Keywords

The Keywords filter lets you filter entries by search terms (ie. keywords). You need to create at least one Collection to use the Keywords filter. It also adds a relevance score to the search results. To further fine-tune a keyword search, the following parameters are available.

Parameters

collection Limit search results to given collection names or IDs. Will use all collections if not specified and keywords are given. collection_lang Limit search results to collections with the given language(s). keywords The search terms to filter by. You can use double quotes to group terms, eg. lion "mighty jungle" keywords:inflect Set to yes to enable singular & plural matching for the keywords, based on the language given in keywords:lang. Eg. searching for lions will also match lion. keywords:lang The language of the given keywords, eg. en. Adding this will enable keyword inflections and stemming, as well as keyword suggestions. keywords:loose`` Enables partial or substring matches:

• right: lion will match lions and lionized.
• left: lion will match dandelion and medallion.
• both: lion will match stallions and millionaire.
• no: lion will only match lion. This is the default.

Note: keywords:mode needs to be set to any or all.

keywords:match If a valid field (native or custom) is given, entries where that field exactly matches the keywords given will be pushed to the top of the search results, regardless of score. Eg. title keywords:mode How to treat the given keywords:

• any: returns entries containing any of the given terms.
• all: returns entries containing all of the given terms.
• exact: returns entries containing the given terms as an exact phrase.
• auto: uses operators to process the given terms. This is the default. keywords:score Limit the search results to entries with a score greater than the given number. Eg. >=1 or 2. keywords:stem Set to yes to enable matching keywords by their stem, based on the language given in keywords:lang. Eg. searching for sleeping will also match words starting with sleep, like sleeps and sleepy.

Note: Setting keywords:loose to left or both will result in a slower query than normal. It is recommended to only use this option with languages that do not have word delimiters, like Thai or Japanese.

The keywords filter also makes these variables available in the Results tag:

{low_search_collection_id} Collection ID for the collection the entry was found in. {low_search_collection_label} Collection label for the collection the entry was found in. {low_search_collection_language} Collection language for the collection the entry was found in. {low_search_collection_name} Collection name for the collection the entry was found in. {low_search_score} Relevance score of the entry.

Order by collection

By default, search results will be ordered by relevance score:

low_search_score. Use orderby="low_search_collection:foo,bar" to order by collection first, where foo and bar are collection names. Search results that belong to other collections than defined here will be grouped together and shown last.

keywords:mode="auto"

The automatic keywords:mode uses operators in keywords for any/all/exact matching, excluding terms, and partial matching. The keywords:loose parameter is ignored when using this mode.

jungle lion Entries containing both “jungle” and “lion”. jungle OR lion Entries containing either “jungle” or “lion” or both. jungle -lion Entries containing “jungle” but not “lion”. "jungle lion" Entries containing the exact phrase “jungle lion”. jun* Entries containing a word that starts with “jun”. *ion Entries containing a word that ends with “ion”. *ung* Entries containing a word that contains “ung”.

Singulars & Plurals

Low Search 4 supports singular and plural matching of keywords (inflections). To enable this, both the keywords:inflect and the keywords:lang parameters must be set. Low Search supports English inflections natively and you can add support for other languages by adding inflection rules to your Config file:

$config['low_search_inflection_rules'][lang] = array( 'plural' => array(), 'singular' => array(), 'irregular' => array(), 'uncountable' => array() );

lang is the 2-letter language code. The plural and singular keys should contain an array of key/value pairs to be used in a preg_replace() call. The irregular key should contain an array of irregular singulars (keys) and plurals (values). The uncountable key should contain a flat array of uncountable terms. Here are Dutch inflection rules as an example. Stems#

Low Search 4 supports matching of keywords by their stem (stemming). To enable this, both the keywords:stem and the keywords:lang parameters must be set. English stemming is supported natively, using a Porter stemmer class, and you can add support for other languages by adding this to your Config file:

$config['low_search_stemmers'][lang] = array( file_path, class_name, method );

lang is the 2-letter language code. file_path is the full path to your the stemmer file, class_name is the class name the file contains, method is the method name that should be called to return the stem of a given word. Here’s a Dutch stemmer class that you could use.

Note: When enabled, inflections and stems are only applied keyword searches that do not contain wildcards or have keywords:loose set to left, right or both. Categories#

You can use the native category parameter to filter by category. For more advanced filtering by category, you can also divide categories into groups. The group syntax lets you combine AND and OR filtering (category 1 or 2 and category 3 or 4), as parameters are always combined with AND.

category Like the native category= parameter, but it also accepts category URL titles. category:group_name Works just like the category parameter. When using category URL titles as the value, you can use the category group ID to limit the conversion to category IDs to that specific group, eg. category:5

Note: Use category IDs instead of URL titles for better performance.

Distance

You can use the Distance filter to limit results by a given maximum distance. This filter requires you to use two channel fields where latitude and longitude values are stored. Alternatively, you can use a single field where the two values are separated by a comma. Using the Distance filter will limit results to entries that actually have latitude and longitude values entered.

Parameters

distance:from Latitude and longitude values separated by a vertical bar, used to calculate the distance. Eg. 52.163298|4.505547 distance:to The one or two channel field names that contain the latitude and longitude values separated by a vertical bar, used to calculate the distance. Eg. cf_entry_lat|cf_entry_long distance:radius The maximum distance between the from and to values. Leave blank for no maximum. Eg. 50. distance:unit The unit for the distances, either km, mi, m or yd. Defaults to km.

