chore(lexer): clean up

I think it is clearer to phrase it this way so it is clear that we are
not unconsuming (i.e. changing the state of the parser).

.envrc

@@ -1 +0,0 @@
-use nix

.forgejo/workflows/ci.yml

@@ -1,2 +0,0 @@
-runs-on: self-hosted
-

.forgejo/workflows/test.yaml

@@ -1,99 +0,0 @@
-name: Haskell CI
-on:
-  pull_request:
-    branches:
-      - dev
-      - main
-  push:
-    branches:
-      - main
-
-jobs:
-  build:
-    runs-on: docker
-    container:
-      image: elland/haddock2:latest
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-      - name: Check versions
-        run: |
-          ghc --version
-          cabal --version
-          node --version
-      - name: Cache Cabal packages
-        uses: actions/cache@v4
-        with:
-          path: |
-            ~/.cabal/packages
-            ~/.cabal/store
-            dist-newstyle
-          key: ${{ runner.os }}-haskell-9.10-cabal-${{ hashFiles('**/*.cabal', '**/cabal.project') }}
-          restore-keys: |
-            ${{ runner.os }}-haskell-9.10-cabal-
-      - name: Update Cabal package index
-        run: cabal update
-      - name: Configure project
-        run: cabal configure --enable-tests --enable-benchmarks
-      - name: Build dependencies
-        run: cabal build --only-dependencies --enable-tests --enable-benchmarks
-      - name: Build project
-        run: cabal build --enable-tests --enable-benchmarks
-      - name: Run documentation build
-        run: cabal haddock
-
-  test:
-    runs-on: docker
-    container:
-      image: elland/haddock2:latest
-    needs: build
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-      - name: Cache Cabal packages
-        uses: actions/cache@v4
-        with:
-          path: |
-            ~/.cabal/packages
-            ~/.cabal/store
-            dist-newstyle
-          key: ${{ runner.os }}-haskell-9.10-cabal-${{ hashFiles('**/*.cabal', '**/cabal.project') }}
-          restore-keys: |
-            ${{ runner.os }}-haskell-9.10-cabal-
-      - name: Update Cabal package index
-        run: cabal update
-      - name: Configure project
-        run: cabal configure --enable-tests --enable-benchmarks
-      - name: Build dependencies
-        run: cabal build --only-dependencies --enable-tests --enable-benchmarks
-      - name: Build project
-        run: cabal build --enable-tests --enable-benchmarks
-      - name: Run tests
-        run: cabal test --test-show-details=direct
-
-  fourmolu:
-    runs-on: docker
-    container:
-      image: elland/haddock2:latest
-    needs: build
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-      - name: Run fourmolu
-        run: |
-          find src test app -name "*.hs" -exec fourmolu --check-idempotence {} \; 2>/dev/null || true
-          find src test app -name "*.hs" -exec fourmolu --mode check {} \;
-
-  hlint:
-    runs-on: docker
-    container:
-      image: elland/haddock2:latest
-    needs: build
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-      - name: Run hlint
-        run: |
-          if [ -d src ]; then hlint src/; fi
-          if [ -d test ]; then hlint test/; fi  
-          if [ -d app ]; then hlint app/; fi

.git-blame-ignore-revs

@@ -1,2 +0,0 @@
-# Fourmolu
-9998ac92263127b05fd1eb607f3b7740c69d3a58

Dockerfile

@@ -1,36 +0,0 @@
-FROM haskell:9.10.2-bullseye AS builder
-
-RUN apt-get update && apt-get install -y curl git && rm -rf /var/lib/apt/lists/*
-
-RUN curl -fsSL https://deb.nodesource.com/setup_22.x | bash - && \
-    apt-get install -y nodejs && \
-    rm -rf /var/lib/apt/lists/*
-
-RUN cabal update && \
-    cabal install --install-method=copy --installdir=/usr/local/bin \
-    fourmolu hlint cabal-gild
-
-WORKDIR /workspace
-
-FROM haskell:9.10.2-bullseye
-
-RUN apt-get update && apt-get install -y \
-    libgmp10 \
-    curl \
-    && rm -rf /var/lib/apt/lists/*
-
-RUN curl -fsSL https://deb.nodesource.com/setup_22.x | bash - && \
-    apt-get install -y nodejs && \
-    rm -rf /var/lib/apt/lists/*
-
-RUN cabal update
-
-COPY --from=builder /usr/local/bin/cabal /usr/local/bin/
-COPY --from=builder /usr/local/bin/fourmolu /usr/local/bin/
-COPY --from=builder /usr/local/bin/hlint /usr/local/bin/
-COPY --from=builder /usr/local/bin/cabal-gild /usr/local/bin/
-
-WORKDIR /workspace
-
-CMD [ "bash" ]
-

