如何解决我的GADT中的歧义

我有两个用于模拟SQL EDSL的GADT。为了使面向API的客户端保持整洁和简单，我想使用OverloadedStrings将字符串文字转换为Column Selection。

因此，您只需键入

select ["a", "b"] $ from tbl

代替

select [Column "a", Column "b"] $ from tbl

问题是select允许Column Selections和Reductions都允许执行聚合的查询。

mean :: Column Selection -> Column Reduction

select :: [Column a] -> Query b -> Query Selection
select [mean "a"] $ from tbl

因此在这种情况下字符串是不明确的[Column a]。但是select [mean "a"] $ from tbl有效，因为mean它提供了必要的上下文来推断字符串文字是“列选择”。

谁能推荐摆脱这种混乱的出路？

我下面的当前代码（忽略了无关的实例）

{-# LANGUAGE 
    GADTs
  , RankNTypes
  , DataKinds
  , TypeFamilies
  , FlexibleContexts
  , FlexibleInstances
  , OverloadedStrings #-}

data Sz = Selection | Reduction deriving Show
data Schema = Schema{name :: Maybe String, spec :: [Column Selection]} 

type family ColOp (a :: Sz) (b :: Sz) where
    ColOp Selection Selection = Selection
    ColOp Selection Reduction = Selection
    ColOp Reduction Selection = Selection
    ColOp Reduction Reduction = Reduction

data Column (a :: Sz) where
    Column  :: String -> Column Selection
    Assign  :: String -> Column a -> Column a
    FKey    :: String -> Schema -> Column Selection
    BinExpr :: BinOp  -> Column a -> Column b -> Column (ColOp a b)
    LogExpr :: LogOp  -> Column a -> Column b -> Column Selection
    AggExpr :: AggOp  -> Column Selection -> Column Reduction

instance IsString (Column Selection) where
    fromString s = Column s

data Query (a :: Sz) where
    Table  :: Schema -> Query Selection
    Select :: [Column a] -> Query b -> Query Selection
    Update :: [Column a] -> Query b -> Query Selection
    Where  :: [Column Selection] -> Query Selection -> Query Selection
    Group  :: [Column Selection] -> Query Selection -> Query Reduction

我也想提出以下签名失败Select/ Update：

[Column Selection] -> Query Reduction -> Query Selection

但这就是蠕虫的另一罐……