core driver

organisation and tidying

CHANGELOG.md

@@ -0,0 +1,19 @@
+# unreleased
+
+* New tag syntax:
+  ```hs
+  case x of
+    { 1 -> something
+    ; 2 -> another
+    }
+  ```
+  is now written as
+  ```hs
+  case x of
+    { <1> -> something
+    ; <2> -> another
+    }
+  ```
+
+# Release 1.0.0
+

Makefile_happysrcs

@@ -1,5 +1,5 @@
 HAPPY = happy
-HAPPY_OPTS = -a -g -c
+HAPPY_OPTS = -a -g -c -i/tmp/t.info
 ALEX = alex
 ALEX_OPTS = -g
 
@@ -8,8 +8,8 @@ CABAL_BUILD = dist-newstyle/build/x86_64-osx/ghc-9.6.2/rlp-0.1.0.0/build
 
 all: parsers lexers
 
-parsers: $(CABAL_BUILD)/Rlp/Parse.hs
-lexers: $(CABAL_BUILD)/Rlp/Lex.hs
+parsers: $(CABAL_BUILD)/Rlp/Parse.hs $(CABAL_BUILD)/Core/Parse.hs
+lexers: $(CABAL_BUILD)/Rlp/Lex.hs $(CABAL_BUILD)/Core/Lex.hs
 
 $(CABAL_BUILD)/Rlp/Parse.hs: $(SRC)/Rlp/Parse.y
 	$(HAPPY) $(HAPPY_OPTS) $< -o $@
@@ -17,3 +17,9 @@ $(CABAL_BUILD)/Rlp/Parse.hs: $(SRC)/Rlp/Parse.y
 $(CABAL_BUILD)/Rlp/Lex.hs: $(SRC)/Rlp/Lex.x
 	$(ALEX) $(ALEX_OPTS) $< -o $@
 
+$(CABAL_BUILD)/Core/Parse.hs: $(SRC)/Core/Parse.y
+	$(HAPPY) $(HAPPY_OPTS) $< -o $@
+
+$(CABAL_BUILD)/Core/Lex.hs: $(SRC)/Core/Lex.x
+	$(ALEX) $(ALEX_OPTS) $< -o $@
+

README.md

@@ -23,19 +23,19 @@ $ cabal test --test-show-details=direct
 $ rlpc -ddump-eval examples/factorial.hs
 # Compile and evaluate t.hs, with evaluation info dumped to t.log
 $ rlpc -ddump-eval -l t.log t.hs
-# Print the raw structure describing the compiler options and die
+# Print the raw structure describing the compiler options
 # (option parsing still must succeed in order to print)
 $ rlpc -ddump-opts t.hs
 ```
 
 ### Potential Features
 Listed in order of importance.
-- [ ] ADTs
-- [ ] First-class functions
+- [x] ADTs
+- [x] First-class functions
 - [ ] Higher-kinded types
 - [ ] Typeclasses
-- [ ] Parametric polymorphism
-- [ ] Hindley-Milner type inference
+- [x] Parametric polymorphism
+- [x] Hindley-Milner type inference
 - [ ] Newtype coercion
 - [ ] Parallelism
 
@@ -66,32 +66,61 @@ Listed in order of importance.
         - [ ] TCO
         - [ ] DCE
 - [ ] Frontend
-    - [ ] High-level language
-        - [ ] AST
-        - [ ] Lexer
-        - [ ] Parser
+    - [x] High-level language
+        - [x] AST
+        - [x] Lexer
+        - [x] Parser
     - [ ] Translation to the core language
         - [ ] Constraint solver
         - [ ] `do`-notation
     - [x] CLI
 - [ ] Documentation
-    - [ ] State transition rules
+    - [x] State transition rules
     - [ ] How does the evaluation model work?
+    - [ ] The Hindley-Milner type system
     - [ ] CLI usage
     - [ ] Tail call optimisation
-    - [x] Parsing rlp
+    - [ ] Parsing rlp
+    - [ ] Trees That Grow
 - [ ] Tests
     - [x] Generic example programs
     - [ ] Parser
 
-### December Release Plan
-- [ ] Tests
+### ~~December Release Plan~~
+- [x] Tests
     - [ ] Core lexer
     - [ ] Core parser
-    - [ ] Evaluation model
+    - [x] Evaluation model
 - [ ] Benchmarks
-- [ ] Stable Core lexer
-- [ ] Stable Core parser
-- [ ] Stable evaluation model
-    - [ ] Garbage Collection
+- [x] Stable Core lexer
+- [x] Stable Core parser
+- [x] Stable evaluation model
+    - [x] Garbage Collection
 - [ ] Stable documentation for the evaluation model
+
+### January Release Plan
+- [ ] Beta rl' to Core
+- [ ] UX improvements
+    - [ ] Actual compiler errors -- no more unexceptional `error` calls
+    - [ ] Better CLI dump flags
+    - [ ] Annotate the AST with token positions for errors
+- [ ] More examples
+
+### March Release Plan
+- [ ] Tests
+    - [ ] rl' parser
+    - [ ] rl' lexer
+
+### Indefinite Release Plan
+
+This list is more concrete than the milestones, but likely further in the future
+than the other release plans.
+
+- [ ] Stable rl' to Core
+- [ ] Core polish
+    - [ ] Better, stable parser
+    - [ ] Better, stable lexer
+    - [ ] Less hacky handling of named data
+    - [ ] Less hacky pragmas
+- [ ] GM to LLVM
+

app/CoreDriver.hs

@@ -0,0 +1,17 @@
+module CoreDriver
+    ( driver
+    )
+    where
+--------------------------------------------------------------------------------
+import Compiler.RLPC
+import Control.Monad
+
+import Core.Lex
+import Core.Parse
+import GM
+--------------------------------------------------------------------------------
+
+driver :: RLPCIO ()
+driver = forFiles_ $ \f ->
+    withSource f (lexCoreR >=> parseCoreProgR >=> evalProgR)
+

app/Main.hs

@@ -10,12 +10,16 @@ import Data.HashSet             qualified as S
 import Data.Text                (Text)
 import Data.Text                qualified as T
 import Data.Text.IO             qualified as TIO
+import Data.List
 import System.IO
 import System.Exit              (exitSuccess)
 import Core
 import TI
 import GM
 import Lens.Micro.Mtl
+
+import CoreDriver               qualified
+import RlpDriver                qualified
 ----------------------------------------------------------------------------------
 
 optParser :: ParserInfo RLPCOptions
@@ -37,9 +41,15 @@ options = RLPCOptions
     {- -d -}
     <*> fmap S.fromList # many # option debugFlagReader
         (  short 'd'
-        <> help "dump evaluation logs"
+        <> help "pass debug flags"
         <> metavar "DEBUG FLAG"
         )
+    {- -f -}
+    <*> fmap S.fromList # many # option compilerFlagReader
+        (  short 'f'
+        <> help "pass compilation flags"
+        <> metavar "COMPILATION FLAG"
+        )
     {- --evaluator, -e -}
     <*> option evaluatorReader
         (  long "evaluator"
@@ -55,96 +65,45 @@ options = RLPCOptions
                 \triggering the garbage collector"
         <> value 50
         )
+    <*> option languageReader
+        (  long "language"
+        <> short 'x'
+        )
     <*> some (argument str $ metavar "FILES...")
     where
         infixr 9 #
         f # x = f x
 
+languageReader :: ReadM Language
+languageReader = maybeReader $ \case
+    "rlp"   -> Just LanguageRlp
+    "core"  -> Just LanguageCore
+    _       -> Nothing
+
+debugFlagReader :: ReadM DebugFlag
+debugFlagReader = str
+
+compilerFlagReader :: ReadM CompilerFlag
+compilerFlagReader = str
+
 evaluatorReader :: ReadM Evaluator
 evaluatorReader = maybeReader $ \case
     "gm"  -> Just EvaluatorGM
-    "tim" -> Just EvaluatorTI
+    "ti"  -> Just EvaluatorTI
     _     -> Nothing
 
 mmany :: (Alternative f, Monoid m) => f m -> f m
 mmany v = liftA2 (<>) v (mmany v)
 
-debugFlagReader :: ReadM DebugFlag
-debugFlagReader = maybeReader $ \case
-    "dump-eval"  -> Just DDumpEval
-    "dump-opts"  -> Just DDumpOpts
-    "dump-ast"   -> Just DDumpAST
-    _            -> Nothing
-
 ----------------------------------------------------------------------------------
 
--- temp
-data CompilerError = CompilerError String
-    deriving Show
-
-instance Exception CompilerError
-
 main :: IO ()
 main = do
     opts <- execParser optParser
-    (_, es) <- evalRLPCIO opts driver
-    forM_ es $ \ (CompilerError e) -> print $ "warning: " <> e
-    pure ()
+    void $ evalRLPCIO opts driver
 
-driver :: RLPCIO CompilerError ()
-driver = sequence_
-    [ dshowFlags
-    , ddumpAST
-    , ddumpEval
-    ]
-
-dshowFlags :: RLPCIO CompilerError ()
-dshowFlags = whenFlag flagDDumpOpts do
-    ask >>= liftIO . print
-
-ddumpAST :: RLPCIO CompilerError ()
-ddumpAST = whenFlag flagDDumpAST $ forFiles_ \o f -> do
-    liftIO $ withFile f ReadMode $ \h -> do
-        s <- TIO.hGetContents h
-        case parseProg o s of
-            Right (a,_) -> hPutStrLn stderr $ show a
-            Left e -> error "todo errors lol"
-
-ddumpEval :: RLPCIO CompilerError ()
-ddumpEval = whenFlag flagDDumpEval do
-    fs <- view rlpcInputFiles
-    forM_ fs $ \f -> liftIO (TIO.readFile f) >>= doProg
-
-    where
-        doProg :: Text -> RLPCIO CompilerError ()
-        doProg s = ask >>= \o -> case parseProg o s of
-            -- TODO: error handling
-            Left e -> addFatal . CompilerError $ show e
-            Right (a,_) -> do
-                log <- view rlpcLogFile
-                dumpEval <- chooseEval
-                case log of
-                    Just f  -> liftIO $ withFile f WriteMode $ dumpEval a
-                    Nothing -> liftIO $ dumpEval a stderr
-
-        -- choose the appropriate model based on the compiler opts
-        chooseEval = do
-            ev <- view rlpcEvaluator
-            pure $ case ev of
-                EvaluatorGM  -> v GM.hdbgProg
-                EvaluatorTI -> v TI.hdbgProg
-            where v f p h = f p h *> pure ()
-
-parseProg :: RLPCOptions
-          -> Text
-          -> Either SrcError (Program', [SrcError])
-parseProg o = evalRLPC o . (lexCore >=> parseCoreProg)
-
-forFiles_ :: (Monad m)
-          => (RLPCOptions -> FilePath -> RLPCT e m a)
-          -> RLPCT e m ()
-forFiles_ k = do
-    fs <- view rlpcInputFiles
-    o <- ask
-    forM_ fs (k o)
+driver :: RLPCIO ()
+driver = view rlpcLanguage >>= \case
+    LanguageCore -> CoreDriver.driver
+    LanguageRlp  -> RlpDriver.driver
 

app/RlpDriver.hs

@@ -0,0 +1,11 @@
+module RlpDriver
+    ( driver
+    )
+    where
+--------------------------------------------------------------------------------
+import Compiler.RLPC
+--------------------------------------------------------------------------------
+
+driver :: RLPCIO ()
+driver = undefined
+

doc/src/commentary/gm.rst

@@ -112,5 +112,3 @@ The way around this is quite simple: simply offset the stack when w
    :end-before: -- << [ref/compileC]
    :caption: src/GM.hs
 
-
-

doc/src/commentary/layout-lexing.rst

@@ -2,16 +2,21 @@ Lexing, Parsing, and Layouts
 ============================
 
 The C-style languages of my previous experiences have all had quite trivial
-lexical analysis stages, peaking in complexity when I streamed tokens lazily in
-C. The task of tokenising a C-style language is very simple in description: you
-ignore all whitespace and point out what you recognise. If you don't recognise
-something, check if it's a literal or an identifier. Should it be neither,
-return an error.
+lexical analysis stages: you ignore all whitespace and point out the symbols you
+recognise. If you don't recognise something, check if it's a literal or an
+identifier. Should it be neither, return an error.
 
-On paper, both lexing and parsing a Haskell-like language seem to pose a few
+In contrast, both lexing and parsing a Haskell-like language poses a number of
 greater challenges. Listed by ascending intimidation factor, some of the
 potential roadblocks on my mind before making an attempt were:
 
+* Context-sensitive keywords; Haskell allows for some words to be used as
+  identifiers in appropriate contexts, such as :code:`family`, :code:`role`,
+  :code:`as`. Reading a note_ found in `GHC's lexer`_, it appears that keywords
+  are only considered in bodies for which their use is relevant, e.g.
+  :code:`family` and :code:`role` in type declarations, :code:`as` after
+  :code:`case`; :code:`if`, :code:`then`, and :code:`else` in expressions, etc.
+
 * Operators; Haskell has not only user-defined infix operators, but user-defined
   precedence levels and associativities. I recall using an algorithm that looked
   up infix, prefix, postfix, and even mixfix operators up in a global table to
@@ -19,17 +24,9 @@ potential roadblocks on my mind before making an attempt were:
   stored in the table). I never modified the table at runtime, however this
   could be a very nice solution for Haskell.
 
-* Context-sensitive keywords; Haskell allows for some words to be used as identifiers in
-  appropriate contexts, such as :code:`family`, :code:`role`, :code:`as`.
-  Reading a note_ found in `GHC's lexer`_,
-  it appears that keywords are only considered in bodies for which their use is
-  relevant, e.g. :code:`family` and :code:`role` in type declarations,
-  :code:`as` after :code:`case`; :code:`if`, :code:`then`, and :code:`else` in
-  expressions, etc.
-
 * Whitespace sensitivity; While I was comfortable with the idea of a system
-  similar to Python's INDENT/DEDENT tokens, Haskell seemed to use whitespace to
-  section code in a way that *felt* different.
+  similar to Python's INDENT/DEDENT tokens, Haskell's layout system is based on
+  alignment and is very generous with line-folding.
 
 .. _note: https://gitlab.haskell.org/ghc/ghc/-/wikis/commentary/coding-style#2-using-notes
 .. _GHC's lexer: https://gitlab.haskell.org/ghc/ghc/-/blob/master/compiler/GHC/Parser/Lexer.x#L1133
@@ -45,9 +42,9 @@ We will compare and contrast with Python's lexical analysis. Much to my dismay,
 Python uses newlines and indentation to separate statements and resolve scope
 instead of the traditional semicolons and braces found in C-style languages (we
 may generally refer to these C-style languages as *explicitly-sectioned*).
-Internally during tokenisation, when the Python lexer begins a new line, they
-compare the indentation of the new line with that of the previous and apply the
-following rules:
+Internally during tokenisation, when the Python lexer encounters a new line, the
+indentation of the new line is compared with that of the previous and the
+following rules are applied:
 
 1. If the new line has greater indentation than the previous, insert an INDENT
    token and push the new line's indentation level onto the indentation stack
@@ -60,44 +57,37 @@ following rules:
 3. If the indentation is equal, insert a NEWLINE token to terminate the previous
    line, and leave it at that!
 
-Parsing Python with the INDENT, DEDENT, and NEWLINE tokens is identical to
-parsing a language with braces and semicolons. This is a solution pretty in line
-with Python's philosophy of the "one correct answer" (TODO: this needs a
-source). In developing our *layout* rules, we will follow in the pattern of
-translating the whitespace-sensitive source language to an explicitly sectioned
-language.
+On the parser's end, the INDENT, DEDENT, and NEWLINE tokens are identical to
+braces and semicolons. In developing our *layout* rules, we will follow in the
+pattern of translating the whitespace-sensitive source language to an explicitly
+sectioned language.
 
 But What About Haskell?
 ***********************
 
-We saw that Python, the most notable example of an implicitly sectioned
-language, is pretty simple to lex. Why then am I so afraid of Haskell's layouts?
-To be frank, I'm far less scared after asking myself this -- however there are
-certainly some new complexities that Python needn't concern. Haskell has
-implicit line *continuation*: forms written over multiple lines; indentation
-styles often seen in Haskell are somewhat esoteric compared to Python's
-"s/[{};]//".
+Parsing Haskell -- and thus rl' -- is only slightly more complex than Python,
+but the design is certainly more sensitive. 
 
 .. code-block:: haskell
 
-   -- line continuation
+   -- line folds
    something = this is a
        single expression
 
    -- an extremely common style found in haskell
-   data Python = Users
-       { are        :: Crying
-       , right      :: About
-       , now        :: Sorry
+   data Some = Data
+       { is    :: Presented
+       , in    :: This
+       , silly :: Style
        }
 
-   -- another formatting oddity
+   -- another style oddity
    -- note that this is not a single
    -- continued line! `look at`,
-   -- `this`, and `alignment` are all
-   -- separate expressions!
+   -- `this odd`, and `alignment` are all
+   -- discrete items!
    anotherThing = do look at
-                     this
+                     this odd
                      alignment
 
 But enough fear, lets actually think about implementation. Firstly, some
@@ -233,3 +223,4 @@ References
 
 * `Haskell syntax reference
   <https://www.haskell.org/onlinereport/haskell2010/haskellch10.html>`_
+

doc/src/commentary/type-inference.rst

@@ -0,0 +1,5 @@
+Type Inference in rl'
+=====================
+
+rl' implements type inference via the Hindley-Milner type system.
