {term_data} t INNER JOIN {term_node} r ON r.tid = t.tid WHERE t.vid = %d AND r.vid = %d ORDER BY t.weight", $vid, $object->vid, 0, 1));
                $term->vid = $vid;
              }
  
              // Ok, if we still don't have a term name maybe this is a pre-taxonomy submit node
              // So if it's a number we can get data from it
              if (!isset($term->name) && is_array($original_term)) {
                $tid = array_shift($original_term);
                if (is_numeric($tid)) {
                  $term = taxonomy_get_term($tid);
                }
              }
              
              $term_array = taxonomy_get_synonyms($term->tid);
              $values['term-synonym'] = $term_array[0];
  
              // The 'catpath' (and 'cat') tokens have been removed, as they caused quite a bit of confusion,
              // and the catpath was a relatively expensive query when the taxonomy tree was deep.
              //
              // It existed only to provide forward-compatability with pathauto module, and
              // for most uses of token.module, it was a relatively useless token -- it exposed
              // a list of term names formatted as a URL/path string. Once pathauto supports
              // tokens, *it* should handle this catpath alias as it's the primary consumer.
              break;
            }
          }
        }
        // It's possible to leave that block and still not have good data.
        // So, we test for these and if not set, set them.
        if (!isset($values['term-synonym'])) {
          $values['term-synonym'] = '';
        }
  
        break;
    }
  
    return $values;
  }
  

I don't doubt that the code I wrote is more than it needs to be but that's primarily because I do not know how to write a patch and didn't want to edit the module itself. I had to pull all the code from the taxonomy tokens function and recreate it in my own module.

I would much rather see this added as a patch like you're attempting.

it does raise the question of what the token should contain when there is no synonym. Any thoughts?

What happens when you use a token that doesn't have a value for the more "core" tokens? It could mimic that which I imagine would just be the print [token] or whatever. I know in my case, if I am using that token it's because every term possible for that node has a synonym.

Re-uploading jaochim's patch with the syntax fix that Dave Reid noticed.

Oops. This should work

[edit] someone want to educate me on how to get the patch to test? I'm a patching noob.

No, thank you for taking a look at this issue. Although my code works it's quite excessive. Going to upgrade to the latest dev now with this new patch. Thanks again