Makefile

@@ -1,44 +0,0 @@
-.PHONY: help
-help: ## Show this help
-	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?## "}; {printf "\033[36m%-30s\033[0m %s\n", $$1, $$2}'
-
-.PHONY: build
-build: ## Build the project
-	cabal build
-
-.PHONY: test
-test: ## Run tests
-	cabal test --test-show-details=direct
-
-.PHONY: clean
-clean: ## Clean build artifacts
-	cabal clean
-
-.PHONY: fourmolu
-fourmolu: ## Format Haskell code
-	find . -type f -name "*.hs" ! -path "./dist-newstyle/*" -exec fourmolu -i {} +
-
-.PHONY: fourmolu-check
-fourmolu-check: ## Check if code is formatted
-	find . -type f -name "*.hs" ! -path "./dist-newstyle/*" -exec fourmolu --mode check {} \;
-
-.PHONY: lint
-lint: ## Run hlint
-	hlint src test app
-
-.PHONY: cabal-gild
-cabal-gild: ## Format cabal file
-	cabal-gild --io=haddock2.cabal
-
-.PHONY: format
-format: fourmolu cabal-gild ## Run all formatters
-
-.PHONY: check
-check: fourmolu-check lint ## Run all checks (CI-style)
-
-.PHONY: ci
-ci: build test check ## Run full CI pipeline locally
-
-.PHONY: docs
-docs: ## Generate documentation
-	cabal haddock --haddock-hyperlink-source

fourmolu.yaml

@@ -1,72 +0,0 @@
-# Number of spaces per indentation step
-indentation: 2
-
-# Max line length for automatic line breaking
-column-limit: none
-
-# Styling of arrows in type signatures (choices: trailing, leading, or leading-args)
-function-arrows: trailing
-
-# How to place commas in multi-line lists, records, etc. (choices: leading or trailing)
-comma-style: leading
-
-# Styling of import/export lists (choices: leading, trailing, or diff-friendly)
-import-export-style: diff-friendly
-
-# Rules for grouping import declarations
-import-grouping: legacy
-
-# Whether to full-indent or half-indent 'where' bindings past the preceding body
-indent-wheres: false
-
-# Whether to leave a space before an opening record brace
-record-brace-space: false
-
-# Number of spaces between top-level declarations
-newlines-between-decls: 1
-
-# How to print Haddock comments (choices: single-line, multi-line, or multi-line-compact)
-haddock-style: multi-line
-
-# How to print module docstring
-haddock-style-module: null
-
-# Styling of let blocks (choices: auto, inline, newline, or mixed)
-let-style: auto
-
-# How to align the 'in' keyword with respect to the 'let' keyword (choices: left-align, right-align, or no-space)
-in-style: right-align
-
-# Whether to put parentheses around a single constraint (choices: auto, always, or never)
-single-constraint-parens: always
-
-# Whether to put parentheses around a single deriving class (choices: auto, always, or never)
-single-deriving-parens: always
-
-# Whether to sort constraints
-sort-constraints: false
-
-# Whether to sort derived classes
-sort-derived-classes: false
-
-# Whether to sort deriving clauses
-sort-deriving-clauses: false
-
-# Whether to place section operators (those that are infixr 0, such as $) in trailing position, continuing the expression indented below
-trailing-section-operators: true
-
-# Output Unicode syntax (choices: detect, always, or never)
-unicode: never
-
-# Give the programmer more choice on where to insert blank lines
-respectful: true
-
-# Fixity information for operators
-fixities: []
-
-# Module reexports Fourmolu should know about
-reexports: []
-
-# Modules defined by the current Cabal package for import grouping
-local-modules: []
-

haddock2.cabal

@@ -46,10 +46,10 @@ test-suite haddock2-test
   type: exitcode-stdio-1.0
   main-is: Spec.hs
   build-depends:
+    parsec ^>=3.1.18.0,
     base >=4.20.1.0,
     haddock2:{haddock2-lib},
     hspec ^>=2.11.0,
-    parsec ^>=3.1.18.0,
     text ^>=2.1.2,
 
   hs-source-dirs: test

npins/default.nix

@@ -1,146 +0,0 @@
-/*
-  This file is provided under the MIT licence:
-
-  Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
-
-  The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
-
-  THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
-*/
-# Generated by npins. Do not modify; will be overwritten regularly
-let
-  data = builtins.fromJSON (builtins.readFile ./sources.json);
-  version = data.version;
-
-  # https://github.com/NixOS/nixpkgs/blob/0258808f5744ca980b9a1f24fe0b1e6f0fecee9c/lib/lists.nix#L295
-  range =
-    first: last: if first > last then [ ] else builtins.genList (n: first + n) (last - first + 1);
-
-  # https://github.com/NixOS/nixpkgs/blob/0258808f5744ca980b9a1f24fe0b1e6f0fecee9c/lib/strings.nix#L257
-  stringToCharacters = s: map (p: builtins.substring p 1 s) (range 0 (builtins.stringLength s - 1));
-
-  # https://github.com/NixOS/nixpkgs/blob/0258808f5744ca980b9a1f24fe0b1e6f0fecee9c/lib/strings.nix#L269
-  stringAsChars = f: s: concatStrings (map f (stringToCharacters s));
-  concatMapStrings = f: list: concatStrings (map f list);
-  concatStrings = builtins.concatStringsSep "";
-
-  # If the environment variable NPINS_OVERRIDE_${name} is set, then use
-  # the path directly as opposed to the fetched source.
-  # (Taken from Niv for compatibility)
-  mayOverride =
-    name: path:
-    let
-      envVarName = "NPINS_OVERRIDE_${saneName}";
-      saneName = stringAsChars (c: if (builtins.match "[a-zA-Z0-9]" c) == null then "_" else c) name;
-      ersatz = builtins.getEnv envVarName;
-    in
-    if ersatz == "" then
-      path
-    else
-      # this turns the string into an actual Nix path (for both absolute and
-      # relative paths)
-      builtins.trace "Overriding path of \"${name}\" with \"${ersatz}\" due to set \"${envVarName}\"" (
-        if builtins.substring 0 1 ersatz == "/" then
-          /. + ersatz
-        else
-          /. + builtins.getEnv "PWD" + "/${ersatz}"
-      );
-
-  mkSource =
-    name: spec:
-    assert spec ? type;
-    let
-      path =
-        if spec.type == "Git" then
-          mkGitSource spec
-        else if spec.type == "GitRelease" then
-          mkGitSource spec
-        else if spec.type == "PyPi" then
-          mkPyPiSource spec
-        else if spec.type == "Channel" then
-          mkChannelSource spec
-        else if spec.type == "Tarball" then
-          mkTarballSource spec
-        else
-          builtins.throw "Unknown source type ${spec.type}";
-    in
-    spec // { outPath = mayOverride name path; };
-
-  mkGitSource =
-    {
-      repository,
-      revision,
-      url ? null,
-      submodules,
-      hash,
-      branch ? null,
-      ...
-    }:
-    assert repository ? type;
-    # At the moment, either it is a plain git repository (which has an url), or it is a GitHub/GitLab repository
-    # In the latter case, there we will always be an url to the tarball
-    if url != null && !submodules then
-      builtins.fetchTarball {
-        inherit url;
-        sha256 = hash; # FIXME: check nix version & use SRI hashes
-      }
-    else
-      let
-        url =
-          if repository.type == "Git" then
-            repository.url
-          else if repository.type == "GitHub" then
-            "https://github.com/${repository.owner}/${repository.repo}.git"
-          else if repository.type == "GitLab" then
-            "${repository.server}/${repository.repo_path}.git"
-          else
-            throw "Unrecognized repository type ${repository.type}";
-        urlToName =
-          url: rev:
-          let
-            matched = builtins.match "^.*/([^/]*)(\\.git)?$" url;
-
-            short = builtins.substring 0 7 rev;
-
-            appendShort = if (builtins.match "[a-f0-9]*" rev) != null then "-${short}" else "";
-          in
-          "${if matched == null then "source" else builtins.head matched}${appendShort}";
-        name = urlToName url revision;
-      in
-      builtins.fetchGit {
-        rev = revision;
-        inherit name;
-        # hash = hash;
-        inherit url submodules;
-      };
-
-  mkPyPiSource =
-    { url, hash, ... }:
-    builtins.fetchurl {
-      inherit url;
-      sha256 = hash;
-    };
-
-  mkChannelSource =
-    { url, hash, ... }:
-    builtins.fetchTarball {
-      inherit url;
-      sha256 = hash;
-    };
-
-  mkTarballSource =
-    {
-      url,
-      locked_url ? url,
-      hash,
-      ...
-    }:
-    builtins.fetchTarball {
-      url = locked_url;
-      sha256 = hash;
-    };
-in
-if version == 5 then
-  builtins.mapAttrs mkSource data.pins
-else
-  throw "Unsupported format version ${toString version} in sources.json. Try running `npins upgrade`"

