Seperated detection of dismax and field search

Exact match: need to change full text field to String to be effective
This commit is contained in:
Kyle Huynh 2022-05-04 13:27:07 +00:00
parent 51028bdf48
commit 4d3a4df647
2 changed files with 74 additions and 48 deletions

View file

@ -152,79 +152,91 @@ class AdvancedSearchQuery {
// disable for Lucene and wildcard
//$q[] = "{!boost b=boost_document}";
// create a flag for active/inactive dismax
$isDismax = false;
// To support negative queries we must first bring in all documents.
$q[] = $this->negativeQuery($terms) ? "*:*" : "";
$term = array_shift($terms);
$q[] = $term->toSolrQuery($field_mapping);
// set dismax is enabled if the field set to "all"
if ($term->getField() === "all") {
$isDismax = true;
}
// for multiple conditions
foreach ($terms as $term) {
$q[] = $term->getConjunction();
$q[] = $term->toSolrQuery($field_mapping);
// set dismax is enabled if the field set to "all"
if ($term->getField() === "all") {
$isDismax = true;
}
}
$q = implode(' ', $q);
$case_insensitive_field = $this::getConfig(SettingsForm::SOLR_CASE_INSENSITIVE_FIELD_PREFIX, '');
// Limit extra processing if Luncene Search is enable
if ($isDismax) {
$case_insensitive_field = $this::getConfig(SettingsForm::SOLR_CASE_INSENSITIVE_FIELD_PREFIX, '');
/** @var Solarium\QueryType\Select\Query\Query $solarium_query */
if ((strpos($q, "*") !== false || strpos($q, "?") !== false)) {
// if the query string contain '*', '?', OR is a single world, enable wildcard
$tmp = str_replace('"', "", trim($q));
$query_fields = [];
foreach ($field_mapping as $key => $field) {
foreach ($field as $f => $item) {
// bs_ are boolean fields, do not work well with text search
if (substr($item, 0, 3) !== "bs_" && !in_array($item, ['score', 'random', 'boost_document'])
&& ((strpos( $item, "sm_" ) === 0) || (strpos( $item, "tm_" ) === 0) || (strpos($item, "sort_ss_") === 0) || (strpos($item, "ts_") === 0)
|| (strpos($item, "ss_") === 0)
)){
array_push($query_fields, '('.$item. ':'. $tmp .')');
/** @var Solarium\QueryType\Select\Query\Query $solarium_query */
if ((strpos($q, "*") !== false || strpos($q, "?") !== false)) {
// if the query string contain '*', '?', OR is a single world, enable wildcard
$tmp = str_replace('"', "", trim($q));
$query_fields = [];
foreach ($field_mapping as $key => $field) {
foreach ($field as $f => $item) {
// bs_ are boolean fields, do not work well with text search
if (substr($item, 0, 3) !== "bs_" && !in_array($item, ['score', 'random', 'boost_document'])
&& ((strpos( $item, "sm_" ) === 0) || (strpos( $item, "tm_" ) === 0) || (strpos($item, "sort_ss_") === 0) || (strpos($item, "ts_") === 0)
|| (strpos($item, "ss_") === 0)
)){
array_push($query_fields, '('.$item. ':'. $tmp .')');
// Add case insensitive fields in the query search fields
if ($case_insensitive_field !== '') {
$item_prefix = substr($item, 0, 3);
if ($item_prefix == "ss_" || $item_prefix == "sm_") {
$case_insensitive_item = str_replace($item_prefix,$case_insensitive_field, $item);
array_push($query_fields, '('.$case_insensitive_item. ':'. $tmp .')');
// Add case insensitive fields in the query search fields
if ($case_insensitive_field !== '') {
$item_prefix = substr($item, 0, 3);
if ($item_prefix == "ss_" || $item_prefix == "sm_") {
$case_insensitive_item = str_replace($item_prefix,$case_insensitive_field, $item);
array_push($query_fields, '('.$case_insensitive_item. ':'. $tmp .')');
}
}
}
}
}
}
$q = implode(" ", array_unique($query_fields));
}
$q = implode(" ", array_unique($query_fields));
}
else {
else {
// enable dismax search query option
/** @var Solarium\QueryType\Select\Query\Component\DisMax $dismax */
$dismax = $solarium_query->getDisMax();
$dismax->setQueryParser('dismax');
$query_fields = [];
foreach ($field_mapping as $key => $field) {
foreach ($field as $f => $item) {
// bs_ are boolean fields, do not work well with text search
if (substr($item, 0, 3) !== "bs_") {
array_push($query_fields, $item);
// enable dismax search query option
/** @var Solarium\QueryType\Select\Query\Component\DisMax $dismax */
$dismax = $solarium_query->getDisMax();
$dismax->setQueryParser('dismax');
$query_fields = [];
foreach ($field_mapping as $key => $field) {
foreach ($field as $f => $item) {
// bs_ are boolean fields, do not work well with text search
if (substr($item, 0, 3) !== "bs_") {
array_push($query_fields, $item);
if ($case_insensitive_field !== '') {
$item_prefix = substr($item, 0, 3);
if ($item_prefix == "ss_" || $item_prefix == "sm_") {
$case_insensitive_item = str_replace($item_prefix,$case_insensitive_field, $item);
array_push($query_fields, $case_insensitive_item);
if ($case_insensitive_field !== '') {
$item_prefix = substr($item, 0, 3);
if ($item_prefix == "ss_" || $item_prefix == "sm_") {
$case_insensitive_item = str_replace($item_prefix,$case_insensitive_field, $item);
array_push($query_fields, $case_insensitive_item);
}
}
}
}
}
$query_fields = implode(" ", array_unique($query_fields));
$dismax->setQueryFields($query_fields);
}
$query_fields = implode(" ", array_unique($query_fields));
$dismax->setQueryFields($query_fields);
// Looks like, somewhere in the previous steps, double quotes get added, we are removing them!
// ToDo: Look into where the original quotes are being added, and remove it there!
$q = trim($q);
$q = substr($q, 1);
$q = substr($q, 0, -1);
}
$solarium_query->setQuery($q);
}
}

View file

@ -284,6 +284,13 @@ class AdvancedSearchQueryTerm {
$terms = [];
$query_helper = \Drupal::service('solarium.query_helper');
$value = $query_helper->escapePhrase(trim($this->value));
// Added to handle exact matches keyword (surrounded by "")
if (preg_match('#^(\'|").+\1$#', $value) == 1) {
trim($value);
$value = str_replace('"\\', '', $value);
$value = str_replace('\\"', '', $value);
}
if ($this->field === "all") {
return $value;
}
@ -295,5 +302,12 @@ class AdvancedSearchQueryTerm {
$terms = implode(' ', $terms);
return $this->include ? "($terms)" : "-($terms)";
}
/**
* Get Field search
*/
public function getField() {
return $this->field;
}
}