+

doc/src/references/rlp-inference-rules.rst

@@ -0,0 +1,17 @@
+rl' Inference Rules
+===================
+
+.. rubric::
+   [Var]
+
+.. math::
+   \frac{x : \tau \in \Gamma}
+        {\Gamma \vdash x : \tau}
+
+.. rubric::
+   [App]
+
+.. math::
+   \frac{\Gamma \vdash f : \alpha \to \beta \qquad \Gamma \vdash x : \alpha}
+        {\Gamma \vdash f x : \beta}
+

examples/factorial.hs

@@ -1,6 +1,6 @@
 fac n = case (==#) n 0 of
-    { 1 -> 1
-    ; 0 -> (*#) n (fac ((-#) n 1))
+    { <1> -> 1
+    ; <0> -> (*#) n (fac ((-#) n 1))
     };
 
 main = fac 3;

examples/sumList.hs

@@ -2,8 +2,8 @@ nil = Pack{0 0};
 cons x y = Pack{1 2} x y;
 list = cons 1 (cons 2 (cons 3 nil));
 sum l = case l of
-    { 0      -> 0
-    ; 1 x xs -> (+#) x (sum xs)
+    { <0>      -> 0
+    ; <1> x xs -> (+#) x (sum xs)
     };
 main = sum list;
 

programming-language-checklist

@@ -0,0 +1,105 @@
+Programming Language Checklist
+by Colin McMillen, Jason Reed, and Elly Fong-Jones, 2011-10-10.
+
+You appear to be advocating a new:
+[x] functional  [ ] imperative  [ ] object-oriented  [ ] procedural [ ] stack-based
+[ ] "multi-paradigm"  [x] lazy  [ ] eager  [x] statically-typed  [ ] dynamically-typed
+[x] pure  [ ] impure  [ ] non-hygienic  [ ] visual  [x] beginner-friendly
+[ ] non-programmer-friendly  [ ] completely incomprehensible
+programming language.  Your language will not work.  Here is why it will not work.
+
+You appear to believe that:
+[ ] Syntax is what makes programming difficult
+[x] Garbage collection is free                [x] Computers have infinite memory
+[x] Nobody really needs:
+    [x] concurrency  [x] a REPL  [x] debugger support  [x] IDE support  [x] I/O
+    [x] to interact with code not written in your language
+[ ] The entire world speaks 7-bit ASCII
+[ ] Scaling up to large software projects will be easy
+[ ] Convincing programmers to adopt a new language will be easy
+[ ] Convincing programmers to adopt a language-specific IDE will be easy
+[ ] Programmers love writing lots of boilerplate
+[ ] Specifying behaviors as "undefined" means that programmers won't rely on them
+[ ] "Spooky action at a distance" makes programming more fun
+
+Unfortunately, your language (has/lacks):
+[x] comprehensible syntax  [ ] semicolons  [x] significant whitespace  [ ] macros
+[ ] implicit type conversion  [ ] explicit casting  [x] type inference
+[ ] goto  [ ] exceptions  [x] closures  [x] tail recursion  [ ] coroutines
+[ ] reflection  [ ] subtyping  [ ] multiple inheritance  [x] operator overloading
+[x] algebraic datatypes  [x] recursive types  [x] polymorphic types
+[ ] covariant array typing  [x] monads  [ ] dependent types
+[x] infix operators  [x] nested comments  [ ] multi-line strings  [ ] regexes
+[ ] call-by-value  [x] call-by-name  [ ] call-by-reference  [ ] call-cc
+
+The following philosophical objections apply:
+[ ] Programmers should not need to understand category theory to write "Hello, World!"
+[ ] Programmers should not develop RSI from writing "Hello, World!"
+[ ] The most significant program written in your language is its own compiler
+[x] The most significant program written in your language isn't even its own compiler
+[x] No language spec
+[x] "The implementation is the spec"
+   [ ] The implementation is closed-source  [ ] covered by patents  [ ] not owned by you
+[ ] Your type system is unsound  [ ] Your language cannot be unambiguously parsed
+   [ ] a proof of same is attached
+   [ ] invoking this proof crashes the compiler
+[x] The name of your language makes it impossible to find on Google
+[x] Interpreted languages will never be as fast as C
+[ ] Compiled languages will never be "extensible"
+[ ] Writing a compiler that understands English is AI-complete
+[ ] Your language relies on an optimization which has never been shown possible
+[ ] There are less than 100 programmers on Earth smart enough to use your language
+[ ] ____________________________ takes exponential time
+[ ] ____________________________ is known to be undecidable
+
+Your implementation has the following flaws:
+[ ] CPUs do not work that way
+[ ] RAM does not work that way
+[ ] VMs do not work that way
+[ ] Compilers do not work that way
+[ ] Compilers cannot work that way
+[ ] Shift-reduce conflicts in parsing seem to be resolved using rand()
+[ ] You require the compiler to be present at runtime
+[ ] You require the language runtime to be present at compile-time
+[ ] Your compiler errors are completely inscrutable
+[ ] Dangerous behavior is only a warning
+[ ] The compiler crashes if you look at it funny
+[x] The VM crashes if you look at it funny
+[x] You don't seem to understand basic optimization techniques
+[x] You don't seem to understand basic systems programming
+[ ] You don't seem to understand pointers
+[ ] You don't seem to understand functions
+
+Additionally, your marketing has the following problems:
+[x] Unsupported claims of increased productivity
+[x] Unsupported claims of greater "ease of use"
+[ ] Obviously rigged benchmarks
+   [ ] Graphics, simulation, or crypto benchmarks where your code just calls
+       handwritten assembly through your FFI
+   [ ] String-processing benchmarks where you just call PCRE
+   [ ] Matrix-math benchmarks where you just call BLAS
+[x] Noone really believes that your language is faster than:
+    [x] assembly  [x] C  [x] FORTRAN  [x] Java  [x] Ruby  [ ] Prolog
+[ ] Rejection of orthodox programming-language theory without justification
+[x] Rejection of orthodox systems programming without justification
+[ ] Rejection of orthodox algorithmic theory without justification
+[ ] Rejection of basic computer science without justification
+
+Taking the wider ecosystem into account, I would like to note that:
+[x] Your complex sample code would be one line in: examples/
+[ ] We already have an unsafe imperative language
+[ ] We already have a safe imperative OO language
+[x] We already have a safe statically-typed eager functional language
+[ ] You have reinvented Lisp but worse
+[ ] You have reinvented Javascript but worse
+[ ] You have reinvented Java but worse
+[ ] You have reinvented C++ but worse
+[ ] You have reinvented PHP but worse
+[ ] You have reinvented PHP better, but that's still no justification
+[ ] You have reinvented Brainfuck but non-ironically
+
+In conclusion, this is what I think of you:
+[ ] You have some interesting ideas, but this won't fly.
+[x] This is a bad language, and you should feel bad for inventing it.
+[ ] Programming in this language is an adequate punishment for inventing it.
+

rlp.cabal

@@ -7,7 +7,7 @@ license:            GPL-2.0-only
 -- license-file:       LICENSE
 author:             crumbtoo
 maintainer:         crumb@disroot.org
--- copyright:
+copyright:          Madeleine Sydney Ślaga
 category:           Language
 build-type:         Simple
 extra-doc-files:    README.md
@@ -37,8 +37,8 @@ library
                     , Rlp.Parse.Associate
                     , Rlp.Lex
                     , Rlp.Parse.Types
-
-    other-modules:    Data.Heap
+                    , Compiler.Types
+                    , Data.Heap
                     , Data.Pretty
                     , Core.Parse
                     , Core.Lex
@@ -48,7 +48,7 @@ library
     build-tool-depends: happy:happy, alex:alex
 
     -- other-extensions:
-    build-depends:    base ^>=4.18.0.0
+    build-depends:    base >=4.17 && <4.20
                     -- required for happy
                     , array >= 0.5.5 && < 0.6
                     , containers >= 0.6.7 && < 0.7
@@ -69,16 +69,26 @@ library
                     , data-fix >= 0.3.2 && < 0.4
                     , utf8-string >= 1.0.2 && < 1.1
                     , extra >= 1.7.0 && < 2
+                    , semigroupoids
+                    , comonad
+                    , lens
+                    , text-ansi
 
     hs-source-dirs:   src
     default-language: GHC2021
 
+    default-extensions:
+        OverloadedStrings
+        TypeFamilies
+        LambdaCase
+
 executable rlpc
     import:           warnings
     main-is:          Main.hs
-    -- other-modules:
-    -- other-extensions:
-    build-depends: base ^>=4.18.0.0
+    other-modules:    RlpDriver
+                    , CoreDriver
+
+    build-depends: base >=4.17.0.0 && <4.20.0.0
                  , rlp
                  , optparse-applicative >= 0.18.1 && < 0.19
                  , microlens >= 0.4.13 && < 0.5

src/Compiler/JustRun.hs

@@ -26,21 +26,23 @@ import Data.Function                    ((&))
 import GM
 ----------------------------------------------------------------------------------
 
-justLexSrc :: String -> Either RlpcError [CoreToken]
+justLexSrc :: String -> Either [MsgEnvelope RlpcError] [CoreToken]
 justLexSrc s = lexCoreR (T.pack s)
              & fmap (map $ \ (Located _ _ _ t) -> t)
              & rlpcToEither
 
-justParseSrc :: String -> Either RlpcError Program'
+justParseSrc :: String -> Either [MsgEnvelope RlpcError] Program'
 justParseSrc s = parse (T.pack s)
                & rlpcToEither
     where parse = lexCoreR >=> parseCoreProgR
 
-justTypeCheckSrc :: String -> Either RlpcError Program'
+justTypeCheckSrc :: String -> Either [MsgEnvelope RlpcError] Program'
 justTypeCheckSrc s = typechk (T.pack s)
                    & rlpcToEither
     where typechk = lexCoreR >=> parseCoreProgR >=> checkCoreProgR
 
-rlpcToEither :: RLPC e a -> Either e a
-rlpcToEither = evalRLPC def >>> fmap fst
+rlpcToEither :: RLPC a -> Either [MsgEnvelope RlpcError] a
+rlpcToEither r = case evalRLPC def r of
+    (Just a,   _) -> Right a
+    (Nothing, es) -> Left es
 

src/Compiler/RLPC.hs

@@ -11,101 +11,110 @@ errors and the family of RLPC monads.
 -- only used for mtl instances
 {-# LANGUAGE UndecidableInstances #-}
 {-# LANGUAGE DeriveGeneric, DerivingStrategies, DerivingVia #-}
+{-# LANGUAGE BlockArguments, ViewPatterns #-}
 module Compiler.RLPC
-    ( RLPC
-    , RLPCT(..)
-    , RLPCIO
-    , RLPCOptions(RLPCOptions)
-    , RlpcError(..)
-    , IsRlpcError(..)
-    , rlpc
-    , addFatal
-    , addWound
-    , MonadErrorful
-    , Severity(..)
-    , Evaluator(..)
-    , evalRLPCT
-    , evalRLPCIO
-    , evalRLPC
-    , addRlpcWound
-    , addRlpcFatal
-    , liftRlpcErrs
-    , rlpcLogFile
-    , rlpcDebugOpts
-    , rlpcEvaluator
-    , rlpcInputFiles
-    , DebugFlag(..)
-    , whenFlag
-    , flagDDumpEval
-    , flagDDumpOpts
-    , flagDDumpAST
-    , def
+    (
+    -- * Rlpc Monad transformer
+      RLPCT(RLPCT),
+    -- ** Special cases
+      RLPC, RLPCIO
+    -- ** Running
+    , runRLPCT
+    , evalRLPCT, evalRLPCIO, evalRLPC
+    -- * Rlpc options
+    , Language(..), Evaluator(..)
+    , DebugFlag(..), CompilerFlag(..)
+    -- ** Lenses
+    , rlpcLogFile, rlpcDFlags, rlpcEvaluator, rlpcInputFiles, rlpcLanguage
+    -- * Misc. MTL-style functions
+    , liftErrorful, hoistRlpcT
+    -- * Misc. Rlpc Monad -related types
+    , RLPCOptions(RLPCOptions), IsRlpcError(..), RlpcError(..)
+    , MsgEnvelope(..), Severity(..)
+    , addDebugMsg
+    , whenDFlag, whenFFlag
+    -- * Misc. Utilities
+    , forFiles_, withSource
+    -- * Convenient re-exports
+    , addFatal, addWound, def
     )
     where
 ----------------------------------------------------------------------------------
 import Control.Arrow            ((>>>))
 import Control.Exception
+import Control.Monad
 import Control.Monad.Reader
 import Control.Monad.State      (MonadState(state))
 import Control.Monad.Errorful
+import Control.Monad.IO.Class
 import Compiler.RlpcError
+import Compiler.Types
 import Data.Functor.Identity
 import Data.Default.Class
+import Data.Foldable
 import GHC.Generics             (Generic)
+import Data.Maybe
 import Data.Hashable            (Hashable)
 import Data.HashSet             (HashSet)
 import Data.HashSet             qualified as S
 import Data.Coerce
-import Lens.Micro
-import Lens.Micro.TH
+import Data.Text                (Text)
+import Data.Text                qualified as T
+import Data.Text.IO             qualified as T
+import Text.ANSI                qualified as Ansi
+import Text.PrettyPrint         hiding ((<>))
+import Lens.Micro.Platform
+import Lens.Micro.Platform.Internal
+import System.Exit
 ----------------------------------------------------------------------------------
 
--- TODO: fancy errors
-newtype RLPCT e m a = RLPCT {
-        runRLPCT :: ReaderT RLPCOptions (ErrorfulT e m) a
+newtype RLPCT m a = RLPCT {
+        runRLPCT :: ReaderT RLPCOptions (ErrorfulT (MsgEnvelope RlpcError) m) a
     }
-    -- TODO: incorrect ussage of MonadReader. RLPC should have its own
-    -- environment access functions
-    deriving (Functor, Applicative, Monad, MonadReader RLPCOptions)
+    deriving ( Functor, Applicative, Monad
+             , MonadReader RLPCOptions, MonadErrorful (MsgEnvelope RlpcError))
 
-deriving instance (MonadIO m) => MonadIO (RLPCT e m)
+rlpc :: (IsRlpcError e, Monad m)
+     => (RLPCOptions -> (Maybe a, [MsgEnvelope e]))
+     -> RLPCT m a
+rlpc f = RLPCT . ReaderT $ \opt ->
+    ErrorfulT . pure $ f opt & _2 . each . mapped %~ liftRlpcError
 
-instance MonadTrans (RLPCT e) where
-    lift = RLPCT . lift . lift
+type RLPC = RLPCT Identity
 
-instance (MonadState s m) => MonadState s (RLPCT e m) where
-    state = lift . state
+type RLPCIO = RLPCT IO
 
-type RLPC e = RLPCT e Identity
-
-type RLPCIO e = RLPCT e IO
-
-evalRLPCT :: RLPCOptions
-          -> RLPCT e m a
-          -> m (Either e (a, [e]))
-evalRLPCT o = runRLPCT >>> flip runReaderT o >>> runErrorfulT
+instance (MonadIO m) => MonadIO (RLPCT m) where
 
 evalRLPC :: RLPCOptions
-         -> RLPC e a
-         -> Either e (a, [e])
-evalRLPC o m = coerce $ evalRLPCT o m
+         -> RLPC a
+         -> (Maybe a, [MsgEnvelope RlpcError])
+evalRLPC opt r = runRLPCT r
+               & flip runReaderT opt
+               & runErrorful
 
-evalRLPCIO :: (Exception e)
-           => RLPCOptions
-           -> RLPCIO e a
-           -> IO (a, [e])
-evalRLPCIO o m = do
-    m' <- evalRLPCT o m
-    case m' of
-        -- TODO: errors
-        Left e -> throwIO e
-        Right a -> pure a
+evalRLPCT :: RLPCOptions
+          -> RLPCT m a
+          -> m (Maybe a, [MsgEnvelope RlpcError])
+evalRLPCT opt r = runRLPCT r
+                & flip runReaderT opt
+                & runErrorfulT
+
+liftErrorful :: (Monad m, IsRlpcError e) => ErrorfulT (MsgEnvelope e) m a -> RLPCT m a
+liftErrorful e = RLPCT $ lift (fmap liftRlpcError `mapErrorful` e)
+
+hoistRlpcT :: (forall a. m a -> n a)
+         -> RLPCT m a -> RLPCT n a
+hoistRlpcT f rma = RLPCT $ ReaderT $ \opt ->
+    ErrorfulT $ f $ evalRLPCT opt rma
 
 data RLPCOptions = RLPCOptions
     { _rlpcLogFile     :: Maybe FilePath
-    , _rlpcDebugOpts   :: DebugOpts
+    , _rlpcDFlags      :: HashSet DebugFlag
+    , _rlpcFFlags      :: HashSet CompilerFlag
     , _rlpcEvaluator   :: Evaluator
     , _rlpcHeapTrigger :: Int
+    , _rlpcLanguage    :: Language
     , _rlpcInputFiles  :: [FilePath]
     }
     deriving Show
@@ -113,69 +122,117 @@ data RLPCOptions = RLPCOptions
 data Evaluator = EvaluatorGM | EvaluatorTI
     deriving Show
 
-data Severity = Error
-              | Warning
-              | Debug
-              deriving Show
-
--- temporary until we have a new doc building system
-type ErrorDoc = String
-
-instance (Monad m) => MonadErrorful e (RLPCT e m) where
-    addWound = RLPCT . lift . addWound
-    addFatal = RLPCT . lift . addFatal
-
-liftRlpcErrs :: (IsRlpcError e, Monad m)
-             => RLPCT e m a -> RLPCT RlpcError m a
-liftRlpcErrs m = RLPCT . ReaderT $ \r ->
-    mapErrors liftRlpcErr $ runRLPCT >>> (`runReaderT` r) $ m
-
-addRlpcWound :: (IsRlpcError e, Monad m) => e -> RLPCT RlpcError m ()
-addRlpcWound = addWound . liftRlpcErr
-
-addRlpcFatal :: (IsRlpcError e, Monad m) => e -> RLPCT RlpcError m ()
-addRlpcFatal = addWound . liftRlpcErr
-
-rlpc :: (Monad m) => ErrorfulT e m a -> RLPCT e m a
-rlpc = RLPCT . ReaderT . const
+data Language = LanguageRlp | LanguageCore
+    deriving Show
 
 ----------------------------------------------------------------------------------
 
 instance Default RLPCOptions where
     def = RLPCOptions
         { _rlpcLogFile = Nothing
-        , _rlpcDebugOpts = mempty
+        , _rlpcDFlags = mempty
+        , _rlpcFFlags = mempty
         , _rlpcEvaluator = EvaluatorGM
         , _rlpcHeapTrigger = 200
         , _rlpcInputFiles = []
+        , _rlpcLanguage = LanguageRlp
         }
 
-type DebugOpts = HashSet DebugFlag
+-- debug flags are passed with -dFLAG
+type DebugFlag = Text
 
-data DebugFlag = DDumpEval
-               | DDumpOpts
-               | DDumpAST
-               deriving (Show, Eq, Generic)
-
-instance Hashable DebugFlag
+type CompilerFlag = Text
 
 makeLenses ''RLPCOptions
 pure []
 
-whenFlag :: (MonadReader s m) => SimpleGetter s Bool -> m () -> m ()
-whenFlag l m = asks (^. l) >>= \a -> if a then m else pure ()
+addDebugMsg :: (Monad m, IsText e) => Text -> e -> RLPCT m ()
+addDebugMsg tag e = addWound . debugMsg tag $ Text [e ^. unpacked . packed]
 
--- there's probably a better way to write this. my current knowledge of lenses
--- is too weak.
