Get rid of template preprocessing

Instead of preprocessing an outer layer of CPP when building Alex, just
always produce code that uses CPP. Combined with from-source
bootstrapping, this means Alex an have a perfectly bog standard build
system, with Makefiles and extra steps strictly optional.

I gather Hugs, and possibly other Haskell implementations, out of the
box doesn't support CPP, but I don't want this to stop us. Those can
just manually run CPP on the generated code first.

Author: Ericson2314

Makefile

@@ -22,7 +22,6 @@ sdist ::
 	$(ALEX) $(ALEX_OPTS) src/Scan.x -o src/Scan.hs
 	mv src/Parser.y src/Parser.y.boot
 	mv src/Scan.x src/Scan.x.boot
-	$(CABAL) v2-run gen-alex-sdist
 	$(CABAL) v2-sdist
 	@if [ ! -f "${SDIST_DIR}/alex-$(ALEX_VER).tar.gz" ]; then \
 		echo "Error: source tarball not found: dist/alex-$(ALEX_VER).tar.gz"; \

README.md

@@ -50,14 +50,7 @@ they are in your `$PATH` for the next steps!).
 
 ### Variant A
 
-First you need to generate the pre-processed templates via
-
-    $ cabal new-run gen-alex-sdist
-
-(otherwise `cabal install` will complain about
-"`data/AlexTemplate: copyFile: does not exist (No such file or directory)`")
-
-And then you can install `alex` simply by invoking
+You can install `alex` simply by invoking
 
     $ cabal install
 
@@ -67,8 +60,7 @@ from inside the Git folder.
 
 Alternatively, you can use the `Makefile` which automates the steps of
 producing a self-contained pre-bootstrapped source distribution with
-pre-generated lexer/scanners (and which also performs the `cabal
-new-run gen-alex-sdist` pre-preprocessing step):
+pre-generated lexer/scanners:
 
     $ make sdist
     $ cabal install dist/alex-*.tar.gz

alex.cabal

@@ -38,32 +38,8 @@ tested-with:
 data-dir: data/
 
 data-files:
-        AlexTemplate
-        AlexTemplate-debug
-        AlexTemplate-nopred
-        AlexTemplate-nopred-debug
-        AlexTemplate-latin1
-        AlexTemplate-latin1-debug
-        AlexTemplate-latin1-nopred
-        AlexTemplate-latin1-nopred-debug
-        AlexTemplate-ghc
-        AlexTemplate-ghc-debug
-        AlexTemplate-ghc-nopred
-        AlexTemplate-ghc-nopred-debug
-        AlexTemplate-ghc-latin1
-        AlexTemplate-ghc-latin1-debug
-        AlexTemplate-ghc-latin1-nopred
-        AlexTemplate-ghc-latin1-nopred-debug
-        AlexWrapper-basic
-        AlexWrapper-basic-bytestring
-        AlexWrapper-strict-bytestring
-        AlexWrapper-posn
-        AlexWrapper-posn-bytestring
-        AlexWrapper-monad
-        AlexWrapper-monad-bytestring
-        AlexWrapper-monadUserState
-        AlexWrapper-monadUserState-bytestring
-        AlexWrapper-gscan
+        AlexTemplate.hs
+        AlexWrappers.hs
 
 extra-source-files:
         CHANGELOG.md
@@ -94,8 +70,6 @@ extra-source-files:
         src/Parser.y.boot
         src/Scan.x.boot
         src/ghc_hooks.c
-        templates/GenericTemplate.hs
-        templates/wrappers.hs
         tests/Makefile
         tests/simple.x
         tests/null.x

