-
-
Couldn't load subscription status.
- Fork 113
Search Query Syntax
We can see a search query as a bunch of terms and operators, let's explore several ways to use our Lucene search engine:
The following fields (with examples) are searchable using the field-name:search-term syntax (without the field prefix, all fields are searched):
-
group-id-group-id:com.cemerick -
artifact-id-artifact-id:clj* -
description- the description pulled from the pom:description:useful -
licenses- a concatenation of any licenses listed in the pom:licenses:mit -
url- the homepage listed in the pom:url:gitlab.com -
at- the time the release was created. Times are best queried as ranges, either using the default Lucene syntax of specifying epoch times:at:[1515200516 TO 1515274516]or as ISO-8601 datetime strings (a Clojars extension to Lucene's search syntax):at:[2017-12-24T00:00:00Z TO 2018-01-01T00:00:00Z]
As a convenience, the search API will expand foo/bar to ((group-id:foo AND artifact-id:bar) OR "foo/bar". If you want to match foo/bar exactly, provide it as a quoted phrase ("foo/bar").
There are 2 types of terms:
- A
termis a single word such asoauthorsql - A
phraseis just a bunch of words grouped by quotes, like"sql builder"
Terms can be combined with operators in order to build more complex queries (note that these have to be uppercase):
-
ANDmatches both terms, i.e.google AND oauth -
NOTexcludes the term after NOT, i.e.auth NOT SNAPSHOT -
ORmatches if either of the terms is present, i.e.ldap OR saml
Parentheses can be used to form sub queries, i.e. (ldap OR saml) AND ADFS
-
?single character wildcard search for terms that matches with a single character replaced, i.e.oauth? -
*multiple character wildcard searches for 0 or more characters, i.e.oauth*
Do not use ? or * as the first character of a search
See also: Lucene's own Query Syntax Tutorial.
All clojars-web wiki content licensed under a Creative Commons Attribution-ShareAlike 4.0 International License