-flagGetter :: DebugFlag -> SimpleGetter RLPCOptions Bool
-flagGetter d = to $ \s -> s ^. rlpcDebugOpts & S.member d
+-- TODO: rewrite this with prisms once microlens-pro drops :3
+whenDFlag :: (Monad m) => DebugFlag -> RLPCT m () -> RLPCT m ()
+whenDFlag f m = do
+    -- mfw no `At` instance for HashSet
+    fs <- view rlpcDFlags
+    let a = S.member f fs
+    when a m
 
-flagDDumpEval :: SimpleGetter RLPCOptions Bool
-flagDDumpEval = flagGetter DDumpEval
+whenFFlag :: (Monad m) => CompilerFlag -> RLPCT m () -> RLPCT m ()
+whenFFlag f m = do
+    -- mfw no `At` instance for HashSet
+    fs <- view rlpcFFlags
+    let a = S.member f fs
+    when a m
 
-flagDDumpOpts :: SimpleGetter RLPCOptions Bool
-flagDDumpOpts = flagGetter DDumpOpts
+--------------------------------------------------------------------------------
 
-flagDDumpAST :: SimpleGetter RLPCOptions Bool
-flagDDumpAST = flagGetter DDumpAST
+evalRLPCIO :: RLPCOptions -> RLPCIO a -> IO a
+evalRLPCIO opt r = do
+    (ma,es) <- evalRLPCT opt r
+    putRlpcErrs opt es
+    case ma of
+        Just x  -> pure x
+        Nothing -> die "Failed, no code compiled."
+
+putRlpcErrs :: RLPCOptions -> [MsgEnvelope RlpcError] -> IO ()
+putRlpcErrs opts = filter byTag
+               >>> traverse_ (putStrLn . ('\n':) . prettyRlpcMsg)
+    where
+        dflags = opts ^. rlpcDFlags
+
+        byTag :: MsgEnvelope RlpcError -> Bool
+        byTag (view msgSeverity -> SevDebug t) =
+            t `S.member` dflags
+
+prettyRlpcMsg :: MsgEnvelope RlpcError -> String
+prettyRlpcMsg m@(view msgSeverity -> SevDebug _) = prettyRlpcDebugMsg m
+prettyRlpcMsg m                                  = render $ docRlpcErr m
+
+prettyRlpcDebugMsg :: MsgEnvelope RlpcError -> String
+prettyRlpcDebugMsg msg =
+        T.unpack . foldMap mkLine $ [ t' | t <- ts, t' <- T.lines t ]
+    where
+        mkLine s = "-d" <> tag <> ": " <> s <> "\n"
+        Text ts = msg ^. msgDiagnostic
+        SevDebug tag = msg ^. msgSeverity
+
+docRlpcErr :: MsgEnvelope RlpcError -> Doc
+docRlpcErr msg = header
+                 $$ nest 2 bullets
+                 $$ source
+    where
+        source = vcat $ zipWith (<+>) rule srclines
+            where
+                rule = repeat (ttext . Ansi.blue . Ansi.bold $ "|")
+                srclines = ["", "<problematic source code>", ""]
+        filename = msgColour "<input>"
+        pos = msgColour $ tshow (msg ^. msgSpan . srcspanLine)
+                       <> ":"
+                       <> tshow (msg ^. msgSpan . srcspanColumn)
+
+        header = ttext $ filename <> msgColour ":" <> pos <> msgColour ": "
+                        <> errorColour "error" <> msgColour ":"
+
+        bullets = let Text ts = msg ^. msgDiagnostic
+                  in vcat $ hang "•" 2 . ttext . msgColour <$> ts
+
+        msgColour = Ansi.white . Ansi.bold
+        errorColour = Ansi.red . Ansi.bold
+        ttext = text . T.unpack
+        tshow :: (Show a) => a -> Text
+        tshow = T.pack . show
+
+--------------------------------------------------------------------------------
+
+forFiles_ :: (Monad m)
+          => (FilePath -> RLPCT m a)
+          -> RLPCT m ()
+forFiles_ k = do
+    fs <- view rlpcInputFiles
+    forM_ fs k
+
+-- TODO: catch any exceptions, i.e. non-existent files should be handled by the
+-- compiler
+withSource :: (MonadIO m) => FilePath -> (Text -> RLPCT m a) -> RLPCT m a
+withSource f k = liftIO (T.readFile f) >>= k
 

src/Compiler/RlpcError.hs

@@ -1,15 +1,77 @@
+{-# LANGUAGE TemplateHaskell #-}
+{-# LANGUAGE PatternSynonyms, ViewPatterns #-}
 module Compiler.RlpcError
-    ( RlpcError(..)
-    , IsRlpcError(..)
+    ( IsRlpcError(..)
+    , MsgEnvelope(..)
+    , Severity(..)
+    , RlpcError(..)
+    , msgSpan
+    , msgDiagnostic
+    , msgSeverity
+    , liftRlpcErrors
+    , errorMsg
+    , debugMsg
+    -- * Located Comonad
+    , Located(..)
+    , SrcSpan(..)
     )
     where
 ----------------------------------------------------------------------------------
 import Control.Monad.Errorful
+import Data.Text                (Text)
+import Data.Text                qualified as T
+import GHC.Exts                 (IsString(..))
+import Lens.Micro.Platform
+import Lens.Micro.Platform.Internal
+import Compiler.Types
 ----------------------------------------------------------------------------------
 
-data RlpcError = RlpcErr String -- temp
+data MsgEnvelope e = MsgEnvelope
+   { _msgSpan        :: SrcSpan
+   , _msgDiagnostic  :: e
+   , _msgSeverity    :: Severity
+   }
+   deriving (Functor, Show)
+
+newtype RlpcError = Text [Text]
     deriving Show
 
-class IsRlpcError a where
-    liftRlpcErr :: a -> RlpcError
+instance IsString RlpcError where
+    fromString = Text . pure . T.pack
+
+class IsRlpcError e where
+    liftRlpcError :: e -> RlpcError
+
+instance IsRlpcError RlpcError where
+    liftRlpcError = id
+
+data Severity = SevWarning
+              | SevError
+              | SevDebug Text
+              deriving Show
+
+makeLenses ''MsgEnvelope
+
+liftRlpcErrors :: (Functor m, IsRlpcError e)
+               => ErrorfulT e m a
+               -> ErrorfulT RlpcError m a
+liftRlpcErrors = mapErrorful liftRlpcError
+
+instance (IsRlpcError e) => IsRlpcError (MsgEnvelope e) where
+    liftRlpcError msg = msg ^. msgDiagnostic & liftRlpcError
+
+errorMsg :: SrcSpan -> e -> MsgEnvelope e
+errorMsg s e = MsgEnvelope
+    { _msgSpan = s
+    , _msgDiagnostic = e
+    , _msgSeverity = SevError
+    }
+
+debugMsg :: Text -> e -> MsgEnvelope e
+debugMsg tag e = MsgEnvelope
+    -- TODO: not pretty, but it is a debug message after all
+    { _msgSpan = SrcSpan 0 0 0 0
+    , _msgDiagnostic = e
+    , _msgSeverity = SevDebug tag
+    }
 

src/Compiler/Types.hs

@@ -0,0 +1,78 @@
+module Compiler.Types
+    ( SrcSpan(..)
+    , srcspanLine, srcspanColumn, srcspanAbs, srcspanLen
+    , Located(..)
+    , (<<~), (<~>)
+
+    -- * Re-exports
+    , Comonad
+    , Apply
+    , Bind
+    )
+    where
+--------------------------------------------------------------------------------
+import Control.Comonad
+import Data.Functor.Apply
+import Data.Functor.Bind
+import Control.Lens             hiding ((<<~))
+--------------------------------------------------------------------------------
+
+-- | Token wrapped with a span (line, column, absolute, length)
+data Located a = Located SrcSpan a
+    deriving (Show, Functor)
+
+instance Apply Located where
+    liftF2 f (Located sa p) (Located sb q)
+            = Located (sa <> sb) (p `f` q)
+
+instance Bind Located where
+    Located sa a >>- k = Located (sa <> sb) b
+        where
+            Located sb b = k a
+
+instance Comonad Located where
+    extract (Located _ a) = a
+    extend ck w@(Located p _) = Located p (ck w)
+
+data SrcSpan = SrcSpan
+    !Int -- ^ Line
+    !Int -- ^ Column
+    !Int -- ^ Absolute
+    !Int -- ^ Length
+    deriving Show
+
+tupling :: Iso' SrcSpan (Int, Int, Int, Int)
+tupling = iso (\ (SrcSpan a b c d) -> (a,b,c,d))
+              (\ (a,b,c,d) -> SrcSpan a b c d)
+
+srcspanLine, srcspanColumn, srcspanAbs, srcspanLen :: Lens' SrcSpan Int
+srcspanLine = tupling . _1
+srcspanColumn = tupling . _2
+srcspanAbs = tupling . _3
+srcspanLen = tupling . _4
+
+instance Semigroup SrcSpan where
+    SrcSpan la ca aa sa <> SrcSpan lb cb ab sb = SrcSpan l c a s where
+        l = min la lb
+        c = min ca cb
+        a = min aa ab
+        s = case aa `compare` ab of
+            EQ -> max sa sb
+            LT -> max sa (ab + lb - aa)
+            GT -> max sb (aa + la - ab)
+
+-- | A synonym for '(<<=)' with a tighter precedence and left-associativity for
+-- use with '(<~>)' in a sort of, comonadic pseudo-applicative style.
+
+(<<~) :: (Comonad w) => (w a -> b) -> w a -> w b
+(<<~) = (<<=)
+
+infixl 4 <<~
+
+-- | Similar to '(<*>)', but with a cokleisli arrow.
+
+(<~>) :: (Comonad w, Bind w) => w (w a -> b) -> w a -> w b
+mc <~> ma = mc >>- \f -> ma =>> f
+
+infixl 4 <~>
+

src/Control/Monad/Errorful.hs

@@ -1,73 +1,83 @@
-{-# LANGUAGE StandaloneDeriving #-}
 {-# LANGUAGE FunctionalDependencies #-}
-{-# LANGUAGE TupleSections, PatternSynonyms #-}
+{-# LANGUAGE PatternSynonyms #-}
+{-# LANGUAGE UndecidableInstances #-}
 module Control.Monad.Errorful
-    ( ErrorfulT
-    , runErrorfulT
+    ( ErrorfulT(..)
     , Errorful
+    , pattern Errorful
+    , errorful
     , runErrorful
-    , mapErrors
+    , mapErrorful
     , MonadErrorful(..)
     )
     where
 ----------------------------------------------------------------------------------
+import Control.Monad.State.Strict
+import Control.Monad.Reader
 import Control.Monad.Trans
 import Data.Functor.Identity
 import Data.Coerce
+import Data.HashSet                 (HashSet)
+import Data.HashSet                 qualified as H
 import Lens.Micro
 ----------------------------------------------------------------------------------
 
-newtype ErrorfulT e m a = ErrorfulT { runErrorfulT :: m (Either e (a, [e])) }
+newtype ErrorfulT e m a = ErrorfulT { runErrorfulT :: m (Maybe a, [e]) }
 
 type Errorful e = ErrorfulT e Identity
 
-pattern Errorful :: (Either e (a, [e])) -> Errorful e a
+pattern Errorful :: (Maybe a, [e]) -> Errorful e a
 pattern Errorful a = ErrorfulT (Identity a)
 
-runErrorful :: Errorful e a -> Either e (a, [e])
+errorful :: (Applicative m) => (Maybe a, [e]) -> ErrorfulT e m a
+errorful = ErrorfulT . pure
+
+runErrorful :: Errorful e a -> (Maybe a, [e])
 runErrorful m = coerce (runErrorfulT m)
 
 class (Applicative m) => MonadErrorful e m | m -> e where
-    addWound   :: e -> m ()
-    addFatal   :: e -> m a
-
-    -- not sure if i want to add this yet...
-    -- catchWound :: m a -> (e -> m a) -> m a
+    addWound :: e -> m ()
+    addFatal :: e -> m a
 
 instance (Applicative m) => MonadErrorful e (ErrorfulT e m) where
-    addWound e = ErrorfulT $ pure . Right $ ((), [e])
-    addFatal e = ErrorfulT $ pure . Left  $ e
+    addWound e = ErrorfulT $ pure (Just (), [e])
+    addFatal e = ErrorfulT $ pure (Nothing, [e])
 
 instance MonadTrans (ErrorfulT e) where
-    lift m = ErrorfulT (Right . (,[]) <$> m)
+    lift m = ErrorfulT ((\x -> (Just x,[])) <$> m)
 
 instance (MonadIO m) => MonadIO (ErrorfulT e m) where
     liftIO = lift . liftIO
 
 instance (Functor m) => Functor (ErrorfulT e m) where
-    fmap f (ErrorfulT m) = ErrorfulT $ fmap (_1 %~ f) <$> m
+    fmap f (ErrorfulT m) = ErrorfulT (m & mapped . _1 . _Just %~ f)
 
 instance (Applicative m) => Applicative (ErrorfulT e m) where
-    pure a = ErrorfulT (pure . Right $ (a, []))
+    pure a = ErrorfulT . pure $ (Just a, [])
 
-    m <*> a = ErrorfulT (m' `apply` a')
-        where
-            m' = runErrorfulT m
-            a' = runErrorfulT a
-            -- TODO: strict concatenation
-            apply = liftA2 $ liftA2 (\ (f,e1) (x,e2) -> (f x, e1 ++ e2)) 
+    ErrorfulT m <*> ErrorfulT n = ErrorfulT $ m `apply` n where
+        apply :: m (Maybe (a -> b), [e]) -> m (Maybe a, [e]) -> m (Maybe b, [e])
+        apply = liftA2 $ \ (mf,e1) (ma,e2) -> (mf <*> ma, e1 <> e2)
 
 instance (Monad m) => Monad (ErrorfulT e m) where
     ErrorfulT m >>= k = ErrorfulT $ do
-        m' <- m
-        case m' of
-            Right (a,es) -> runErrorfulT (k a)
-            Left e       -> pure (Left e)
+        (a,es) <- m
+        case a of
+            Just x      -> runErrorfulT (k x)
+            Nothing     -> pure (Nothing, es)
 
-mapErrors :: (Monad m) => (e -> e') -> ErrorfulT e m a -> ErrorfulT e' m a
-mapErrors f m = ErrorfulT $ do
-    x <- runErrorfulT m
-    case x of
-        Left  e      -> pure . Left  $ f e
-        Right (a,es) -> pure . Right $ (a, f <$> es)
+mapErrorful :: (Functor m) => (e -> e') -> ErrorfulT e m a -> ErrorfulT e' m a
+mapErrorful f (ErrorfulT m) = ErrorfulT $
+    m & mapped . _2 . mapped %~ f
+
+-- when microlens-pro drops we can write this as
+--    mapErrorful f = coerced . mapped . _2 . mapped %~ f
+-- lol
+
+--------------------------------------------------------------------------------
+-- daily dose of n^2 instances
+
+instance (Monad m, MonadErrorful e m) => MonadErrorful e (ReaderT r m) where
+    addWound = lift . addWound
+    addFatal = lift . addFatal
 

src/Core/Examples.hs

@@ -4,17 +4,19 @@ Description : Core examples (may eventually be unit tests)
 -}
 {-# LANGUAGE QuasiQuotes #-}
 {-# LANGUAGE OverloadedStrings #-}
-module Core.Examples
-    ( fac3
-    , sumList
-    , constDivZero
-    , idCase
-    ) where
+module Core.Examples where
 ----------------------------------------------------------------------------------
 import Core.Syntax
 import Core.TH
 ----------------------------------------------------------------------------------
 
+-- fac3 = undefined
+-- sumList = undefined
+-- constDivZero = undefined
+-- idCase = undefined
+
+---
+
 letrecExample :: Program'
 letrecExample = [coreProg|
     pair x y f = f x y;
@@ -140,8 +142,8 @@ simple1 = [coreProg|
 caseBool1 :: Program'
 caseBool1 = [coreProg|
     _if c x y = case c of
-        { 1 -> x
-        ; 0 -> y
+        { <1> -> x
+        ; <0> -> y
         };
 
     false = Pack{0 0};
@@ -153,8 +155,8 @@ caseBool1 = [coreProg|
 fac3 :: Program'
 fac3 = [coreProg|
     fac n = case (==#) n 0 of
-        { 1 -> 1
-        ; 0 -> (*#) n (fac ((-#) n 1))
+        { <1> -> 1
+        ; <0> -> (*#) n (fac ((-#) n 1))
         };
 
     main = fac 3;
@@ -168,8 +170,8 @@ sumList = [coreProg|
         cons x y = Pack{1 2} x y;
         list = cons 1 (cons 2 (cons 3 nil));
         sum l = case l of
-            { 0      -> 0
-            ; 1 x xs -> (+#) x (sum xs)
+            { <0>      -> 0
+            ; <1> x xs -> (+#) x (sum xs)
             };
         main = sum list;
     |]
@@ -185,10 +187,36 @@ idCase = [coreProg|
         id x = x;
 
         main = id (case Pack{1 0} of
-            { 1 -> (+#) 2 3
+            { <1> -> (+#) 2 3
             })
     |]
 
+-- NOTE: the GM primitive (==#) returns an untyped constructor with tag 1 for
+-- true, and 0 for false. See: GM.boxBool
+namedBoolCase :: Program'
+namedBoolCase = [coreProg|
+        {-# PackData True 1 0 #-}
+        {-# PackData False 0 0 #-}
+        main = case (==#) 1 1 of
+            { True -> 123
+            ; False -> 456
+            }
+    |]
+
+namedConsCase :: Program'
+namedConsCase = [coreProg|
+        {-# PackData Nil  0 0 #-}
+        {-# PackData Cons 1 2 #-}
+        Nil = Pack{0 0};
+        Cons = Pack{1 2};
+        foldr f z l = case l of
+            { Nil       -> z
+            ; Cons x xs -> f x (foldr f z xs)
+            };
+        list = Cons 1 (Cons 2 (Cons 3 Nil));
+        main = foldr (+#) 0 list
+    |]
+
 -- corePrelude :: Module Name
 -- corePrelude = Module (Just ("Prelude", [])) $
 --     -- non-primitive defs
@@ -216,3 +244,4 @@ idCase = [coreProg|
 --         , ScDef "Cons" [] $ Con 2 2
 --         ]
 
+--}

src/Core/HindleyMilner.hs

@@ -3,6 +3,7 @@ Module      : Core.HindleyMilner
 Description : Hindley-Milner type system
 -}
 {-# LANGUAGE LambdaCase #-}
+{-# LANGUAGE OverloadedStrings #-}
 module Core.HindleyMilner
     ( Context'
     , infer
@@ -16,15 +17,17 @@ module Core.HindleyMilner
 ----------------------------------------------------------------------------------
 import Lens.Micro
 import Lens.Micro.Mtl
+import Lens.Micro.Platform
 import Data.Maybe               (fromMaybe)
 import Data.Text                qualified as T
 import Data.HashMap.Strict      qualified as H
 import Data.Foldable            (traverse_)
 import Compiler.RLPC
-import Control.Monad            (foldM, void)
+import Control.Monad            (foldM, void, forM)
 import Control.Monad.Errorful   (Errorful, addFatal)
 import Control.Monad.State
 import Control.Monad.Utils      (mapAccumLM)
+import Text.Printf
 import Core.Syntax
 ----------------------------------------------------------------------------------
 
@@ -48,9 +51,23 @@ data TypeError
     | TyErrMissingTypeSig Name
     deriving (Show, Eq)
 
--- TODO:
 instance IsRlpcError TypeError where
-    liftRlpcErr = RlpcErr . show
+    liftRlpcError = \case
+        -- todo: use anti-parser instead of show
+        TyErrCouldNotUnify t u -> Text
+            [ T.pack $ printf "Could not match type `%s` with `%s`."
+                              (show t) (show u)
+            , "Expected: " <> tshow t
+            , "Got: " <> tshow u
+            ]
+        TyErrUntypedVariable n -> Text
+            [ "Untyped (likely undefined) variable `" <> n <> "`"
+            ]
+        TyErrRecursiveType t x -> Text
+            [ T.pack $ printf "recursive type error lol"
+            ]
+
+        where tshow = T.pack . show
 
 -- | Synonym for @Errorful [TypeError]@. This means an @HMError@ action may
 -- throw any number of fatal or nonfatal errors. Run with @runErrorful@.
@@ -88,10 +105,10 @@ checkCoreProg p = scDefs
              where scname = sc ^. _lhs._1
 
 -- | @checkCoreProgR p@ returns @p@ if @p@ successfully typechecks.
-checkCoreProgR :: Program' -> RLPC RlpcError Program'
-checkCoreProgR p = do
-    liftRlpcErrs . rlpc . checkCoreProg $ p
-    pure p
+checkCoreProgR :: (Applicative m) => Program' -> RLPCT m Program'
+checkCoreProgR p = undefined
+
+{-# WARNING checkCoreProgR "unimpl" #-}
 
 -- | Infer the type of an expression under some context.
 --
@@ -140,7 +157,32 @@ gather = \g e -> runStateT (go g e) ([],0) <&> \ (t,(cs,_)) -> (t,cs) where
         Let NonRec bs e -> do
             g' <- buildLetContext g bs
             go g' e
-        -- TODO letrec, lambda, case
+        Let Rec bs e -> do
+            g' <- buildLetrecContext g bs
+            go g' e
+        Lam bs e -> case bs of
+            [x] -> do
+                tx <- uniqueVar
+                let g' = (x,tx) : g
+                te <- go g' e
+                pure (tx :-> te)
+        -- TODO lambda, case
+
+    buildLetrecContext :: Context' -> [Binding']
+                       -> StateT ([Constraint], Int) HMError Context'
+    buildLetrecContext g bs = do
+        let f ag (k := _) = do
+                n <- uniqueVar
+                pure ((k,n) : ag)
+        rg <- foldM f g bs
+        let k ag (k := v) = do
+                t <- go rg v
+                pure ((k,t) : ag)
+        foldM k g bs
+
+    -- | augment a context with the inferred types of each binder. the returned
+    -- context is linearly accumulated, meaning that the context used to infer each binder
+    -- will include the inferred types of all previous binder
 
     buildLetContext :: Context' -> [Binding']
                     -> StateT ([Constraint], Int) HMError Context'
@@ -218,3 +260,20 @@ subst x t (TyVar y) | x == y  = t
 subst x t (a :-> b)           = subst x t a :-> subst x t b
 subst _ _ e                   = e
 
+--------------------------------------------------------------------------------
+
+demoContext :: Context'
+demoContext =
+    [ ("fix", (TyVar "a" :-> TyVar "a") :-> TyVar "a")
+    , ("add", TyInt :-> TyInt :-> TyInt)
+    , ("==", TyInt :-> TyInt :-> TyCon "Bool")
+    , ("True", TyCon "Bool")
+    , ("False", TyCon "Bool")
+    ]
+
+pprintType :: Type -> String
+pprintType (s :-> t) = "(" <> pprintType s <> " -> " <> pprintType t <> ")"
+pprintType TyFun = "(->)"
+pprintType (TyVar x) = x ^. unpacked
+pprintType (TyCon t) = t ^. unpacked
+

src/Core/Lex.x

@@ -20,9 +20,11 @@ import Debug.Trace
 import Data.Text            (Text)
 import Data.Text            qualified as T
 import Data.String          (IsString(..))
+import Data.Functor.Identity
 import Core.Syntax
 import Compiler.RLPC
-import Compiler.RlpcError
+-- TODO: unify Located definitions
+import Compiler.RlpcError   hiding (Located(..))
 import Lens.Micro
 import Lens.Micro.TH
 }
@@ -65,6 +67,8 @@ $white_no_nl = $white # $nl
 
 @decimal = $digit+
 
+@alttag    = "<" $digit+ ">"
+
 rlp :-
 
 <0>
@@ -92,6 +96,8 @@ rlp :-
     "="                     { constTok TokenEquals }
     "->"                    { constTok TokenArrow }
 
+    @alttag                 { lexWith ( TokenAltTag . read @Int . T.unpack
+                                      . T.drop 1 . T.init ) }
     @varname                { lexWith TokenVarName }
     @conname                { lexWith TokenConName }
     @varsym                 { lexWith TokenVarSym }
@@ -135,6 +141,7 @@ data CoreToken = TokenLet
                | TokenConName Name
                | TokenVarSym Name
                | TokenConSym Name
+               | TokenAltTag Tag
                | TokenEquals
                | TokenLParen
                | TokenRParen
@@ -167,24 +174,22 @@ lexWith :: (Text -> CoreToken) -> Lexer
 lexWith f (AlexPn _ y x,_,_,s) l = pure $ Located y x l (f $ T.take l s)
 
 -- | The main lexer driver.
-lexCore :: Text -> RLPC SrcError [Located CoreToken]
+lexCore :: Text -> RLPC [Located CoreToken]
 lexCore s = case m of
-    Left e   -> addFatal err
-        where err = SrcError
-                { _errSpan       = (0,0,0) -- TODO: location
-                , _errSeverity   = Error
-                , _errDiagnostic = SrcErrLexical e
-                }
+    Left e   -> error "core lex error"
     Right ts -> pure ts
     where
         m = runAlex s lexStream
 
-lexCoreR :: Text -> RLPC RlpcError [Located CoreToken]
-lexCoreR = liftRlpcErrs . lexCore
+lexCoreR :: forall m. (Applicative m) => Text -> RLPCT m [Located CoreToken]
+lexCoreR = hoistRlpcT generalise . lexCore
+    where
+        generalise :: forall a. Identity a -> m a
+        generalise (Identity a) = pure a
 
 -- | @lexCore@, but the tokens are stripped of location info. Useful for
 -- debugging
-lexCore' :: Text -> RLPC SrcError [CoreToken]
+lexCore' :: Text -> RLPC [CoreToken]
 lexCore' s = fmap f <$> lexCore s
     where f (Located _ _ _ t) = t
 
@@ -201,11 +206,11 @@ data ParseError = ParErrLexical String
 
 -- TODO:
 instance IsRlpcError SrcError where
-    liftRlpcErr = RlpcErr . show
+    liftRlpcError = Text . pure . T.pack . show
 
 -- TODO:
 instance IsRlpcError ParseError where
-    liftRlpcErr = RlpcErr . show
+    liftRlpcError = Text . pure . T.pack . show
 
 alexEOF :: Alex (Located CoreToken)
 alexEOF = Alex $ \ st@(AlexState { alex_pos = AlexPn _ y x }) ->

src/Core/Parse.y

@@ -3,14 +3,13 @@
 Module : Core.Parse
 Description : Parser for the Core language
 -}
-{-# LANGUAGE OverloadedStrings #-}
+{-# LANGUAGE OverloadedStrings, ViewPatterns #-}
 module Core.Parse
     ( parseCore
     , parseCoreExpr
     , parseCoreProg
     , parseCoreProgR
     , module Core.Lex -- temp convenience
-    , parseTmp
     , SrcError
     , Module
     )
@@ -18,13 +17,17 @@ module Core.Parse
 
 import Control.Monad        ((>=>))
 import Data.Foldable        (foldl')
+import Data.Functor.Identity
 import Core.Syntax
 import Core.Lex
 import Compiler.RLPC
+import Control.Monad
 import Lens.Micro
 import Data.Default.Class   (def)
 import Data.Hashable        (Hashable)
+import Data.List.Extra
 import Data.Text.IO         qualified as TIO
+import Data.Text            (Text)
 import Data.Text            qualified as T
 import Data.HashMap.Strict  qualified as H
 }
@@ -34,7 +37,7 @@ import Data.HashMap.Strict  qualified as H
 %name parseCoreProg StandaloneProgram
 %tokentype { Located CoreToken }
 %error { parseError }
-%monad { RLPC SrcError }
+%monad { RLPC } { happyBind } { happyPure }
 
 %token
       let             { Located _ _ _ TokenLet }
@@ -50,6 +53,7 @@ import Data.HashMap.Strict  qualified as H
       varsym          { Located _ _ _ (TokenVarSym $$) }
       conname         { Located _ _ _ (TokenConName $$) }
       consym          { Located _ _ _ (TokenConSym $$) }
+      alttag          { Located _ _ _ (TokenAltTag $$) }
       word            { Located _ _ _ (TokenWord $$) }
       'λ'             { Located _ _ _ TokenLambda }
       '->'            { Located _ _ _ TokenArrow }
@@ -83,6 +87,15 @@ Program         : ScTypeSig ';' Program         { insTypeSig $1 $3 }
                 | ScTypeSig OptSemi             { singletonTypeSig $1 }
                 | ScDef     ';' Program         { insScDef $1 $3 }
                 | ScDef     OptSemi             { singletonScDef $1 }
+                | TLPragma  Program             {% doTLPragma $1 $2 }
+                | TLPragma                      {% doTLPragma $1 mempty }
+
+TLPragma        :: { Pragma }
+                : '{-#' Words '#-}'             { Pragma $2 }
+
+Words           :: { [Text] }
+                : Words word                    { $1 `snoc` $2 }
+                | word                          { [$1] }
 
 OptSemi         :: { () }
 OptSemi         : ';'                           { () }
@@ -95,10 +108,11 @@ ScDefs          :: { [ScDef Name] }
 ScDefs          : ScDef ';' ScDefs              { $1 : $3 }
                 | ScDef ';'                     { [$1] }
                 | ScDef                         { [$1] }
-                | {- epsilon -}                 { [] }
 
 ScDef           :: { ScDef Name }
 ScDef           : Var ParList '=' Expr          { ScDef $1 $2 $4 }
+                -- hack to allow constructors to be compiled into scs
+                | Con ParList '=' Expr          { ScDef $1 $2 $4 }
 
 Type            :: { Type }
 Type            : Type1                         { $1 }
@@ -148,22 +162,15 @@ Alters          : Alter ';' Alters              { $1 : $3 }
                 | Alter                         { [$1] }
 
 Alter           :: { Alter Name }
-Alter           : litint ParList '->' Expr      { Alter (AltData $1) $2 $4 }
+Alter           : alttag ParList '->' Expr      { Alter (AltTag  $1) $2 $4 }
+                | Con    ParList '->' Expr      { Alter (AltData $1) $2 $4 }
 
 Expr1           :: { Expr Name }
 Expr1           : litint                        { Lit $ IntL $1 }
                 | Id                            { Var $1 }
                 | PackCon                       { $1 }
-                | ExprPragma                    { $1 }
                 | '(' Expr ')'                  { $2 }
 
-ExprPragma      :: { Expr Name }
-ExprPragma      : '{-#' Words '#-}'      {% exprPragma $2 }
-
-Words           :: { [String] }
-Words           : word Words                    { T.unpack $1 : $2 }
-                | word                          { [T.unpack $1] }
-
 PackCon         :: { Expr Name }
 PackCon         : pack '{' litint litint '}'    { Con $3 $4 }
 
@@ -189,34 +196,23 @@ Con             : '(' consym ')'                    { $2 }
 
 {
 
-parseError :: [Located CoreToken] -> RLPC SrcError a
-parseError (Located y x l _ : _) = addFatal err
-    where err = SrcError
-            { _errSpan       = (y,x,l)
-            , _errSeverity   = Error
-            , _errDiagnostic = SrcErrParse
-            }
+parseError :: [Located CoreToken] -> RLPC a
+parseError (Located y x l t : _) =
+    error $ show y <> ":" <> show x
+         <> ": parse error at token `" <> show t <> "'"
 
-parseTmp :: IO (Module Name)
-parseTmp = do
-    s <- TIO.readFile "/tmp/t.hs"
-    case parse s of
-        Left e -> error (show e)
-        Right (ts,_) -> pure ts
-    where
-        parse = evalRLPC def . (lexCore >=> parseCore)
+{-# WARNING parseError "unimpl" #-}
 
-exprPragma :: [String] -> RLPC SrcError (Expr Name)
-exprPragma ("AST" : e) = astPragma e
-exprPragma _           = addFatal err
-    where err = SrcError
-            { _errSpan       = (0,0,0) -- TODO: span
-            , _errSeverity   = Warning
-            , _errDiagnostic = SrcErrUnknownPragma "" -- TODO: missing pragma
-            }
+exprPragma :: [String] -> RLPC (Expr Name)
+exprPragma ("AST" : e) = undefined
+exprPragma _           = undefined
 
-astPragma :: [String] -> RLPC SrcError (Expr Name)
-astPragma = pure . read . unwords
+{-# WARNING exprPragma "unimpl" #-}
+
+astPragma :: [String] -> RLPC (Expr Name)
+astPragma _ = undefined
+
+{-# WARNING astPragma "unimpl" #-}
 
 insTypeSig :: (Hashable b) => (b, Type) -> Program b -> Program b
 insTypeSig ts = programTypeSigs %~ uncurry H.insert ts
@@ -230,8 +226,34 @@ insScDef sc = programScDefs %~ (sc:)
 singletonScDef :: (Hashable b) => ScDef b -> Program b
 singletonScDef sc = insScDef sc mempty
 
-parseCoreProgR :: [Located CoreToken] -> RLPC RlpcError Program'
-parseCoreProgR = liftRlpcErrs . parseCoreProg
+parseCoreProgR :: forall m. (Monad m) => [Located CoreToken] -> RLPCT m Program'
+parseCoreProgR = ddumpast <=< (hoistRlpcT generalise . parseCoreProg)
+    where
+        generalise :: forall a. Identity a -> m a
+        generalise (Identity a) = pure a
+
+        ddumpast :: Program' -> RLPCT m Program'
+        ddumpast p = do
+            addDebugMsg "dump-ast" . show $ p
+            pure p
+
+happyBind :: RLPC a -> (a -> RLPC b) -> RLPC b
+happyBind m k = m >>= k
+
+happyPure :: a -> RLPC a
+happyPure a = pure a
+
+doTLPragma :: Pragma -> Program' -> RLPC Program'
+-- TODO: warn unrecognised pragma
+doTLPragma (Pragma []) p = pure p
+
+doTLPragma (Pragma pr) p = case pr of
+    -- TODO: warn on overwrite
+    ["PackData", n, readt -> t, readt -> a] ->
+        pure $ p & programDataTags . at n ?~ (t,a)
+
+readt :: (Read a) => Text -> a
+readt = read . T.unpack
 
 }
 

src/Core/Syntax.hs

@@ -5,8 +5,14 @@ Description : Core ASTs and the like
 {-# LANGUAGE PatternSynonyms, OverloadedStrings #-}
 {-# LANGUAGE FunctionalDependencies #-}
 {-# LANGUAGE TemplateHaskell #-}
+{-# LANGUAGE DerivingStrategies, DerivingVia #-}
+-- for recursion-schemes
+{-# LANGUAGE DeriveFunctor, DeriveFoldable, DeriveTraversable
+  , TemplateHaskell, TypeFamilies #-}
 module Core.Syntax
     ( Expr(..)
