Created
February 19, 2015 20:30
-
-
Save grantr/d60bc01afaaa41657e31 to your computer and use it in GitHub Desktop.
Elasticsearch query parser experiments
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
module Elastomer | |
module Query | |
class ElasticsearchQueryParser < LuceneQueryParser | |
rule(:unbounded_range) { gte | lte | gt | lt } | |
#TODO range groups e.g. age:(>=10 AND < 20) age:(+>=10 +<20) | |
#TODO eliminate duplication | |
rule(:field) { word.as(:field) >> str(':') >> (((group | regexp | inclusive_range.as(:inclusive_range) | exclusive_range.as(:exclusive_range)) >> boost.maybe) | ((phrase | term) >> (distance | boost).maybe)) } | |
end | |
end | |
end |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
module Elastomer | |
module Query | |
class LuceneQueryParser < Parslet::Parser | |
rule(:space) { match('\s').repeat(1) } | |
rule(:word) { match('\w').repeat(1) } | |
rule(:digit) { match('[0-9]') } | |
rule(:integer) { digit.repeat(1) } | |
rule(:float) { integer >> str('.') >> integer } | |
rule(:reserved) { match('[+\-!(){}[]^"~*?:\\/]') | string('&&') | string('||') } | |
rule(:wildcard) { str('*') } | |
rule(:operator) { (str('AND') | str('OR')).as(:operator) } | |
rule(:distance) { str('~') >> digit.as(:distance) } | |
rule(:boost) { str('^') >> (integer | float).as(:boost) } | |
#TODO disallow unmatched quotes | |
rule(:term) { (match('[^?*]') >> match('[\w?*]').repeat).as(:term) >> (distance | boost).maybe } | |
#TODO allow escaped quotes | |
rule(:phrase) { str('"') >> match('[^"]').repeat(1).as(:phrase) >> str('"') >> (distance | boost).maybe } | |
rule(:group) { str('(') >> space.maybe >> query.as(:group) >> space.maybe >> str(')') >> boost.maybe } | |
rule(:gt) { str('>') >> space.maybe >> word.as(:gt) } | |
rule(:lt) { str('<') >> space.maybe >> word.as(:lt) } | |
rule(:gte) { str('>=') >> space.maybe >> word.as(:gte) } | |
rule(:lte) { str('<=') >> space.maybe >> word.as(:lte) } | |
rule(:bounded_range) { (word | wildcard).as(:from) >> space >> str('TO') >> space >> (word | wildcard).as(:to) } | |
rule(:inclusive_range) { str('[') >> space.maybe >> bounded_range >> space.maybe >> str(']') } | |
rule(:exclusive_range) { str('{') >> space.maybe >> bounded_range >> space.maybe >> str('}') } | |
#TODO allow escaped / | |
rule(:regexp) { str('/') >> match('[^/]').repeat(1).as(:regexp) >> str('/') >> boost.maybe} | |
rule(:unary_operator) { str('+').as(:required) | str('-').as(:prohibited) | (str('NOT').as(:prohibited) >> space) } | |
rule(:field) { word.as(:field) >> str(':') >> (((group | regexp | inclusive_range.as(:inclusive_range) | exclusive_range.as(:exclusive_range)) >> boost.maybe) | ((phrase | term) >> (distance | boost).maybe)) } | |
rule(:clause) { unary_operator.maybe >> (group | field | phrase | regexp | term) } | |
rule(:query) { space.maybe >> clause >> (space >> (operator >> space >> clause | clause)).repeat >> space.maybe } | |
root :query | |
end | |
end | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment