数据日志中的参数化和不区分大小写的查询（数据）

Question

我想编写一个以名字作为输入参数和 returns 所有匹配记录的查询。匹配应该不区分大小写。例如，我想提取所有名为 Douglass 的人。参数化，但区分大小写，这将是：

(d/q '[:find (pull ?e [*])
       :in $ ?par
       :where
       [?e :person/firstname ?par]
       ] db "Douglass")

无论大小写如何，以下查询都会产生所有匹配项，但未进行参数化（尚未使用 ?par，而是参数化查询的占位符）：

(d/q '[:find (pull ?e [*])
       :in $ ?par
       :where
       [?e :person/firstname ?bfn]
       [(re-find (re-pattern "(?i)DouGLASS") ?bfn)]
       ] db "")

但我无法将它们组合起来。一个 - 可能是天真的 - 方法是抛出 Unable to resolve symbol: ?par in this context :

(d/q '[:find (pull ?e [*])
       :in $ ?par
       :where
       [?e :person/firstname ?bfn]
       [ (re-find (re-pattern (str "(?i)" ?par)) ?bfn)]
       ] db "Douglass")

那么：如何为这种情况传递名字？

Answer 1

如 here, the issue is that function expressions in Datomic Datalog do not nest. You could break it out like this (using a tested query against the mbrainz 数据库所述）。

(d/q '[:find ?name ?year
       :in $ ?match
       :where [(str "(?i)" ?match) ?matcher]
              [(re-pattern ?matcher) ?regex]
              [(re-find ?regex ?aname)]
              [?a :artist/name ?aname]
              [?r :release/artists ?a]
              [?r :release/name ?name]
              [?r :release/year ?year]]
     (d/db conn) "pink floyd")

Returns:

#{["Point Me at the Sky" 1972]
  ["Any Colour You Like" 1972]
  ["The Dark Side of the Moon" 1973]
  ["Obscured by Clouds" 1972]
  ...

您也可以编写自己的函数并从 Datalog 中调用它。

数据日志中的参数化和不区分大小写的查询（数据）

Parameterized and case insensitive query in datalog (datomic)

datalog

datomic