+    , ExprF(..)
+    , ExprF'(..)
     , Type(..)
     , pattern TyInt
     , Lit(..)
@@ -24,9 +30,11 @@ module Core.Syntax
     , Module(..)
     , Program(..)
     , Program'
+    , Pragma(..)
     , unliftScDef
     , programScDefs
     , programTypeSigs
+    , programDataTags
     , Expr'
     , ScDef'
     , Alter'
@@ -40,11 +48,15 @@ import Data.Coerce
 import Data.Pretty
 import Data.List                    (intersperse)
 import Data.Function                ((&))
+import Data.Functor.Foldable
+import Data.Functor.Foldable.TH     (makeBaseFunctor)
 import Data.String
+import Data.HashMap.Strict          (HashMap)
 import Data.HashMap.Strict          qualified as H
 import Data.Hashable
 import Data.Text                    qualified as T
 import Data.Char
+import GHC.Generics
 -- Lift instances for the Core quasiquoters
 import Language.Haskell.TH.Syntax   (Lift)
 import Lens.Micro.TH                (makeLenses)
@@ -91,7 +103,7 @@ data Binding b = Binding b (Expr b)
 deriving instance (Eq b) => Eq (Binding b)
 
 infixl 1 :=
-pattern (:=) :: b -> (Expr b) -> (Binding b)
+pattern (:=) :: b -> Expr b -> Binding b
 pattern k := v = Binding k v
 
 data Alter b = Alter AltCon [b] (Expr b)
@@ -99,16 +111,19 @@ data Alter b = Alter AltCon [b] (Expr b)
 
 deriving instance (Eq b) => Eq (Alter b)
 
+newtype Pragma = Pragma [T.Text]
+
 data Rec = Rec
          | NonRec
          deriving (Show, Read, Eq, Lift)
 
-data AltCon = AltData Tag
+data AltCon = AltData Name
+            | AltTag Tag
             | AltLit Lit
             | Default
             deriving (Show, Read, Eq, Lift)
 
-data Lit = IntL Int
+newtype Lit = IntL Int
     deriving (Show, Read, Eq, Lift)
 
 type Name = T.Text
@@ -125,13 +140,20 @@ data Module b = Module (Maybe (Name, [Name])) (Program b)
 
 data Program b = Program
     { _programScDefs   :: [ScDef b]
-    , _programTypeSigs :: H.HashMap b Type
+    , _programTypeSigs :: HashMap b Type
+    -- map constructors to their tag and arity
+    , _programDataTags :: HashMap b (Tag, Int)
     }
-    deriving (Show, Lift)
+    deriving (Show, Lift, Generic)
+    deriving (Semigroup, Monoid)
+        via Generically (Program b)
 
 makeLenses ''Program
+makeBaseFunctor ''Expr
 pure []
 
+type ExprF' = ExprF Name
+
 type Program' = Program Name
 type Expr' = Expr Name
 type ScDef' = ScDef Name
@@ -148,12 +170,6 @@ instance IsString Type where
         | otherwise     = TyVar . fromString $ s
         where (c:_) = s
 
-instance (Hashable b) => Semigroup (Program b) where
-    (<>) = undefined
-
-instance (Hashable b) => Monoid (Program b) where
-    mempty = Program mempty mempty
-
 ----------------------------------------------------------------------------------
 
 class HasRHS s t a b | s -> a, t -> b, s b -> t, t a -> s where
@@ -185,5 +201,10 @@ instance HasLHS (Alter b) (Alter b) (AltCon, [b]) (AltCon, [b]) where
 instance HasLHS (ScDef b) (ScDef b) (b, [b]) (b, [b]) where
     _lhs = lens
         (\ (ScDef n as _) -> (n,as))
-        (\ (ScDef _ _ e) (n',as') -> (ScDef n' as' e))
+        (\ (ScDef _ _ e) (n',as') -> ScDef n' as' e)
+
+instance HasLHS (Binding b) (Binding b) b b where
+    _lhs = lens
+        (\ (k := _) -> k)
+        (\ (_ := e) k' -> k' := e)
 

src/Core/TH.hs

@@ -6,7 +6,6 @@ module Core.TH
     ( coreExpr
     , coreProg
     , coreProgT
-    , core
     )
     where
 ----------------------------------------------------------------------------------
@@ -14,74 +13,38 @@ import Language.Haskell.TH
 import Language.Haskell.TH.Syntax   hiding (Module)
 import Language.Haskell.TH.Quote
 import Control.Monad                ((>=>))
+import Control.Monad.IO.Class
+import Control.Arrow                ((>>>))
 import Compiler.RLPC
 import Data.Default.Class           (def)
+import Data.Text                    (Text)
 import Data.Text                    qualified as T
 import Core.Parse
 import Core.Lex
+import Core.Syntax
 import Core.HindleyMilner           (checkCoreProgR)
 ----------------------------------------------------------------------------------
 
--- TODO: write in terms of a String -> QuasiQuoter
-
-core :: QuasiQuoter
-core = QuasiQuoter
-    { quoteExp = qCore
-    , quotePat = error "core quasiquotes may only be used in expressions"
-    , quoteType = error "core quasiquotes may only be used in expressions"
-    , quoteDec = error "core quasiquotes may only be used in expressions"
-    }
-
 coreProg :: QuasiQuoter
-coreProg = QuasiQuoter
-    { quoteExp = qCoreProg
-    , quotePat = error "core quasiquotes may only be used in expressions"
-    , quoteType = error "core quasiquotes may only be used in expressions"
-    , quoteDec = error "core quasiquotes may only be used in expressions"
-    }
+coreProg = mkqq $ lexCoreR >=> parseCoreProgR
 
 coreExpr :: QuasiQuoter
-coreExpr = QuasiQuoter
-    { quoteExp = qCoreExpr
-    , quotePat = error "core quasiquotes may only be used in expressions"
-    , quoteType = error "core quasiquotes may only be used in expressions"
-    , quoteDec = error "core quasiquotes may only be used in expressions"
-    }
+coreExpr = mkqq $ lexCoreR >=> parseCoreExpr
 
 -- | Type-checked @coreProg@
 coreProgT :: QuasiQuoter
-coreProgT = QuasiQuoter
-    { quoteExp = qCoreProgT
+coreProgT = mkqq $ lexCoreR >=> parseCoreProgR >=> checkCoreProgR
+
+mkqq :: (Lift a) => (Text -> RLPC a) -> QuasiQuoter
+mkqq p = QuasiQuoter
+    { quoteExp = mkq p
     , quotePat = error "core quasiquotes may only be used in expressions"
     , quoteType = error "core quasiquotes may only be used in expressions"
     , quoteDec = error "core quasiquotes may only be used in expressions"
     }
 
-qCore :: String -> Q Exp
-qCore s = case parse (T.pack s) of
-    Left e       -> error (show e)
-    Right (m,ts) -> lift m
-    where
-        parse = evalRLPC def . (lexCore >=> parseCore)
-
-qCoreExpr :: String -> Q Exp
-qCoreExpr s = case parseExpr (T.pack s) of
-    Left e       -> error (show e)
-    Right (m,ts) -> lift m
-    where
-        parseExpr = evalRLPC def . (lexCore >=> parseCoreExpr)
-
-qCoreProg :: String -> Q Exp
-qCoreProg s = case parse (T.pack s) of
-    Left e       -> error (show e)
-    Right (m,ts) -> lift m
-    where
-        parse = evalRLPC def . (lexCoreR >=> parseCoreProgR)
-
-qCoreProgT :: String -> Q Exp
-qCoreProgT s = case parse (T.pack s) of
-    Left e      -> error (show e)
-    Right (m,_) -> lift m
-    where
-        parse = evalRLPC def . (lexCoreR >=> parseCoreProgR >=> checkCoreProgR)
+mkq :: (Lift a) => (Text -> RLPC a) -> String -> Q Exp
+mkq parse s = case evalRLPC def (parse $ T.pack s) of
+    (Just a,  _) -> lift a
+    (Nothing, _) -> error "todo: aaahhbbhjhbdjhabsjh"
 

src/Core/Utils.hs

@@ -1,16 +1,10 @@
--- for recursion schemes
-{-# LANGUAGE DeriveFunctor, DeriveFoldable, DeriveTraversable #-}
--- for recursion schemes
-{-# LANGUAGE TemplateHaskell, TypeFamilies #-}
-
 module Core.Utils
-    ( bindersOf
-    , rhssOf
+    ( programRhss
+    , programGlobals
     , isAtomic
     -- , insertModule
     , extractProgram
     , freeVariables
-    , ExprF(..)
     )
     where
 ----------------------------------------------------------------------------------
@@ -23,13 +17,11 @@ import Lens.Micro
 import GHC.Exts                     (IsList(..))
 ----------------------------------------------------------------------------------
 
-bindersOf :: (IsList l, Item l ~ b) => [Binding b] -> l
-bindersOf bs = fromList $ fmap f bs
-    where f (k := _) = k
+programGlobals :: Traversal' (Program b) b
+programGlobals = programScDefs . each . _lhs . _1
 
-rhssOf :: (IsList l, Item l ~ Expr b) => [Binding b] -> l
-rhssOf = fromList . fmap f
-    where f (_ := v) = v
+programRhss :: Traversal' (Program b) (Expr b)
+programRhss = programScDefs . each . _rhs
 
 isAtomic :: Expr b -> Bool
 isAtomic (Var _)  = True
@@ -47,8 +39,6 @@ extractProgram (Module _ p) = p
 
 ----------------------------------------------------------------------------------
 
-makeBaseFunctor ''Expr
-
 freeVariables :: Expr' -> Set Name
 freeVariables = cata go
     where
@@ -57,8 +47,8 @@ freeVariables = cata go
         -- TODO: collect free vars in rhss of bs
         go (LetF _ bs e)    = (e `S.union` esFree) `S.difference` ns
             where
-                es = rhssOf bs :: [Expr']
-                ns = bindersOf bs
+                es = bs ^.. each . _rhs :: [Expr']
+                ns = S.fromList $ bs ^.. each . _lhs
                 -- TODO: this feels a little wrong. maybe a different scheme is
                 -- appropriate
                 esFree = foldMap id $ freeVariables <$> es

src/Core2Core.hs

@@ -1,3 +1,4 @@
+{-# LANGUAGE ImplicitParams #-}
 {-# LANGUAGE LambdaCase #-}
 module Core2Core
     ( core2core
@@ -15,11 +16,12 @@ import Data.Set                 (Set)
 import Data.Set                 qualified as S
 import Data.List
 import Control.Monad.Writer
-import Control.Monad.State
+import Control.Monad.State.Lazy
 import Control.Arrow            ((>>>))
 import Data.Text                qualified as T
+import Data.HashMap.Strict      (HashMap)
 import Numeric                  (showHex)
-import Lens.Micro
+import Lens.Micro.Platform
 import Core.Syntax
 import Core.Utils
 ----------------------------------------------------------------------------------
@@ -28,22 +30,35 @@ core2core :: Program' -> Program'
 core2core p = undefined
 
 gmPrep :: Program' -> Program'
-gmPrep p = p' & programScDefs %~ (<>caseScs)
+gmPrep p = p & appFloater (floatNonStrictCases globals)
+             & tagData
     where
-        rhss :: Applicative f => (Expr z -> f (Expr z)) -> Program z -> f (Program z)
-        rhss = programScDefs . each . _rhs
         globals = p ^.. programScDefs . each . _lhs . _1
                 & S.fromList
 
-        -- i kinda don't like that we're calling floatNonStrictCases twice tbh
-        p'      = p & rhss %~ fst . runFloater . floatNonStrictCases globals
-        caseScs = (p ^.. rhss)
-              <&> snd . runFloater . floatNonStrictCases globals
-                & mconcat
+tagData :: Program' -> Program'
+tagData p = let ?dt = p ^. programDataTags
+            in p & programRhss %~ cata go where
+    go :: (?dt :: HashMap Name (Tag, Int)) => ExprF' Expr' -> Expr'
+    go (CaseF  e as)    = Case e (tagAlts <$> as)
+    go x                = embed x
+
+    tagAlts :: (?dt :: HashMap Name (Tag, Int)) => Alter' -> Alter'
+    tagAlts (Alter (AltData c) bs e) = Alter (AltTag tag) bs e
+        where tag = case ?dt ^. at c of
+                Just (t,_) -> t
+                -- TODO: errorful
+                Nothing    -> error $ "unknown constructor " <> show c
+    tagAlts x = x
 
 -- | Auxilary type used in @floatNonSrictCases@
 type Floater = StateT [Name] (Writer [ScDef'])
 
+appFloater :: (Expr' -> Floater Expr') -> Program' -> Program'
+appFloater fl p = p & traverseOf programRhss fl
+                    & runFloater
+                    & \ (me,floats) -> me & programScDefs %~ (<>floats)
+
 runFloater :: Floater a -> (a, [ScDef'])
 runFloater = flip evalStateT ns >>> runWriter
     where

src/GM.hs

@@ -8,6 +8,7 @@ Description : The G-Machine
 module GM
     ( hdbgProg
     , evalProg
+    , evalProgR
     , Node(..)
     , gmEvalProg
     , finalStateOf
@@ -34,6 +35,7 @@ import System.IO                    (Handle, hPutStrLn)
 import Data.String                  (IsString)
 import Data.Heap
 import Debug.Trace
+import Compiler.RLPC
 import Core2Core
 import Core
 ----------------------------------------------------------------------------------
@@ -156,6 +158,21 @@ hdbgProg p hio = do
         [resAddr] = final ^. gmStack
         res = hLookupUnsafe resAddr h
 
+evalProgR :: (Monad m) => Program' -> RLPCT m (Node, Stats)
+evalProgR p = do
+    (renderOut . showState) `traverse_` states
+    renderOut . showStats $ sts
+    pure (res, sts)
+    where
+        renderOut r = addDebugMsg "dump-eval" $ render r ++ "\n"
+        states = eval . compile $ p
+        final = last states
+
+        sts = final ^. gmStats
+        -- the address of the result should be the one and only stack entry
+        [resAddr] = final ^. gmStack
+        res = hLookupUnsafe resAddr (final ^. gmHeap)
+
 eval :: GmState -> [GmState]
 eval st = st : rest
     where
@@ -661,7 +678,8 @@ buildInitialHeap (view programScDefs -> ss) = mapAccumL allocateSc mempty compil
         compileC _ (Con t n) = [PushConstr t n]
 
         compileC _ (Case _ _) =
-            error "case expressions may not appear in non-strict contexts :/"
+            error "GM compiler found a non-strict case expression, which should\
+                  \ have been floated by Core2Core.gmPrep. This is a bug!"
 
         compileC _ _ = error "yet to be implemented!"