cabal.project

@@ -1,6 +1,2 @@
-
 packages:
   ./
-  ./gen-alex-sdist/
-
-

templates/GenericTemplate.hs renamed to data/AlexTemplate.hs

@@ -8,71 +8,58 @@
 -- INTERNALS and main scanner engine
 
 #ifdef ALEX_GHC
-#undef __GLASGOW_HASKELL__
-#define ALEX_IF_GHC_GT_500 #if __GLASGOW_HASKELL__ > 500
-#define ALEX_IF_GHC_LT_503 #if __GLASGOW_HASKELL__ < 503
-#define ALEX_IF_GHC_GT_706 #if __GLASGOW_HASKELL__ > 706
-#define ALEX_IF_GHC_GE_901 #if __GLASGOW_HASKELL__ >= 901
-#define ALEX_ELIF_GHC_500 #elif __GLASGOW_HASKELL__ == 500
-#define ALEX_IF_BIGENDIAN #ifdef WORDS_BIGENDIAN
-#define ALEX_ELSE #else
-#define ALEX_ENDIF #endif
-#define ALEX_DEFINE #define
-#endif
-
-#ifdef ALEX_GHC
-#define ILIT(n) n#
-#define IBOX(n) (I# (n))
-#define FAST_INT Int#
+#  define ILIT(n) n#
+#  define IBOX(n) (I# (n))
+#  define FAST_INT Int#
 -- Do not remove this comment. Required to fix CPP parsing when using GCC and a clang-compiled alex.
-ALEX_IF_GHC_GT_706
-ALEX_DEFINE GTE(n,m) (tagToEnum# (n >=# m))
-ALEX_DEFINE EQ(n,m) (tagToEnum# (n ==# m))
-ALEX_ELSE
-ALEX_DEFINE GTE(n,m) (n >=# m)
-ALEX_DEFINE EQ(n,m) (n ==# m)
-ALEX_ENDIF
-#define PLUS(n,m) (n +# m)
-#define MINUS(n,m) (n -# m)
-#define TIMES(n,m) (n *# m)
-#define NEGATE(n) (negateInt# (n))
-#define IF_GHC(x) (x)
+#  if __GLASGOW_HASKELL__ > 706
+#    define GTE(n,m) (tagToEnum# (n >=# m))
+#    define EQ(n,m) (tagToEnum# (n ==# m))
+#  else
+#    define GTE(n,m) (n >=# m)
+#    define EQ(n,m) (n ==# m)
+#  endif
+#  define PLUS(n,m) (n +# m)
+#  define MINUS(n,m) (n -# m)
+#  define TIMES(n,m) (n *# m)
+#  define NEGATE(n) (negateInt# (n))
+#  define IF_GHC(x) (x)
 #else
-#define ILIT(n) (n)
-#define IBOX(n) (n)
-#define FAST_INT Int
-#define GTE(n,m) (n >= m)
-#define EQ(n,m) (n == m)
-#define PLUS(n,m) (n + m)
-#define MINUS(n,m) (n - m)
-#define TIMES(n,m) (n * m)
-#define NEGATE(n) (negate (n))
-#define IF_GHC(x)
+#  define ILIT(n) (n)
+#  define IBOX(n) (n)
+#  define FAST_INT Int
+#  define GTE(n,m) (n >= m)
+#  define EQ(n,m) (n == m)
+#  define PLUS(n,m) (n + m)
+#  define MINUS(n,m) (n - m)
+#  define TIMES(n,m) (n * m)
+#  define NEGATE(n) (negate (n))
+#  define IF_GHC(x)
 #endif
 
 #ifdef ALEX_GHC
 data AlexAddr = AlexA# Addr#
 -- Do not remove this comment. Required to fix CPP parsing when using GCC and a clang-compiled alex.
-ALEX_IF_GHC_LT_503
+#if __GLASGOW_HASKELL__ < 503
 uncheckedShiftL# = shiftL#
-ALEX_ENDIF
+#endif
 
 {-# INLINE alexIndexInt16OffAddr #-}
 alexIndexInt16OffAddr :: AlexAddr -> Int# -> Int#
 alexIndexInt16OffAddr (AlexA# arr) off =
-ALEX_IF_BIGENDIAN
+#ifdef WORDS_BIGENDIAN
   narrow16Int# i
   where
         i    = word2Int# ((high `uncheckedShiftL#` 8#) `or#` low)
         high = int2Word# (ord# (indexCharOffAddr# arr (off' +# 1#)))
         low  = int2Word# (ord# (indexCharOffAddr# arr off'))
         off' = off *# 2#
-ALEX_ELSE
-ALEX_IF_GHC_GE_901
+#else
+#if __GLASGOW_HASKELL__ >= 901
   int16ToInt#
-ALEX_ENDIF
+#endif
     (indexInt16OffAddr# arr off)
-ALEX_ENDIF
+#endif
 #else
 alexIndexInt16OffAddr arr off = arr ! off
 #endif
@@ -81,7 +68,7 @@ alexIndexInt16OffAddr arr off = arr ! off
 {-# INLINE alexIndexInt32OffAddr #-}
 alexIndexInt32OffAddr :: AlexAddr -> Int# -> Int#
 alexIndexInt32OffAddr (AlexA# arr) off =
-ALEX_IF_BIGENDIAN
+#ifdef WORDS_BIGENDIAN
   narrow32Int# i
   where
    i    = word2Int# ((b3 `uncheckedShiftL#` 24#) `or#`
@@ -92,24 +79,24 @@ ALEX_IF_BIGENDIAN
    b1   = int2Word# (ord# (indexCharOffAddr# arr (off' +# 1#)))
    b0   = int2Word# (ord# (indexCharOffAddr# arr off'))
    off' = off *# 4#
-ALEX_ELSE
-ALEX_IF_GHC_GE_901
+#else
+#if __GLASGOW_HASKELL__ >= 901
   int32ToInt#
-ALEX_ENDIF
+#endif
     (indexInt32OffAddr# arr off)
-ALEX_ENDIF
+#endif
 #else
 alexIndexInt32OffAddr arr off = arr ! off
 #endif
 
