ARDA Query Syntax Options and Information

Query terms are entered in the text entry field, separated by blanks.
Searches are generally case-insensitive, and it is strongly recommended that all query terms be entered in lowercase.
The only uppercase characters necessary are for specifying boolean operators which are explained further below.
By default, only items that contain matches against all of the query terms are returned.
For example, a query with terms god jesus returns items that contain both god and jesus.
See explanation of boolean terms below for options to override this default behavior.

Boolean terms AND, OR, and NOT can be used with query terms (Note - the keywords AND, OR, and NOT must be fully capitalized as shown).
For example, a query with god OR jesus returns items that contain either the term god or the term jesus.
Left and Right parens can be used to form sub-expressions when using boolean terms; for example, (god AND jesus) OR methodist

Phrases can be enclosed in double-quotes; for example, "god jesus" only returns items that contain the two word phrase "god jesus"
A term proximity value can be added to phrases (with a ~ followed by a proximity numeric value);
for example in the previous example, to permit matches where the terms are no more than 3 words apart, use "god jesus"~3.

Wildcards can specified in query terms.
A question mark character is a single-character wildcard; for example chil? would match items with child, chill, chili.
An asterisk character is a multi-character wildcard; for example g*d would match items with god, good, glad, etc.

A fuzzy search can be specified by adding a ~ character after the query term; fuzzy searches allow some differences in characters in determining matches.
For example, christ~ could match items with christ, christmas, christian, crisp, etc.
A proximty threshold value between 0.0 and 0.99 can be added after the ~ character (default value is 0.5).
The lower the proximity value, the more variance allowed in matching terms.
For example christ~0.4 could match items with christ, christmas, christian, crisp, and charisma.

Queries employing features such as booleans, wildcards, phrases, and fuzzy typically take several to many more seconds than simple term-only queries.

Click here for information on entering international characters

The search feature is implemented using the Apache SOLR open-source software, which uses the Lucene search engine.
The lexical analyzer ignores certain common English terms within the data (such as a, the, and, or, but, etc.),
and performs stemming of data to facilitate faster searching (e.g., may trim terms 'guiding' to 'guid' and 'guide' to 'guid').
This can result in unanticipated search results, for example a query with g*d might return guide as well as god, good, etc.