 
@@ -724,12 +742,16 @@ buildInitialHeap (view programScDefs -> ss) = mapAccumL allocateSc mempty compil
         compileD g as = fmap (compileA g) as
 
         compileA :: Env -> Alter' -> (Tag, Code)
-        compileA g (Alter (AltData t) as e) = (t, [Split n] <> c <> [Slide n])
+        compileA g (Alter (AltTag t) as e) = (t, [Split n] <> c <> [Slide n])
             where
                 n = length as
                 binds = (NameKey <$> as) `zip` [0..]
                 g' = binds ++ argOffset n g
                 c = compileE g' e
+        compileA _ (Alter _ as e) = error "GM.compileA found an untagged\
+                                          \ constructor, which should have\
+                                          \ been handled by Core2Core.gmPrep.\
+                                          \ This is a bug!"
 
         inlineOp1 :: Env -> Instr -> Expr' -> Code
         inlineOp1 g i a = compileE g a <> [i]

src/Rlp/Lex.x

@@ -7,12 +7,17 @@ module Rlp.Lex
     , RlpToken(..)
     , Located(..)
     , lexToken
+    , lexStream
     , lexDebug
     , lexCont
+    , popLexState
+    , programInitState
+    , runP'
     )
     where
 import Codec.Binary.UTF8.String (encodeChar)
 import Control.Monad
+import Control.Monad.Errorful
 import Core.Syntax              (Name)
 import Data.Functor.Identity
 import Data.Char                (digitToInt)
@@ -52,9 +57,10 @@ $asciisym       = [\!\#\$\%\&\*\+\.\/\<\=\>\?\@\\\^\|\-\~\:]
 
 @reservedname = 
     case|data|do|import|in|let|letrec|module|of|where
+    |infixr|infixl|infix
 
 @reservedop =
-    "=" | \\ | "->" | "|"
+    "=" | \\ | "->" | "|" | "::"
 
 rlp :-
     
@@ -70,6 +76,17 @@ $white_no_nl+       ;
 -- for the definition of `doBol`
 <0> \n                  { beginPush bol }
 
+<layout>
+{
+    
+}
+
+-- layout keywords
+<0>
+{
+    "let"               { constToken TokenLet `thenBeginPush` layout_let }
+}
+
 -- scan various identifiers and reserved words. order is important here!
 <0>
 {
@@ -107,6 +124,14 @@ $white_no_nl+       ;
     ()                  { doBol }
 }
 
+<layout_let>
+{
+    \n                  { beginPush bol }
+    "{"                 { explicitLBrace }
+    "in"                { constToken TokenIn `thenDo` (popLexState *> popLayout) }
+    ()                  { doLayout }
+}
+
 <layout_top>
 {
     \n                  ;
@@ -123,6 +148,9 @@ lexReservedName = \case
     "of"        -> TokenOf
     "let"       -> TokenLet
     "in"        -> TokenIn
+    "infix"     -> TokenInfix
+    "infixl"    -> TokenInfixL
+    "infixr"    -> TokenInfixR
 
 lexReservedOp :: Text -> RlpToken
 lexReservedOp = \case
@@ -138,6 +166,12 @@ thenBegin act c inp l = do
     psLexState . _head .= c
     pure a
 
+thenBeginPush :: LexerAction a -> Int -> LexerAction a
+thenBeginPush act c inp l = do
+    a <- act inp l
+    pushLexState c
+    pure a
+
 andBegin :: LexerAction a -> Int -> LexerAction a
 andBegin act c inp l = do
     psLexState . _head .= c
@@ -158,10 +192,10 @@ alexGetByte inp = case inp ^. aiBytes of
                        -- report the previous char
                        & aiPrevChar .~ c
                        -- update the position
-                       & aiPos %~ \ (ln,col) ->
+                       & aiPos %~ \ (ln,col,a) ->
                                    if c == '\n'
-                                   then (ln+1,1)
-                                   else (ln,col+1)
+                                   then (ln+1, 1, a+1)
+                                   else (ln, col+1, a+1)
         pure (b, inp')
     
     _ -> Just (head bs, inp')
@@ -181,19 +215,19 @@ pushLexState :: Int -> P ()
 pushLexState n = psLexState %= (n:)
 
 readInt :: Text -> Int
-readInt = T.foldr f 0 where
-    f c n = digitToInt c + 10*n
+readInt = T.foldl f 0 where
+    f n c = 10*n + digitToInt c
 
 constToken :: RlpToken -> LexerAction (Located RlpToken)
 constToken t inp l = do
     pos <- use (psInput . aiPos)
-    pure (Located (pos,l) t)
+    pure (Located (spanFromPos pos l) t)
 
 tokenWith :: (Text -> RlpToken) -> LexerAction (Located RlpToken)
 tokenWith tf inp l = do
     pos <- getPos
     let t = tf (T.take l $ inp ^. aiSource)
-    pure (Located (pos,l) t)
+    pure (Located (spanFromPos pos l) t)
 
 getPos :: P Position
 getPos = use (psInput . aiPos)
@@ -201,32 +235,12 @@ getPos = use (psInput . aiPos)
 alexEOF :: P (Located RlpToken)
 alexEOF = do
     inp <- getInput
-    pure (Located undefined TokenEOF)
+    pos <- getPos
+    pure (Located (spanFromPos pos 0) TokenEOF)
 
-execP :: P a -> ParseState -> Maybe a
-execP p st = runP p st & snd
-
-execP' :: P a -> Text -> Maybe a
-execP' p s = execP p st where
-    st = initParseState s
-
-initParseState :: Text -> ParseState
-initParseState s = ParseState
-    { _psLayoutStack = []
-    -- IMPORTANT: the initial state is `bol` to begin the top-level layout,
-    -- which then returns to state 0 which continues the normal lexing process.
-    , _psLexState = [layout_top,0]
-    , _psInput = initAlexInput s
-    , _psOpTable = mempty
-    }
-
-initAlexInput :: Text -> AlexInput
-initAlexInput s = AlexInput
-    { _aiPrevChar   = '\0'
-    , _aiSource     = s
-    , _aiBytes      = []
-    , _aiPos        = (1,1)
-    }
+runP' :: P a -> Text -> (ParseState, [MsgEnvelope RlpParseError], Maybe a)
+runP' p s = runP p st where
+    st = initParseState [layout_top,0] s
 
 lexToken :: P (Located RlpToken)
 lexToken = do
@@ -235,13 +249,14 @@ lexToken = do
     st <- use id
     -- traceM $ "st: " <> show st
     case alexScan inp c of
-        AlexEOF -> pure $ Located (inp ^. aiPos, 0) TokenEOF
+        AlexEOF -> pure $ Located (spanFromPos (inp^.aiPos) 0) TokenEOF
         AlexSkip inp' l -> do
             psInput .= inp'
             lexToken
         AlexToken inp' l act -> do
             psInput .= inp'
             act inp l
+        AlexError inp' -> addFatalHere 1 RlpParErrLexical
 
 lexCont :: (Located RlpToken -> P a) -> P a
 lexCont = (lexToken >>=)
@@ -260,7 +275,7 @@ lexDebug k = do
     k t
 
 lexTest :: Text -> Maybe [RlpToken]
-lexTest s = execP' lexStream s
+lexTest s = runP' lexStream s ^. _3
 
 indentLevel :: P Int
 indentLevel = do
@@ -270,7 +285,7 @@ indentLevel = do
 insertToken :: RlpToken -> P (Located RlpToken)
 insertToken t = do
     pos <- use (psInput . aiPos)
-    pure (Located (pos, 0) t)
+    pure (Located (spanFromPos pos 0) t)
 
 popLayout :: P Layout
 popLayout = do
@@ -279,7 +294,7 @@ popLayout = do
     psLayoutStack %= (drop 1)
     case ctx of
         Just l      -> pure l
-        Nothing     -> error "uhh"
+        Nothing     -> error "popLayout: layout stack empty! this is a bug."
 
 pushLayout :: Layout -> P ()
 pushLayout l = do
@@ -337,9 +352,13 @@ explicitRBrace inp l = do
 doLayout :: LexerAction (Located RlpToken)
 doLayout _ _ = do
     i <- indentLevel
+    -- traceM $ "doLayout: i: " <> show i
     pushLayout (Implicit i)
     popLexState
     insertLBrace
 
+programInitState :: Text -> ParseState
+programInitState = initParseState [layout_top,0]
+
 }
 

src/Rlp/Parse.y

@@ -1,37 +1,51 @@
 {
-{-# LANGUAGE LambdaCase #-}
+{-# LANGUAGE LambdaCase, ViewPatterns #-}
 module Rlp.Parse
     ( parseRlpProg
+    , parseRlpProgR
+    , parseRlpExpr
+    , parseRlpExprR
     )
     where
+import Compiler.RlpcError
+import Compiler.RLPC
 import Rlp.Lex
 import Rlp.Syntax
 import Rlp.Parse.Types
 import Rlp.Parse.Associate
-import Lens.Micro
-import Lens.Micro.Mtl
-import Lens.Micro.Platform          ()
+import Lens.Micro.Platform
 import Data.List.Extra
 import Data.Fix
 import Data.Functor.Const
+import Data.Functor.Apply
+import Data.Functor.Bind
+import Control.Comonad
+import Data.Functor
+import Data.Semigroup.Traversable
+import Data.Text                    (Text)
+import Data.Text                    qualified as T
+import Data.Void
 }
 
 %name parseRlpProg StandaloneProgram
+%name parseRlpExpr StandaloneExpr
 
 %monad { P }
 %lexer { lexCont } { Located _ TokenEOF }
 %error { parseError }
+%errorhandlertype explist
 %tokentype { Located RlpToken }
 
 %token
-    varname         { Located _ (TokenVarName $$) }
-    conname         { Located _ (TokenConName $$) }
-    consym          { Located _ (TokenConSym $$) }
-    varsym          { Located _ (TokenVarSym $$) }
+    varname         { Located _ (TokenVarName _) }
+    conname         { Located _ (TokenConName _) }
+    consym          { Located _ (TokenConSym _) }
+    varsym          { Located _ (TokenVarSym _) }
     data            { Located _ TokenData }
-    litint          { Located _ (TokenLitInt $$) }
+    litint          { Located _ (TokenLitInt _) }
     '='             { Located _ TokenEquals }
     '|'             { Located _ TokenPipe }
+    '::'            { Located _ TokenHasType }
     ';'             { Located _ TokenSemicolon }
     '('             { Located _ TokenLParen }
     ')'             { Located _ TokenRParen }
@@ -44,15 +58,22 @@ import Data.Functor.Const
     infixl          { Located _ TokenInfixL }
     infixr          { Located _ TokenInfixR }
     infix           { Located _ TokenInfix }
+    let             { Located _ TokenLet }
+    in              { Located _ TokenIn }
 
+%nonassoc '='
 %right '->'
+%right in
 
 %%
 
-StandaloneProgram   :: { RlpProgram' }
+StandaloneProgram   :: { RlpProgram RlpcPs }
 StandaloneProgram   : '{' Decls  '}'        {% mkProgram $2 }
                     | VL  DeclsV VR         {% mkProgram $2 }
 
+StandaloneExpr      :: { RlpExpr RlpcPs }
+                    : VL Expr VR            { extract $2 }
+
 VL  :: { () }
 VL  : vlbrace       { () }
 
@@ -60,117 +81,164 @@ VR  :: { () }
 VR  : vrbrace       { () }
     | error         { () }
 
-Decls               :: { [PartialDecl'] }
+Decls               :: { [Decl' RlpcPs] }
 Decls               : Decl ';' Decls        { $1 : $3 }
                     | Decl ';'              { [$1] }
                     | Decl                  { [$1] }
 
-DeclsV              :: { [PartialDecl'] }
+DeclsV              :: { [Decl' RlpcPs] }
 DeclsV              : Decl VS Decls         { $1 : $3 }
                     | Decl VS               { [$1] }
                     | Decl                  { [$1] }
+                    | {- epsilon -}         { [] }
 
 VS                  :: { Located RlpToken }
 VS                  : ';'                   { $1 }
                     | vsemi                 { $1 }
 
-Decl        :: { PartialDecl' }
+Decl        :: { Decl' RlpcPs }
             : FunDecl                   { $1 }
+            | TySigDecl                 { $1 }
             | DataDecl                  { $1 }
             | InfixDecl                 { $1 }
 
-InfixDecl   :: { PartialDecl' }
-            : InfixWord litint InfixOp  {% mkInfixD $1 $2 $3 }
+TySigDecl   :: { Decl' RlpcPs }
+            : Var '::' Type             { (\e -> TySigD [extract e]) <<~ $1 <~> $3 }
 
-InfixWord   :: { Assoc }
-            : infixl                    { InfixL }
-            | infixr                    { InfixR }
-            | infix                     { Infix  }
+InfixDecl   :: { Decl' RlpcPs }
+            : InfixWord litint InfixOp  { $1 =>> \w ->
+                                          InfixD (extract $1) (extractInt $ extract $2)
+                                          (extract $3) }
 
-DataDecl    :: { PartialDecl' }
-            : data Con TyParams '=' DataCons    { DataD $2 $3 $5 }
+InfixWord   :: { Located Assoc }
+            : infixl                    { $1 \$> InfixL }
+            | infixr                    { $1 \$> InfixR }
+            | infix                     { $1 \$> Infix  }
 
-TyParams    :: { [Name] }
+DataDecl    :: { Decl' RlpcPs }
+            : data Con TyParams '=' DataCons    { $1 \$> DataD (extract $2) $3 $5 }
+
+TyParams    :: { [PsName] }
             : {- epsilon -}             { [] }
-            | TyParams varname          { $1 `snoc` $2 }
+            | TyParams varname          { $1 `snoc` (extractName . extract $ $2) }
 
-DataCons    :: { [ConAlt] }
+DataCons    :: { [ConAlt RlpcPs] }
             : DataCons '|' DataCon      { $1 `snoc` $3 }
             | DataCon                   { [$1] }
 
-DataCon     :: { ConAlt }
-            : Con Type1s                { ConAlt $1 $2 }
+DataCon     :: { ConAlt RlpcPs }
+            : Con Type1s                { ConAlt (extract $1) $2 }
 
-Type1s      :: { [Type] }
+Type1s      :: { [RlpType' RlpcPs] }
             : {- epsilon -}             { [] }
             | Type1s Type1              { $1 `snoc` $2 }
 
-Type1       :: { Type }
+Type1       :: { RlpType' RlpcPs }
             : '(' Type ')'              { $2 }
-            | conname                   { TyCon $1 }
-            | varname                   { TyVar $1 }
+            | conname                   { fmap ConT (mkPsName $1) }
+            | varname                   { fmap VarT (mkPsName $1) }
 
-Type        :: { Type }
-            : Type '->' Type            { $1 :-> $3 }
+Type        :: { RlpType' RlpcPs }
+            : Type '->' Type            { FunT <<~ $1 <~> $3 }
             | Type1                     { $1 }
 
-FunDecl     :: { PartialDecl' }
-FunDecl     : Var Params '=' Expr       { FunD $1 $2 (Const $4) Nothing }
+FunDecl     :: { Decl' RlpcPs }
+FunDecl     : Var Params '=' Expr       { $4 =>> \e ->
+                                          FunD (extract $1) $2 e Nothing }
 
-Params      :: { [Pat'] }
+Params      :: { [Pat' RlpcPs] }
 Params      : {- epsilon -}             { [] }
             | Params Pat1               { $1 `snoc` $2 }
 
-Pat1        :: { Pat' }
-            : Var                       { VarP $1 }
-            | Lit                       { LitP $1 }
+Pat1        :: { Pat' RlpcPs }
+            : Var                       { fmap VarP $1 }
+            | Lit                       { LitP <<= $1 }
 
-Expr        :: { PartialExpr' }
-            : Expr1 varsym Expr         { Fix $ B $2 (unFix $1) (unFix $3) }
+Expr        :: { RlpExpr' RlpcPs }
+            : Expr1 InfixOp Expr        { $2 =>> \o ->
+                                          OAppE (extract o) $1 $3 }
             | Expr1                     { $1 }
+            | LetExpr                   { $1 }
 
-Expr1       :: { PartialExpr' }
-            : '(' Expr ')'              { wrapFix . Par . unwrapFix $ $2 }
-            | Lit                       { Fix . E $ LitEF $1 }
-            | Var                       { Fix . E $ VarEF $1 }
+LetExpr     :: { RlpExpr' RlpcPs }
+            : let layout1(Binding) in Expr { $1 \$> LetE $2 $4 }
 
--- TODO: happy prefers left-associativity. doing such would require adjusting
--- the code in Rlp.Parse.Associate to expect left-associative input rather than
--- right.