Note: Use two separate fields instead of a single one for better performance.

The Distance filter also makes this variable available in the Results tag:

{low_search_distance} The calculated distance in the given unit for this entry.

Note: Using the Distance filter will return the search results ordered by distance, ignoring keyword relevance if applicable. Override by explicitly setting the orderby parameter. Field Search#

You can use the native search:field_name parameter to target specific fields. Additionally, Low Search can target the entry’s title, url_title, status, target Grid / Matrix columns, use multiple values for numeric matching (in combination with the gt, gte, lt and lte params), and use starts / ends with matching.

Parameters

search:field_name Like the native search:field_name= parameter. search:field_name:column_name Works just like the search:field_name parameter, but targets a specific column in a Grid/Matrix field. search:title Works just like the search:field_name parameter, but targets titles. search:url_title Works just like the search:field_name parameter, but targets url_titles. search:status Works just like the search:field_name parameter, but targets statuses. contains_words Accepts parameter names. Force the given parameter values to match the full term, to ensure that the values are not contained within other words. Also possible by using search:field_name="value\W". ends_with Accepts parameter names. Force the given parameter values to match the end of the target field. Also possible by using search:field_name="value$". exact Accepts parameter names. Force the given parameter values to exactly match the target field. Also possible by using search:field_name="=value". gt Accepts parameter names. Force the given parameter values to have greater than prepended to it. Also possible by using search:field_name=">value". gte Accepts parameter names. Force the given parameter values to have greater than or equal to prepended to it. Also possible by using search:field_name=">=value". lt Accepts parameter names. Force the given parameter values to have less than prepended to it. Also possible by using search:field_name="value". lte Accepts parameter names. Force the given parameter values to have less than or equal to prepended to it. Also possible by using search:field_name="<=value". starts_with Accepts parameter names. Force the given parameter values to match the beginning of the target field. Also possible by using search:field_name="^value". smart_field_search Set to yes to make search:field_name parameters that target non-Grid/Matrix custom channel fields aware of the channels they belong to. Use this if your search results contain multiple channels.

Note: using smart_field_search="yes" can affect performance, depending on the total amount of entries. Using Collections and collection="" parameter can help in such cases.

Ranges

You can use the Ranges filter to limit results by a given range, targeting a numeric or date field. If Low Search detects that the given channel field for the range is a date field, it will try and convert the given range values to timestamps. This will allow custom date ranges as well.

Parameters

range:field_name Takes a from and to value, separated by a vertical bar: |. Use field_name:column_name to target Grid/Matrix columns (v4.2.0+). range-from:field_name Takes a single from value. Use field_name:column_name to target Grid/Matrix columns (v4.2.0+). range-to:field_name Takes a single to value. Use field_name:column_name to target Grid/Matrix columns (v4.2.0+). range:min_field:max_field Where min and max are two separate (non-Grid/Matrix) fields. If a single value is given, entries will be returned where the value is between the min and max fields. If a from and to value is given, entries will be returned where the min and max fields overlap the given range. range-from:min_field:max_field Where min and max are two separate fields. Takes a single from value. range-to:min_field:max_field Where min and max are two separate fields. Takes a single to value. exclude`` Accepts parameter names. Excludes a given parameter value from the range itself. For example, range:field_name="0|10" exclude="range:field_name" will result in values > 0 and < 10 rather than >= 0 and <= 10.

Note: For numeric fields, make sure the Field Content option in the field’s settings is set to Number, Integer or Decimal.

Apart from any custom numeric or date field, the following stardard channel fields are supported:

entry_date
edit_date
expiration_date
comment_expiration_date
recent_comment_date
view_count_one
view_count_two
view_count_thee
view_count_four
comment_total

Relationships

You can use the Relationships filter to limit results by given parent or child entry IDs. The filter works for both native Relationships and Playa.

Parameters

child:field_name Limit results by entries that have the given entry IDs or URL titles as a child for the Relationships/Playa field defined. parent:field_name Limit results by entries that have the given entry IDs or URL titles as a parent for the Relationships/Playa field defined. child:field_name:column_name Limit results by entries that have the given entry IDs or URL titles as a child for the Relationships/Playa column defined in a Grid field. parent:field_name:column_name Limit results by entries that have the given entry IDs or URL titles as a parent for the Relationships/Playa column defined in a Grid field.

Note: Use entry IDs instead of URL titles for better performance.

Tags

You can use the Tags filter to limit results by given tag names or IDs. The filter works for both Solspace Tag and DevDemon Tagger. For more advanced filtering by tags, you can also divide tags into groups. The group syntax lets you combine AND and OR filtering (tag 1 or 2 and tag 3 or 4), as parameters are always combined with AND.

Parameters

tag_id Limit results by entries that have the given tag IDs assigned to them. tag_name Limit results by entries that have the given tag names assigned to them. tag_id:group_name Works just like the tag_id parameter. tag_name:group_name Works just like the tag_name parameter. websafe_separator The websafe separator character for multi-word tags. Defaults to +.

Note: Use tag_id instead of tag_name for better performance.