组合 SPARQL 查询

Question

我想获取人口最多和最少的州的人口。我知道如何使用ORDER BY(ASC和DESC)。如何将这两个 (ASC和DESC) 组合在一个查询中？

SELECT  ?population
{
?state rdf:type :State
?state :hasPopulation ?population.
} ORDER BY DESC(?population) LIMIT 1
AND

SELECT  ?population
{
?state rdf:type :State
?state :hasPopulation ?population.
} ORDER BY ASC(?population) LIMIT 1

Answer 1

SELECT ?population_max ?population_min {
    ?state_max rdf:type :State .
    ?state_max :hasPopulation ?population_max .
    ?state_min rdf:type :State .
    ?state_min :hasPopulation ?population_min .
} ORDER BY DESC(?population_max) ASC(?population_min) LIMIT 1

也许更有效：

SELECT * {
    {
        SELECT (?population AS ?population_max) {
            ?state rdf:type :State .
            ?state :hasPopulation ?population .
        } ORDER BY DESC(?population) LIMIT 1 
    }
    {
        SELECT (?population AS ?population_min) {
            ?state rdf:type :State .
            ?state :hasPopulation ?population .
        } ORDER BY ASC(?population) LIMIT 1
    }
}

使用 AnzoGraph 或 Blazegraph，在这种情况下可以使用命名 子查询：

SELECT *
WITH {
    SELECT ?pop { [] a :State ; :hasPopulation ?pop }
} AS %unsorted
WHERE {
    { SELECT (?pop AS ?max) { INCLUDE %unsorted } ORDER BY DESC(?pop) LIMIT 1 }
    { SELECT (?pop AS ?min) { INCLUDE %unsorted } ORDER BY  ASC(?pop) LIMIT 1 }
}

最后，而不是重写而不是组合：

SELECT (MAX(?population) AS ?population_max) (MIN(?population) AS ?population_min) {
    ?state rdf:type :State .
    ?state :hasPopulation ?population
} 

如果使用聚合...但未GROUP BY使用该术语，则将其视为所有解决方案所属的单个隐式组。