Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Autosuggest: Major rewrite of adjust_fuzzy_fields #2347

Merged
merged 6 commits into from
Sep 21, 2021
Merged

Autosuggest: Major rewrite of adjust_fuzzy_fields #2347

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
a7814f9
Autosuggest: Major rewrite of adjust_fuzzy_fields
felipeelia Sep 14, 2021
c16d07e
Autosuggest: restrict the addition of term_suggest
felipeelia Sep 14, 2021
54b55e9
Escape dots in regex
felipeelia Sep 14, 2021
331cc83
Merge branch 'develop' into fix/issue-2346
Rahmon Sep 14, 2021
90cd6ca
Merge branch 'develop' into fix/issue-2346
felipeelia Sep 20, 2021
fcdb21f
Merge branch 'develop' into fix/issue-2346
felipeelia Sep 21, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
148 changes: 104 additions & 44 deletions includes/classes/Feature/Autosuggest/Autosuggest.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -237,57 +237,117 @@ public function set_fuzziness( $fuzziness, $search_fields, $args ) {
* @return array $query adjusted ES Query arguments
*/
public function adjust_fuzzy_fields( $query, $post_type, $args ) {
if ( Utils\is_integrated_request( $this->slug, [ 'public' ] ) && ! empty( $args['s'] ) ) {
/**
* Filter autosuggest ngram fields
*
* @hook ep_autosuggest_ngram_fields
* @param {array} $fields Fields available to ngram
* @return {array} New fields array
*/
$ngram_fields = apply_filters(
'ep_autosuggest_ngram_fields',
[
'post_title' => 'post_title.suggest',
]
);
if ( ! Utils\is_integrated_request( $this->slug, [ 'public' ] ) || empty( $args['s'] ) ) {
return $query;
}

if ( isset( $query['bool'] ) && isset( $query['bool']['must'] ) ) {
foreach ( $query['bool']['must'] as $q_index => $must_query ) {
if ( isset( $must_query['bool'] ) && isset( $must_query['bool']['should'] ) ) {
foreach ( $must_query['bool']['should'] as $index => $current_bool_should ) {
if (
isset( $current_bool_should['multi_match'] ) &&
isset( $current_bool_should['multi_match']['fields'] ) &&
(
(
isset( $current_bool_should['multi_match']['fuzziness'] ) &&
0 !== $current_bool_should['multi_match']['fuzziness']
) ||
(
isset( $current_bool_should['multi_match']['slop'] ) &&
0 !== $current_bool_should['multi_match']['slop']
)
)
) {
foreach ( $current_bool_should['multi_match']['fields'] as $key => $field ) {
foreach ( $ngram_fields as $plain_field => $ngram_field ) {
if ( preg_match( '/^(' . $plain_field . ')(\^(\d+))?$/', $field, $match ) ) {
if ( isset( $match[3] ) && $match[3] > 1 ) {
$weight = $match[3] - 1;
} else {
$weight = 1;
}
$query['bool']['must'][ $q_index ]['bool']['should'][ $index ]['multi_match']['fields'][] = $ngram_field . '^' . $weight;
}
}
}
if ( ! isset( $query['bool'] ) || ! isset( $query['bool']['must'] ) ) {
return $query;
}

/**
* Filter autosuggest ngram fields
*
* @hook ep_autosuggest_ngram_fields
* @param {array} $fields Fields available to ngram
* @return {array} New fields array
*/
$ngram_fields = apply_filters(
'ep_autosuggest_ngram_fields',
[
'post_title' => 'post_title.suggest',
'terms\.(.+)\.name' => 'term_suggest',
]
);

/**
* At this point, `$query` might look like this (using the 3.5 search algorithm):
*
* [
* [bool] => [
* [must] => [
* [0] => [
* [bool] => [
* [should] => [
* [0] => [
* [multi_match] => [
* [query] => ep_autosuggest_placeholder
* [type] => phrase
* [fields] => [
* [0] => post_title^1
* ...
* [n] => terms.category.name^27
* ]
* [boost] => 3
* ]
* ]
* [1] => [
* [multi_match] => [
* [query] => ep_autosuggest_placeholder
* [fields] => [ ... ]
* [type] => phrase
* [slop] => 5
* ]
* ]
* ]
* ]
* ]
* ]
* ]
* ...
* ]
*
* Also, note the usage of `&$must_query`. This means that by changing `$must_query`
* you will be actually changing `$query`.
*/
foreach ( $query['bool']['must'] as &$must_query ) {
if ( ! isset( $must_query['bool'] ) || ! isset( $must_query['bool']['should'] ) ) {
continue;
}
foreach ( $must_query['bool']['should'] as &$current_bool_should ) {
if ( ! isset( $current_bool_should['multi_match'] ) || ! isset( $current_bool_should['multi_match']['fields'] ) ) {
continue;
}

/**
* `fuzziness` is used in the original algorithm.
* `slop` is used in `3.5`.
*
* @see \ElasticPress\Indexable\Post\Post::format_args()
*/
if ( empty( $current_bool_should['multi_match']['fuzziness'] ) && empty( $current_bool_should['multi_match']['slop'] ) ) {
continue;
}

$fields_to_add = [];

/**
* If the regex used in `$ngram_fields` matches more than one field,
* like taxonomies, for example, we use the min value - 1.
*/
foreach ( $current_bool_should['multi_match']['fields'] as $field ) {
foreach ( $ngram_fields as $regex => $ngram_field ) {
if ( preg_match( '/^(' . $regex . ')(\^(\d+))?$/', $field, $match ) ) {
$weight = 1;
if ( isset( $match[4] ) && $match[4] > 1 ) {
$weight = $match[4] - 1;
}

if ( isset( $fields_to_add[ $ngram_field ] ) ) {
$fields_to_add[ $ngram_field ] = min( $fields_to_add[ $ngram_field ], $weight );
} else {
$fields_to_add[ $ngram_field ] = $weight;
}
}
}
}

foreach ( $fields_to_add as $field => $weight ) {
$current_bool_should['multi_match']['fields'][] = "{$field}^{$weight}";
}
}
}

return $query;
}

Expand Down