dataframe-th 2.3.0.0 → 2.3.1.0

raw patch · 3 files changed

+79/−4 lines, 3 filesdep −dataframe-operationsdep ~dataframe-core

Dependencies removed: dataframe-operations

Dependency ranges changed: dataframe-core

Files

• dataframe-th.cabal +3/−3
• src/DataFrame/TH/Identifier.hs +75/−0
• src/DataFrame/TH/Records.hs +1/−1

@@ -1,6 +1,6 @@ cabal-version:      3.4 name:               dataframe-th-version:            2.3.0.0+version:            2.3.1.0 synopsis:           Record-based Template Haskell splices for the dataframe ecosystem. description:     The IO-agnostic part of the @dataframe@ TH machinery — splices that@@ -27,6 +27,7 @@ library     import:             warnings     exposed-modules:+                        DataFrame.TH.Identifier                         DataFrame.TH.Records                         DataFrame.Typed.TH.Records     -- Internal record→schema TH; sealed (superseded publicly by@@ -34,8 +35,7 @@     other-modules:      DataFrame.Internal.Schema.TH     build-depends:      base >= 4 && < 5,                         containers >= 0.6.7 && < 0.10,-                        dataframe-core >= 2.6 && < 2.7,-                        dataframe-operations >= 2.6 && < 2.7,+                        dataframe-core >= 2.7 && < 2.8,                         dataframe-parsing >= 2.2 && < 2.3,                         template-haskell >= 2.0 && < 3,                         text >= 2.1 && < 3,

@@ -0,0 +1,75 @@+{-# LANGUAGE OverloadedStrings #-}++-- | Turning column names into valid Haskell identifiers for generated code.+module DataFrame.TH.Identifier (+    sanitize,+    isHaskellIdentifier,+    isReservedId,+    isVarId,+) where++import qualified Data.Char as Char+import qualified Data.Text as T++-- | Haskell Report 2.4 reserved words.+isReservedId :: T.Text -> Bool+isReservedId t = case t of+    "case" -> True+    "class" -> True+    "data" -> True+    "default" -> True+    "deriving" -> True+    "do" -> True+    "else" -> True+    "foreign" -> True+    "if" -> True+    "import" -> True+    "in" -> True+    "infix" -> True+    "infixl" -> True+    "infixr" -> True+    "instance" -> True+    "let" -> True+    "module" -> True+    "newtype" -> True+    "of" -> True+    "then" -> True+    "type" -> True+    "where" -> True+    _ -> False++isVarId :: T.Text -> Bool+isVarId t = case T.uncons t of+    -- '_' is skipped on purpose so "Name" and "_Name_" cannot collide.+    Just (c, _) -> Char.isLower c && Char.isAlpha c+    Nothing -> False++isHaskellIdentifier :: T.Text -> Bool+isHaskellIdentifier t = Prelude.not (isVarId t) || isReservedId t++sanitize :: T.Text -> T.Text+sanitize t+    | isValid = t+    | isHaskellIdentifier t' = "_" <> t' <> "_"+    | otherwise = t'+  where+    isValid =+        Prelude.not (isHaskellIdentifier t)+            && isVarId t+            && T.all Char.isAlphaNum t+    t' = T.map replaceInvalidCharacters . T.filter (Prelude.not . parentheses) $ t+    replaceInvalidCharacters c+        | Char.isUpper c = Char.toLower c+        | Char.isSpace c = '_'+        | Char.isPunctuation c = '_' -- '-' will also become a '_'+        | Char.isSymbol c = '_'+        | Char.isAlphaNum c = c -- Blanket condition+        | otherwise = '_' -- If we're unsure we'll default to an underscore+    parentheses c = case c of+        '(' -> True+        ')' -> True+        '{' -> True+        '}' -> True+        '[' -> True+        ']' -> True+        _ -> False

@@ -41,7 +41,6 @@ import qualified Language.Haskell.TH.Syntax as TH  import DataFrame.Expression.Operators (col)-import DataFrame.Functions (sanitize) import DataFrame.Internal.Column (columnTypeString) import DataFrame.Internal.DataFrame (     DataFrame (..),@@ -50,6 +49,7 @@ import DataFrame.Internal.Expression (Expr) import DataFrame.Internal.Schema.TH (schemaTypeE) import DataFrame.Schema (Schema, makeSchema)+import DataFrame.TH.Identifier (sanitize) import Prelude as P  knownTypeNames :: M.Map String Name