-InfixExpr   :: { PartialExpr' }
-            : Expr1 varsym Expr         { Fix $ B $2 (unFix $1) (unFix $3) }
+layout1(p)  : '{' layout_list1(';',p) '}'   { $2 }
+            | VL  layout_list1(VS,p) VR     { $2 }
 
-InfixOp     :: { Name }
-            : consym                    { $1 }
-            | varsym                    { $1 }
+layout_list1(sep,p) : p                          { [$1] }
+                    | layout_list1(sep,p) sep p  { $1 `snoc` $3 }
 
-Lit         :: { Lit' }
-Lit         : litint                    { IntL $1 }
+Binding     :: { Binding' RlpcPs }
+            : Pat1 '=' Expr              { PatB <<~ $1 <~> $3 }
 
-Var         :: { VarId }
-Var         : varname                   { NameVar $1 }
+Expr1       :: { RlpExpr' RlpcPs }
+            : '(' Expr ')'              { $1 .> $2 <. $3 }
+            | Lit                       { fmap LitE $1 }
+            | Var                       { fmap VarE $1 }
 
-Con         :: { ConId }
-            : conname                   { NameCon $1 }
+InfixOp     :: { Located PsName }
+            : consym                    { mkPsName $1 }
+            | varsym                    { mkPsName $1 }
+
+-- TODO: microlens-pro save me microlens-pro (rewrite this with prisms)
+Lit         :: { Lit' RlpcPs }
+            : litint                    { $1 <&> (IntL . (\ (TokenLitInt n) -> n)) }
+
+Var         :: { Located PsName }
+Var         : varname                   { mkPsName $1 }
+
+Con         :: { Located PsName }
+            : conname                   { mkPsName $1 }
 
 {
 
-mkProgram :: [PartialDecl'] -> P RlpProgram'
+parseRlpExprR = undefined
+
+parseRlpProgR :: (Monad m) => Text -> RLPCT m (RlpProgram RlpcPs)
+parseRlpProgR s = liftErrorful $ pToErrorful parseRlpProg st
+    where
+        st = programInitState s
+
+mkPsName :: Located RlpToken -> Located PsName
+mkPsName = fmap extractName
+
+extractName :: RlpToken -> PsName
+extractName = \case
+    TokenVarName n -> n
+    TokenConName n -> n
+    TokenConSym  n -> n
+    TokenVarSym  n -> n
+    _              -> error "mkPsName: not an identifier"
+
+extractInt :: RlpToken -> Int
+extractInt (TokenLitInt n) = n
+extractInt _ = error "extractInt: ugh"
+
+mkProgram :: [Decl' RlpcPs] -> P (RlpProgram RlpcPs)
 mkProgram ds = do
     pt <- use psOpTable
     pure $ RlpProgram (associate pt <$> ds)
 
-parseError :: Located RlpToken -> P a
-parseError = error . show
+parseError :: (Located RlpToken, [String]) -> P a
+parseError ((Located ss t), exp) = addFatal $
+    errorMsg ss (RlpParErrUnexpectedToken t exp)
 
-mkInfixD :: Assoc -> Int -> Name -> P PartialDecl'
+mkInfixD :: Assoc -> Int -> PsName -> P (Decl' RlpcPs)
 mkInfixD a p n = do
     let opl :: Lens' ParseState (Maybe OpInfo)
         opl = psOpTable . at n
     opl <~ (use opl >>= \case
-        -- TODO: non-fatal error
-        Just o  -> pure (Just o)
+        Just o  -> addWoundHere l e >> pure (Just o) where
+            e = RlpParErrDuplicateInfixD n
+            l = T.length n
         Nothing -> pure (Just (a,p))
         )
-    pure $ InfixD a p n
+    pos <- use (psInput . aiPos)
+    pure $ Located (spanFromPos pos 0) (InfixD a p n)
+
+intOfToken :: Located RlpToken -> Int
+intOfToken (Located _ (TokenLitInt n)) = n
+
 }
+

src/Rlp/Parse/Associate.hs

@@ -1,6 +1,7 @@
 {-# LANGUAGE OverloadedStrings #-}
 {-# LANGUAGE PatternSynonyms, ViewPatterns, ImplicitParams #-}
 module Rlp.Parse.Associate
+    {-# WARNING "temporarily unimplemented" #-}
     ( associate
     )
     where
@@ -13,88 +14,6 @@ import Rlp.Parse.Types
 import Rlp.Syntax
 --------------------------------------------------------------------------------
 
-associate :: OpTable -> PartialDecl' -> Decl' RlpExpr
-associate pt (FunD n as b w)    = FunD n as b' w
-    where b' = let ?pt = pt in completeExpr (getConst b)
-associate pt (TySigD ns t)      = TySigD ns t
-associate pt (DataD n as cs)    = DataD n as cs
-associate pt (InfixD a p n)     = InfixD a p n
-
-completeExpr :: (?pt :: OpTable) => PartialExpr' -> RlpExpr'
-completeExpr = cata completePartial
-
-completePartial :: (?pt :: OpTable) => PartialE -> RlpExpr'
-completePartial (E e)        = completeRlpExpr e
-completePartial p@(B o l r)  = completeB (build p)
-completePartial (Par e)      = completePartial e
-
-completeRlpExpr :: (?pt :: OpTable) => RlpExprF' RlpExpr' -> RlpExpr'
-completeRlpExpr = embed
-
-completeB :: (?pt :: OpTable) => PartialE -> RlpExpr'
-completeB p = case build p of
-    B o l r     -> (o' `AppE` l') `AppE` r'
-        where
-            -- TODO: how do we know it's symbolic?
-            o' = VarE (SymVar o)
-            l' = completeB l
-            r' = completeB r
-    Par e       -> completeB e
-    E e         -> completeRlpExpr e
-
-build :: (?pt :: OpTable) => PartialE -> PartialE
-build e = go id e (rightmost e) where
-    rightmost :: PartialE -> PartialE
-    rightmost (B _ _ r) = rightmost r
-    rightmost p@(E _)   = p
-    rightmost p@(Par _) = p
-
-    go :: (?pt :: OpTable)
-       => (PartialE -> PartialE)
-       -> PartialE -> PartialE -> PartialE
-    go f p@(WithInfo o _ r) = case r of
-            E _     -> mkHole o (f . f')
-            Par _   -> mkHole o (f . f')
-            B _ _ _ -> go (mkHole o (f . f')) r
-        where f' r' = p & pR .~ r'
-    go f _ = id
-
-mkHole :: (?pt :: OpTable)
-       => OpInfo
-       -> (PartialE -> PartialE)
-       -> PartialE
-       -> PartialE
-mkHole _     hole p@(Par _)                 = hole p
-mkHole _     hole p@(E _)                   = hole p
-mkHole (a,d) hole p@(WithInfo (a',d') _ _)
-    | d' < d  = above
-    | d' > d  = below
-    | d == d' = case (a,a') of
-                  -- left-associative operators of equal precedence are
-                  -- associated left
-                  (InfixL,InfixL) -> above
-                  -- right-associative operators are handled similarly
-                  (InfixR,InfixR) -> below
-                  -- non-associative operators of equal precedence, or equal
-                  -- precedence operators of different associativities are
-                  -- invalid
-                  (_,     _)      -> error "invalid expression"
-    where
-        above = p & pL %~ hole
-        below = hole p
-
-examplePrecTable :: OpTable
-examplePrecTable = H.fromList
-    [ ("+",    (InfixL,6))
-    , ("*",    (InfixL,7))
-    , ("^",    (InfixR,8))
-    , (".",    (InfixR,7))
-    , ("~",    (Infix, 9))
-    , ("=",    (Infix, 4))
-    , ("&&",   (Infix, 3))
-    , ("||",   (Infix, 2))
-    , ("$",    (InfixR,0))
-    , ("&",    (InfixL,0))
-    ]
-
+associate x y = y
+{-# WARNING associate "temporarily undefined" #-}
 

src/Rlp/Parse/Types.hs

@@ -1,11 +1,37 @@
 {-# LANGUAGE TemplateHaskell #-}
 {-# LANGUAGE ImplicitParams, ViewPatterns, PatternSynonyms #-}
 {-# LANGUAGE LambdaCase #-}
-module Rlp.Parse.Types where
+module Rlp.Parse.Types
+    (
+    -- * Trees That Grow
+      RlpcPs
+
+    -- * Parser monad and state
+    , P(..), ParseState(..), Layout(..), OpTable, OpInfo
+    , initParseState, initAlexInput
+    , pToErrorful
+    -- ** Lenses
+    , psLayoutStack, psLexState, psInput, psOpTable
+
+    -- * Other parser types
+    , RlpToken(..), AlexInput(..), Position(..), spanFromPos, LexerAction
+    , Located(..), PsName
+    -- ** Lenses
+    , aiPrevChar, aiSource, aiBytes, aiPos, posLine, posColumn
+
+    , (<<~), (<~>)
+
+    -- * Error handling
+    , MsgEnvelope(..), RlpcError(..), RlpParseError(..)
+    , addFatal, addWound, addFatalHere, addWoundHere
+    )
+    where
 --------------------------------------------------------------------------------
 import Core.Syntax                  (Name)
 import Control.Monad
-import Control.Monad.State.Class
+import Control.Monad.State.Strict
+import Control.Monad.Errorful
+import Compiler.RlpcError
 import Data.Text                    (Text)
 import Data.Maybe
 import Data.Fix
@@ -13,12 +39,47 @@ import Data.Functor.Foldable
 import Data.Functor.Const
 import Data.Functor.Classes
 import Data.HashMap.Strict          qualified as H
+import Data.Void
 import Data.Word                    (Word8)
+import Data.Text                    qualified as T
 import Lens.Micro.TH
 import Lens.Micro
 import Rlp.Syntax
+import Compiler.Types
 --------------------------------------------------------------------------------
 
+-- | Phantom type identifying rlpc's parser phase
+
+data RlpcPs
+
+type instance XRec RlpcPs f = Located (f RlpcPs)
+type instance IdP RlpcPs = PsName
+
+type instance XFunD RlpcPs = ()
+type instance XDataD RlpcPs = ()
+type instance XInfixD RlpcPs = ()
+type instance XTySigD RlpcPs = ()
+type instance XXDeclD RlpcPs = ()
+
+type instance XLetE RlpcPs = ()
+type instance XVarE RlpcPs = ()
+type instance XLamE RlpcPs = ()
+type instance XCaseE RlpcPs = ()
+type instance XIfE RlpcPs = ()
+type instance XAppE RlpcPs = ()
+type instance XLitE RlpcPs = ()
+type instance XParE RlpcPs = ()
+type instance XOAppE RlpcPs = ()
+
+type PsName = Text
+
+--------------------------------------------------------------------------------
+
+spanFromPos :: Position -> Int -> SrcSpan
+spanFromPos (l,c,a) s = SrcSpan l c a s
+
+{-# INLINE spanFromPos #-}
+
 type LexerAction a = AlexInput -> Int -> P a
 
 data AlexInput = AlexInput
@@ -30,10 +91,20 @@ data AlexInput = AlexInput
     deriving Show
 
 type Position =
-    ( Int -- line
-    , Int -- column
+    ( Int -- ^ line
+    , Int -- ^ column
+    , Int -- ^ Absolutely
     )
 
+posLine :: Lens' Position Int
+posLine = _1
+
+posColumn :: Lens' Position Int
+posColumn = _2
+
+posAbsolute :: Lens' Position Int
+posAbsolute = _3
+
 data RlpToken
     -- literals
     = TokenLitInt Int
@@ -64,31 +135,46 @@ data RlpToken
     | TokenLParen
     | TokenRParen
     -- 'virtual' control symbols, inserted by the lexer without any correlation
-    -- to a specific symbol
+    -- to a specific part of the input
     | TokenSemicolonV
     | TokenLBraceV
     | TokenRBraceV
     | TokenEOF
     deriving (Show)
 
-newtype P a = P { runP :: ParseState -> (ParseState, Maybe a) }
+newtype P a = P {
+        runP :: ParseState
+             -> (ParseState, [MsgEnvelope RlpParseError], Maybe a)
+    }
     deriving (Functor)
 
+pToErrorful :: (Applicative m)
+            => P a -> ParseState -> ErrorfulT (MsgEnvelope RlpParseError) m a
+pToErrorful p st = ErrorfulT $ pure (ma,es) where
+    (_,es,ma) = runP p st
+
 instance Applicative P where
-    pure a = P $ \st -> (st,Just a)
+    pure a = P $ \st -> (st, [], pure a)
     liftA2 = liftM2
 
 instance Monad P where
     p >>= k = P $ \st ->
-        let (st',a) = runP p st
-        in case a of
-            Just x  -> runP (k x) st'
-            Nothing -> (st', Nothing)
+        let (st',es,ma) = runP p st
+        in case ma of
+            Just a  -> runP (k a) st'
+                     & _2 %~ (es<>)
+            Nothing -> (st',es,Nothing)
+
+    {-# INLINE (>>=) #-}
 
 instance MonadState ParseState P where
     state f = P $ \st ->
         let (a,st') = f st
-        in (st', Just a)
+        in (st', [], Just a)
+
+instance MonadErrorful (MsgEnvelope RlpParseError) P where
+    addWound e = P $ \st -> (st, [e], Just ())
+    addFatal e = P $ \st -> (st, [e], Nothing)
 
 data ParseState = ParseState
     { _psLayoutStack        :: [Layout]
@@ -102,62 +188,83 @@ data Layout = Explicit
             | Implicit Int
             deriving (Show, Eq)
 
-data Located a = Located (Position, Int) a
-    deriving (Show)
-
 type OpTable = H.HashMap Name OpInfo
 type OpInfo = (Assoc, Int)
 
 -- data WithLocation a = WithLocation [String] a
 
 data RlpParseError = RlpParErrOutOfBoundsPrecedence Int
-                   | RlpParErrDuplicateInfixD
-    deriving (Eq, Ord, Show)
+                   | RlpParErrDuplicateInfixD Name
+                   | RlpParErrLexical
+                   | RlpParErrUnexpectedToken RlpToken [String]
+                   deriving (Show)
+
+instance IsRlpcError RlpParseError where
+    liftRlpcError = \case
+        RlpParErrOutOfBoundsPrecedence n ->
+            Text [ "Illegal precedence in infixity declaration"
+                 , "rl' currently only allows precedences between 0 and 9."
+                 ]
+        RlpParErrDuplicateInfixD s ->
+            Text [ "Conflicting infixity declarations for operator "
+                    <> tshow s
+                 ]
+        RlpParErrLexical ->
+            Text [ "Unknown lexical error :(" ]
+        RlpParErrUnexpectedToken t exp ->
+            Text [ "Unexpected token " <> tshow t
+                 , "Expected: " <> tshow exp
+                 ]
+        where
+            tshow :: (Show a) => a -> T.Text
+            tshow = T.pack . show
 
 ----------------------------------------------------------------------------------
 
--- absolute psycho shit (partial ASTs)
-
-type PartialDecl' = Decl (Const PartialExpr') Name
-
-data Partial a = E (RlpExprF Name a)
-               | B Name (Partial a) (Partial a)
-               | Par (Partial a)
-               deriving (Show, Functor)
-
-pL :: Traversal' (Partial a) (Partial a)
-pL k (B o l r) = (\l' -> B o l' r) <$> k l
-pL _ x         = pure x
-
-pR :: Traversal' (Partial a) (Partial a)
-pR k (B o l r) = (\r' -> B o l r') <$> k r
-pR _ x         = pure x
-
-type PartialE = Partial RlpExpr'
-
--- i love you haskell
-pattern WithInfo :: (?pt :: OpTable) => OpInfo -> PartialE -> PartialE -> PartialE
-pattern WithInfo p l r <- B (opInfoOrDef -> p) l r
-
-opInfoOrDef :: (?pt :: OpTable) => Name -> OpInfo
-opInfoOrDef c = fromMaybe (InfixL,9) $ H.lookup c ?pt
-
--- required to satisfy constraint on Fix's show instance
-instance Show1 Partial where
-    liftShowsPrec :: forall a. (Int -> a -> ShowS)
-                  -> ([a] -> ShowS)
-                  -> Int -> Partial a -> ShowS
-
-    liftShowsPrec sp sl p m = case m of
-        (E e)       -> showsUnaryWith lshow "E" p e
-        (B f a b)   -> showsTernaryWith showsPrec lshow lshow "B" p f a b
-        (Par e)     -> showsUnaryWith lshow "Par" p e
-        where
-            lshow :: forall f. (Show1 f) => Int -> f a -> ShowS
-            lshow = liftShowsPrec sp sl
-
-type PartialExpr' = Fix Partial
-
 makeLenses ''AlexInput
 makeLenses ''ParseState
 
+addWoundHere :: Int -> RlpParseError -> P ()
+addWoundHere l e = P $ \st ->
+    let e' = MsgEnvelope
+            { _msgSpan = let pos = psInput . aiPos
+                         in SrcSpan (st ^. pos . posLine)
+                                    (st ^. pos . posColumn)
+                                    (st ^. pos . posAbsolute)
+                                    l
+            , _msgDiagnostic = e
+            , _msgSeverity = SevError
+            }
+    in (st, [e'], Just ())
+
+addFatalHere :: Int -> RlpParseError -> P a
+addFatalHere l e = P $ \st ->
+    let e' = MsgEnvelope
+            { _msgSpan = let pos = psInput . aiPos
+                         in SrcSpan (st ^. pos . posLine)
+                                    (st ^. pos . posColumn)
+                                    (st ^. pos . posAbsolute)
+                                    l
+            , _msgDiagnostic = e
+            , _msgSeverity = SevError
+            }
+    in (st, [e'], Nothing)
+
+initParseState :: [Int] -> Text -> ParseState
+initParseState ls s = ParseState
+    { _psLayoutStack = []
+    -- IMPORTANT: the initial state is `bol` to begin the top-level layout,
+    -- which then returns to state 0 which continues the normal lexing process.
+    , _psLexState = ls
+    , _psInput = initAlexInput s
+    , _psOpTable = mempty
+    }
+
+initAlexInput :: Text -> AlexInput
+initAlexInput s = AlexInput
+    { _aiPrevChar   = '\0'
+    , _aiSource     = s
+    , _aiBytes      = []
+    , _aiPos        = (1,1,0)
+    }
+

src/Rlp/Syntax.hs

@@ -1,40 +1,36 @@
 -- recursion-schemes
-{-# LANGUAGE DeriveFunctor, DeriveFoldable, DeriveTraversable #-}
--- recursion-schemes
-{-# LANGUAGE TemplateHaskell, TypeFamilies #-}
+{-# LANGUAGE DeriveFunctor, DeriveFoldable, DeriveTraversable
+  , TemplateHaskell, TypeFamilies #-}
 {-# LANGUAGE OverloadedStrings, PatternSynonyms #-}
+{-# LANGUAGE TypeFamilies, TypeFamilyDependencies #-}
+{-# LANGUAGE UndecidableInstances, ImpredicativeTypes #-}
 module Rlp.Syntax
-    ( RlpModule(..)