 #ifdef ALEX_GHC
 
-ALEX_IF_GHC_LT_503
+#if __GLASGOW_HASKELL__ < 503
 quickIndex arr i = arr ! i
-ALEX_ELSE
+#else
 -- GHC >= 503, unsafeAt is available from Data.Array.Base.
 quickIndex = unsafeAt
-ALEX_ENDIF
+#endif
 #else
 quickIndex arr i = arr ! i
 #endif

templates/wrappers.hs renamed to data/AlexWrappers.hs

@@ -11,7 +11,7 @@
 
 module AbsSyn (
   Code, Directive(..), Scheme(..),
-  wrapperName,
+  wrapperCppDefs,
   Scanner(..),
   RECtx(..),
   RExp(..), nullable,
@@ -70,22 +70,23 @@ strtype :: Bool -> String
 strtype True = "ByteString.ByteString"
 strtype False = "String"
 
-wrapperName :: Scheme -> Maybe String
-wrapperName Default {} = Nothing
-wrapperName GScan {} = Just "gscan"
-wrapperName Basic { basicStrType = Str } = Just "basic"
-wrapperName Basic { basicStrType = Lazy } = Just "basic-bytestring"
-wrapperName Basic { basicStrType = Strict } = Just "strict-bytestring"
-wrapperName Posn { posnByteString = False } = Just "posn"
-wrapperName Posn { posnByteString = True } = Just "posn-bytestring"
-wrapperName Monad { monadByteString = False,
-                    monadUserState = False } = Just "monad"
-wrapperName Monad { monadByteString = True,
-                    monadUserState = False } = Just "monad-bytestring"
-wrapperName Monad { monadByteString = False,
-                    monadUserState = True } = Just "monadUserState"
-wrapperName Monad { monadByteString = True,
-                    monadUserState = True } = Just "monadUserState-bytestring"
+
+wrapperCppDefs :: Scheme -> Maybe [String]
+wrapperCppDefs Default {} = Nothing
+wrapperCppDefs GScan {} = Just ["ALEX_GSCAN"]
+wrapperCppDefs Basic { basicStrType = Str } = Just ["ALEX_BASIC"]
+wrapperCppDefs Basic { basicStrType = Lazy } = Just ["ALEX_BASIC_BYTESTRING"]
+wrapperCppDefs Basic { basicStrType = Strict } = Just ["ALEX_STRICT_BYTESTRING"]
+wrapperCppDefs Posn { posnByteString = False } = Just ["ALEX_POSN"]
+wrapperCppDefs Posn { posnByteString = True } = Just ["ALEX_POSN_BYTESTRING"]
+wrapperCppDefs Monad { monadByteString = False,
+                       monadUserState = False } = Just ["ALEX_MONAD"]
+wrapperCppDefs Monad { monadByteString = True,
+                       monadUserState = False } = Just ["ALEX_MONAD_BYTESTRING"]
+wrapperCppDefs Monad { monadByteString = False,
+                       monadUserState = True } = Just ["ALEX_MONAD", "ALEX_MONAD_USER_STATE"]
+wrapperCppDefs Monad { monadByteString = True,
+                       monadUserState = True } = Just ["ALEX_MONAD_BYTESTRING", "ALEX_MONAD_USER_STATE"]
 
 -- TODO: update this comment
 --