优化器直方图

chusql-core/model/src/ChuSQL/Core/Model.hs

@@ -10,8 +10,11 @@ module ChuSQL.Core.Model
     , Value (..)
     , Row
     , TableMeta (..)
+    , Histogram (..)
     , Table (..)
     , Database
+    , histogramOf
+    , histogramFor
     , colNames
     , colType
     , lookupTable
@@ -42,7 +45,7 @@ module ChuSQL.Core.Model
 import Data.Char (isDigit, isSpace, toLower)
 import Data.Hashable (Hashable (..))
 import Data.List (isSuffixOf, stripPrefix)
-import Data.Maybe (fromMaybe)
+import Data.Maybe (fromMaybe, mapMaybe)
 
 -- 数据模型：列类型、列约束、值、行、表、数据库，外加查表小工具。
 
@@ -441,10 +444,19 @@ instance Hashable Value where
 -- | 一行数据
 type Row = [(String, Value)]
 
--- | 一张表的统计：行数、各列不同值数（capped 表示只数到上限）、有索引的列
+-- | 一列的等宽直方图：区间下界、上界与每桶计数
+data Histogram = Histogram
+    { histLow :: Double
+    , histHigh :: Double
+    , histBuckets :: [Int]
+    }
+    deriving (Show, Eq)
+
+-- | 一张表的统计：行数、各列不同值数、列直方图与索引
 data TableMeta = TableMeta
     { metaRowCount :: Int
     , metaDistinct :: [(String, Int, Bool)]
+    , metaHistograms :: [(String, Histogram)]
     , metaIndexes :: [String]
     }
     deriving (Show)
@@ -461,6 +473,36 @@ data Table = Table
 -- | 库：库名到表的映射
 type Database = [(String, Table)]
 
+-- | 这一列的直方图
+histogramFor :: TableMeta -> String -> Maybe Histogram
+histogramFor m col = lookup col (metaHistograms m)
+
+-- | 数值列按 16 个等宽桶统计，无数值给 Nothing
+histogramOf :: [Value] -> Maybe Histogram
+histogramOf values = case mapMaybe numericOf values of
+    [] -> Nothing
+    vs -> let lo = minimum vs; hi = maximum vs in Just (Histogram lo hi (bucketCounts lo hi vs))
+
+-- | 值转成用于分桶的数；非数值给 Nothing
+numericOf :: Value -> Maybe Double
+numericOf (VInt n) = Just (fromIntegral n)
+numericOf (VFloat d) = Just d
+numericOf _ = Nothing
+
+-- | 每个桶里落了多少个值
+bucketCounts :: Double -> Double -> [Double] -> [Int]
+bucketCounts lo hi vs = [length [v | v <- vs, bucketIndexOf lo hi v == i] | i <- [0 .. bucketTotal - 1]]
+
+-- | 值落在第几个桶；区间退化时都算第 0 桶
+bucketIndexOf :: Double -> Double -> Double -> Int
+bucketIndexOf lo hi v
+    | hi <= lo = 0
+    | otherwise = max 0 (min (bucketTotal - 1) (floor ((v - lo) / (hi - lo) * fromIntegral bucketTotal) :: Int))
+
+-- | 直方图的桶数
+bucketTotal :: Int
+bucketTotal = 16
+
 -- | 取表的列名
 colNames :: Table -> [String]
 colNames = map fst . tableCols

chusql-core/model/src/ChuSQL/Core/Protocol.hs

@@ -15,7 +15,7 @@ module ChuSQL.Core.Protocol (
     queryResultJson,
 ) where
 
-import ChuSQL.Core.Model (Row, Value (..))
+import ChuSQL.Core.Model (Histogram (..), Row, Value (..))
 import Data.Aeson (FromJSON (..), ToJSON (..), object, withObject, (.:), (.:?), (.!=), (.=))
 import qualified Data.Aeson as A
 import qualified Data.Aeson.Key as K
@@ -244,6 +244,7 @@ data TableInfo = TableInfo
     , tiRows :: Int
     , tiIndexes :: [String]
     , tiStats :: [(String, Int, Bool)]
+    , tiHistograms :: [(String, Histogram)]
     }
     deriving (Show, Eq)
 
@@ -255,11 +256,21 @@ instance ToJSON TableInfo where
             , "columns" .= tiColumns info
             , "row_count" .= tiRows info
             , "indexes" .= map (\name -> object ["column" .= name]) (tiIndexes info)
-            , "stats" .= map statJson (tiStats info)
+            , "stats" .= map (statJson (tiHistograms info)) (tiStats info)
             ]
       where
-        statJson (name, distinct, capped) =
-            object ["name" .= name, "distinct" .= distinct, "capped" .= capped]
+        statJson hists (name, distinct, capped) =
+            object (statFields ++ histFields (lookup name hists))
+          where
+            statFields = ["name" .= name, "distinct" .= distinct, "capped" .= capped]
+            histFields Nothing = []
+            histFields (Just h) = ["lo" .= histLow h, "hi" .= histHigh h, "hist" .= histBuckets h]
+
+-- | 线上直方图还原；区间或桶缺了就当作没有
+histogramFromWire :: Maybe Double -> Maybe Double -> [Int] -> Maybe Histogram
+histogramFromWire (Just low) (Just high) counts
+    | not (null counts) = Just (Histogram low high counts)
+histogramFromWire _ _ _ = Nothing
 
 -- | JSON 解回表信息
 instance FromJSON TableInfo where
@@ -270,14 +281,19 @@ instance FromJSON TableInfo where
         indexes <- o .:? "indexes" .!= []
         stats <- o .:? "stats" .!= []
         indexColumns <- mapM (\v -> withObject "IndexWire" (.: "column") v) (indexes :: [A.Value])
-        parsedStats <- mapM parseStat (stats :: [A.Value])
-        pure (TableInfo table columns count indexColumns parsedStats)
+        parsed <- mapM parseStat (stats :: [A.Value])
+        let parsedStats = [(name, distinct, capped) | (name, distinct, capped, _) <- parsed]
+            parsedHistograms = [(name, h) | (name, _, _, Just h) <- parsed]
+        pure (TableInfo table columns count indexColumns parsedStats parsedHistograms)
       where
         parseStat = withObject "ColumnStat" $ \o -> do
             name <- o .: "name"
             distinct <- o .: "distinct"
             capped <- o .:? "capped" .!= False
-            pure (name, distinct, capped)
+            low <- o .:? "lo"
+            high <- o .:? "hi"
+            counts <- o .:? "hist" .!= []
+            pure (name, distinct, capped, histogramFromWire low high counts)
 
 -- | 一条账号记录
 data Account = Account