-    , RlpProgram(..)
-    , RlpProgram'
-    , rlpmodName
-    , rlpmodProgram
-    , RlpExpr(..)
-    , RlpExpr'
-    , RlpExprF(..)
-    , RlpExprF'
-    , Decl(..)
-    , Decl'
-    , Bind(..)
-    , Where
-    , Where'
-    , ConAlt(..)
-    , Type(..)
-    , pattern (:->)
+    (
+    -- * AST
+      RlpProgram(..)
+    , Decl(..), Decl', RlpExpr(..), RlpExpr'
+    , Pat(..), Pat'
     , Assoc(..)
-    , VarId(..)
-    , ConId(..)
-    , Pat(..)
-    , Pat'
-    , Lit(..)
-    , Lit'
-    , Name
+    , Lit(..), Lit'
+    , RlpType(..), RlpType'
+    , ConAlt(..)
+    , Binding(..), Binding'
 
-    -- TODO: ugh move this somewhere else later
-    , showsTernaryWith
-
-    -- * Convenience re-exports
-    , Text
+    -- * Trees That Grow boilerplate
+    -- ** Extension points
+    , IdP, XRec, UnXRec(..), MapXRec(..)
+    -- *** Decl
+    , XFunD, XTySigD, XInfixD, XDataD, XXDeclD
+    -- *** RlpExpr
+    , XLetE, XVarE, XLamE, XCaseE, XIfE, XAppE, XLitE
+    , XParE, XOAppE, XXRlpExprE
+    -- ** Pattern synonyms
+    -- *** Decl
+    , pattern FunD, pattern TySigD, pattern InfixD, pattern DataD
+    -- *** RlpExpr
+    , pattern LetE, pattern VarE, pattern LamE, pattern CaseE, pattern IfE
+    , pattern AppE, pattern LitE, pattern ParE, pattern OAppE
+    , pattern XRlpExprE
     )
     where
 ----------------------------------------------------------------------------------
@@ -43,93 +39,180 @@ import Data.Text                    qualified as T
 import Data.String                  (IsString(..))
 import Data.Functor.Foldable.TH     (makeBaseFunctor)
 import Data.Functor.Classes
+import Data.Kind                    (Type)
 import Lens.Micro
 import Lens.Micro.TH
-import Core.Syntax                  hiding (Lit)
+import Core.Syntax                  hiding (Lit, Type, Binding, Binding')
 import Core                         (HasRHS(..), HasLHS(..))
 ----------------------------------------------------------------------------------
 
-data RlpModule b = RlpModule
+data RlpModule p = RlpModule
     { _rlpmodName       :: Text
-    , _rlpmodProgram    :: RlpProgram b
+    , _rlpmodProgram    :: RlpProgram p
     }
 
-newtype RlpProgram b = RlpProgram [Decl RlpExpr b]
-    deriving Show
+-- | dear god.
+type PhaseShow p =
+    ( Show (XRec p Pat), Show (XRec p RlpExpr)
+    , Show (XRec p Lit), Show (IdP p)
+    , Show (XRec p RlpType)
+    , Show (XRec p Binding)
+    )
 
-type RlpProgram' = RlpProgram Name
+newtype RlpProgram p = RlpProgram [Decl' p]
 
--- | The @e@ parameter is used for partial results. When parsing an input, we
--- first parse all top-level declarations in order to extract infix[lr]
--- declarations. This process yields a @[Decl (Const Text) Name]@, where @Const
--- Text@ stores the remaining unparsed function bodies. Once infixities are
--- accounted for, we may complete the parsing task and get a proper @[Decl
--- RlpExpr Name]@.
+deriving instance (PhaseShow p, Show (XRec p Decl)) => Show (RlpProgram p)
 
-data Decl e b = FunD    VarId [Pat b] (e b) (Maybe (Where b))
-              | TySigD  [VarId] Type
-              | DataD   ConId [Name] [ConAlt]
-              | InfixD  Assoc Int Name
-              deriving Show
+data RlpType p = FunConT
+               | FunT (RlpType' p) (RlpType' p)
+               | AppT (RlpType' p) (RlpType' p)
+               | VarT (IdP p)
+               | ConT (IdP p)
 
-type Decl' e = Decl e Name
+type RlpType' p = XRec p RlpType
+
+deriving instance (PhaseShow p)
+                  => Show (RlpType p)
+
+data Decl p = FunD'   (XFunD p)    (IdP p) [Pat' p] (RlpExpr' p) (Maybe (Where p))
+            | TySigD' (XTySigD p)  [IdP p] (RlpType' p)
+            | DataD'  (XDataD p)   (IdP p) [IdP p] [ConAlt p]
+            | InfixD' (XInfixD p)  Assoc Int (IdP p)
+            | XDeclD' !(XXDeclD p)
+
+deriving instance
+    ( Show (XFunD p), Show (XTySigD p)
+    , Show (XDataD p), Show (XInfixD p)
+    , Show (XXDeclD p)
+    , PhaseShow p
+    ) 
+    => Show (Decl p)
+
+type family XFunD p
+type family XTySigD p
+type family XDataD p
+type family XInfixD p
+type family XXDeclD p
+
+pattern FunD   :: (XFunD p ~ ())
+               => (IdP p) -> [Pat' p] -> (RlpExpr' p) -> (Maybe (Where p))
+               -> Decl p
+pattern TySigD :: (XTySigD p ~ ()) => [IdP p] -> (RlpType' p) -> Decl p
+pattern DataD  :: (XDataD p ~ ()) => (IdP p) -> [IdP p] -> [ConAlt p] -> Decl p
+pattern InfixD :: (XInfixD p ~ ()) => Assoc -> Int -> (IdP p) -> Decl p
+pattern XDeclD :: (XXDeclD p ~ ()) => Decl p
+
+pattern FunD n as e wh = FunD' () n as e wh
+pattern TySigD ns t = TySigD' () ns t
+pattern DataD n as cs = DataD' () n as cs
+pattern InfixD a p n = InfixD' () a p n
+pattern XDeclD = XDeclD' ()
+
+type Decl' p = XRec p Decl
 
 data Assoc = InfixL
            | InfixR
            | Infix
-           deriving Show
+           deriving (Show)
 
-data ConAlt = ConAlt ConId [Type]
-            deriving Show
+data ConAlt p = ConAlt (IdP p) [RlpType' p]
 
-data RlpExpr b = LetE  [Bind b] (RlpExpr b)
-               | VarE  VarId
-               | ConE  ConId
-               | LamE  [Pat b] (RlpExpr b)
-               | CaseE (RlpExpr b) [(Alt b, Where b)]
-               | IfE   (RlpExpr b) (RlpExpr b) (RlpExpr b)
-               | AppE  (RlpExpr b) (RlpExpr b)
-               | LitE  (Lit b)
-               deriving Show
+deriving instance (Show (IdP p), Show (XRec p RlpType)) => Show (ConAlt p)
 
-type RlpExpr' = RlpExpr Name
+data RlpExpr p = LetE'  (XLetE p) [Binding' p] (RlpExpr' p)
+               | VarE'  (XVarE p) (IdP p)
+               | LamE'  (XLamE p) [Pat p] (RlpExpr' p)
+               | CaseE' (XCaseE p) (RlpExpr' p) [(Alt p, Where p)]
+               | IfE'   (XIfE p) (RlpExpr' p) (RlpExpr' p) (RlpExpr' p)
+               | AppE'  (XAppE p) (RlpExpr' p) (RlpExpr' p)
+               | LitE'  (XLitE p) (Lit p)
+               | ParE'  (XParE p) (RlpExpr' p)
+               | OAppE' (XOAppE p) (IdP p) (RlpExpr' p) (RlpExpr' p)
+               | XRlpExprE' !(XXRlpExprE p)
 
-type Where b = [Bind b]
-type Where' = [Bind Name]
+type family XLetE p
+type family XVarE p
+type family XLamE p
+type family XCaseE p
+type family XIfE p
+type family XAppE p
+type family XLitE p
+type family XParE p
+type family XOAppE p
+type family XXRlpExprE p
+
+pattern LetE :: (XLetE p ~ ()) => [Binding' p] -> RlpExpr' p -> RlpExpr p
+pattern VarE :: (XVarE p ~ ()) => IdP p -> RlpExpr p
+pattern LamE :: (XLamE p ~ ()) => [Pat p] -> RlpExpr' p -> RlpExpr p
+pattern CaseE :: (XCaseE p ~ ()) => RlpExpr' p -> [(Alt p, Where p)] -> RlpExpr p
+pattern IfE :: (XIfE p ~ ()) => RlpExpr' p -> RlpExpr' p -> RlpExpr' p -> RlpExpr p
+pattern AppE :: (XAppE p ~ ()) => RlpExpr' p -> RlpExpr' p -> RlpExpr p
+pattern LitE :: (XLitE p ~ ()) => Lit p -> RlpExpr p
+pattern ParE :: (XParE p ~ ()) => RlpExpr' p -> RlpExpr p
+pattern OAppE :: (XOAppE p ~ ()) => IdP p -> RlpExpr' p -> RlpExpr' p -> RlpExpr p
+pattern XRlpExprE :: (XXRlpExprE p ~ ()) => RlpExpr p
+
+pattern LetE bs e = LetE' () bs e
+pattern VarE n = VarE' () n
+pattern LamE as e = LamE' () as e
+pattern CaseE e as = CaseE' () e as
+pattern IfE c a b = IfE' () c a b
+pattern AppE f x = AppE' () f x
+pattern LitE l = LitE' () l
+pattern ParE e = ParE' () e
+pattern OAppE n a b = OAppE' () n a b
+pattern XRlpExprE = XRlpExprE' ()
+
+deriving instance
+    ( Show (XLetE p), Show (XVarE p), Show (XLamE p)
+    , Show (XCaseE p), Show (XIfE p), Show (XAppE p)
+    , Show (XLitE p), Show (XParE p), Show (XOAppE p)
+    , Show (XXRlpExprE p)
+    , PhaseShow p
+    ) => Show (RlpExpr p)
+
+type RlpExpr' p = XRec p RlpExpr
+
+class UnXRec p where
+    unXRec :: XRec p f -> f p
+
+class MapXRec p where
+    mapXRec :: (f p -> f' p') -> XRec p f -> XRec p' f'
+
+type family XRec p (f :: Type -> Type) = (r :: Type) | r -> p f
+
+type family IdP p
+
+type Where p = [Binding p]
 
 -- do we want guards?
-data Alt b = AltA (Pat b) (RlpExpr b)
-           deriving Show
+data Alt p = AltA (Pat' p) (RlpExpr' p)
 
-data Bind b = PatB (Pat b) (RlpExpr b)
-            | FunB VarId [Pat b] (RlpExpr b)
-            deriving Show
+deriving instance (PhaseShow p) => Show (Alt p)
 
-data VarId = NameVar Text
-           | SymVar Text
-           deriving Show
+data Binding p = PatB (Pat' p) (RlpExpr' p)
+               | FunB (IdP p) [Pat' p] (RlpExpr' p)
 
-instance IsString VarId where
-    -- TODO: use symvar if it's an operator
-    fromString = NameVar . T.pack
+type Binding' p = XRec p Binding
 
-data ConId = NameCon Text
-           | SymCon Text
-           deriving Show
+deriving instance (Show (XRec p Pat), Show (XRec p RlpExpr), Show (IdP p)
+                  ) => Show (Binding p)
 
-data Pat b = VarP VarId
-           | LitP (Lit b)
-           | ConP ConId [Pat b]
-           deriving Show
+data Pat p = VarP (IdP p)
+           | LitP (Lit' p)
+           | ConP (IdP p) [Pat' p]
 
-type Pat' = Pat Name
+deriving instance (PhaseShow p) => Show (Pat p)
 
-data Lit b = IntL Int
+type Pat' p = XRec p Pat
+
+data Lit p = IntL Int
            | CharL Char
-           | ListL [RlpExpr b]
-           deriving Show
+           | ListL [RlpExpr' p]
 
-type Lit' = Lit Name
+deriving instance (PhaseShow p) => Show (Lit p)
+
+type Lit' p = XRec p Lit
 
 -- instance HasLHS Alt Alt Pat Pat where
 --     _lhs = lens
@@ -143,33 +226,17 @@ type Lit' = Lit Name
 
 makeBaseFunctor ''RlpExpr
 
-deriving instance (Show b, Show a) => Show (RlpExprF b a)
-
-type RlpExprF' = RlpExprF Name
-
--- society if derivable Show1
-instance (Show b) => Show1 (RlpExprF b) where
-    liftShowsPrec sp _ p m = case m of
-        (LetEF bs e)            -> showsBinaryWith showsPrec sp "LetEF" p bs e
-        (VarEF n)               -> showsUnaryWith showsPrec "VarEF" p n
-        (ConEF n)               -> showsUnaryWith showsPrec "ConEF" p n
-        (LamEF bs e)            -> showsBinaryWith showsPrec sp "LamEF" p bs e
-        (CaseEF e as)           -> showsBinaryWith sp showsPrec "CaseEF" p e as
-        (IfEF a b c)            -> showsTernaryWith sp sp sp "IfEF" p a b c
-        (AppEF f x)             -> showsBinaryWith sp sp "AppEF" p f x
-        (LitEF l)               -> showsUnaryWith showsPrec "LitEF" p l
-
-showsTernaryWith :: (Int -> x -> ShowS)
-                 -> (Int -> y -> ShowS)
-                 -> (Int -> z -> ShowS)
-                 -> String -> Int
-                 -> x -> y -> z
-                 -> ShowS
-showsTernaryWith sa sb sc name p a b c = showParen (p > 10)
-    $ showString name
-    . showChar ' ' . sa 11 a 
-    . showChar ' ' . sb 11 b
-    . showChar ' ' . sc 11 c
+-- showsTernaryWith :: (Int -> x -> ShowS)
+--                  -> (Int -> y -> ShowS)
+--                  -> (Int -> z -> ShowS)
+--                  -> String -> Int
+--                  -> x -> y -> z
+--                  -> ShowS
+-- showsTernaryWith sa sb sc name p a b c = showParen (p > 10)
+--     $ showString name
+--     . showChar ' ' . sa 11 a 
+--     . showChar ' ' . sb 11 b
+--     . showChar ' ' . sc 11 c
 
 --------------------------------------------------------------------------------
 

tst/Core/HindleyMilnerSpec.hs

@@ -38,9 +38,25 @@ spec = do
         let e = [coreExpr|3|]
         in check' [] (TyCon "Bool") e `shouldSatisfy` isLeft
 
-infer' :: Context' -> Expr' -> Either TypeError Type
-infer' g e = fmap fst . runErrorful $ infer g e
+    it "should infer `fix ((+#) 1)` :: Int" $
+        let g = [ ("fix", ("a" :-> "a") :-> "a")
+                , ("+#", TyInt :-> TyInt :-> TyInt) ]
+            e = [coreExpr|fix ((+#) 1)|]
+        in infer' g e `shouldBe` Right TyInt
 
-check' :: Context' -> Type -> Expr' -> Either TypeError ()
-check' g t e = fmap fst . runErrorful $ check g t e
+    it "should infer mutually recursively defined lists" $
+        let g = [ ("cons", TyInt :-> TyCon "IntList" :-> TyCon "IntList") ]
+            e :: Expr'
+            e = [coreExpr|letrec { as = cons 1 bs; bs = cons 2 as } in as|]
+        in infer' g e `shouldBe` Right (TyCon "IntList")
+
+infer' :: Context' -> Expr' -> Either [TypeError] Type
+infer' g e = case runErrorful $ infer g e of
+    (Just t,   _) -> Right t
+    (Nothing, es) -> Left es
+
+check' :: Context' -> Type -> Expr' -> Either [TypeError] ()
+check' g t e = case runErrorful $ check g t e of
+    (Just t,   _) -> Right ()
+    (Nothing, es) -> Left es
 

tst/GMSpec.hs

@@ -27,15 +27,22 @@ spec = do
             in coreRes `shouldBe` arithRes
 
     describe "test programs" $ do
-        it "fac 3" $ do
+        it "fac 3" $
             resultOf Ex.fac3 `shouldBe` Just (NNum 6)
 
-        it "sum [1,2,3]" $ do
+        it "sum [1,2,3]" $
             resultOf Ex.sumList `shouldBe` Just (NNum 6)
 
-        it "k 3 ((/#) 1 0)" $ do
+        it "k 3 ((/#) 1 0)" $
             resultOf Ex.constDivZero `shouldBe` Just (NNum 3)
 
-        it "id (case ... of { ... })" $ do
+        it "id (case ... of { ... })" $
             resultOf Ex.idCase `shouldBe` Just (NNum 5)
 
+        it "bool pattern matching with named constructors" $
+            resultOf Ex.namedBoolCase `shouldBe` Just (NNum 123)
+
+        it "list pattern matching with named constructors" $
+            resultOf Ex.namedConsCase `shouldBe` Just (NNum 6)
+
+