npins/sources.json

@@ -1,11 +0,0 @@
-{
-  "pins": {
-    "nixpkgs": {
-      "type": "Channel",
-      "name": "nixpkgs-unstable",
-      "url": "https://releases.nixos.org/nixpkgs/nixpkgs-25.11pre868532.647e5c14cbd5/nixexprs.tar.xz",
-      "hash": "0i6mgl7pm7y4ydrrll7szmv8hhxb3cyny8x1g1a8sp3g5wl3yd9g"
-    }
-  },
-  "version": 5
-}

shell.nix

@@ -1,25 +0,0 @@
-let
-  sources = import ./npins;
-in
-{
-  pkgs ? import sources.nixpkgs { },
-}:
-
-pkgs.mkShell rec {
-  name = "haddock2";
-
-  packages =
-    with pkgs;
-    [
-      haskell.packages.ghc912.ghc
-      haskell.packages.ghc912.haskell-language-server
-      zlib
-    ]
-    ++ map haskell.lib.justStaticExecutables [
-      haskellPackages.cabal-gild
-      haskellPackages.fourmolu
-      cabal-install
-    ];
-
-  env.LD_LIBRARY_PATH = pkgs.lib.makeLibraryPath packages;
-}

src/Lexer.hs

@@ -1,154 +1,150 @@
 {-# LANGUAGE OverloadedStrings #-}
 
 module Lexer (
-  Token (..),
-  lexer,
-  emphasis,
-)
-where
+    Token (..),
+    lexer,
+    emphasis,
+) where
 
 import Control.Monad (mfilter, void)
-import Data.Char (ord, toLower)
 import Data.Functor (($>))
 import Data.Text (Text, intercalate)
 import Data.Text qualified as Text
-import GHC.Unicode (isAlphaNum, isControl, isDigit, isPrint, isSpace, isUpper)
+import GHC.Unicode (isAlphaNum, isControl, isPrint, isSpace, isUpper)
 import ParserMonad (Parser, initialParserState)
 import Text.Parsec
 import Text.Parsec qualified as Parsec
 import Text.Parsec.Pos (updatePosChar)
 
 type Located a = (SourcePos, a)
-
 type LocatedToken = (SourcePos, Token)
 
 type Lexer = Parser [LocatedToken]
 
 data Level
-  = One
-  | Two
-  | Three
-  | Four
-  | Five
-  | Six
-  deriving (Eq, Show)
+    = One
+    | Two
+    | Three
+    | Four
+    | Five
+    | Six
+    deriving (Eq, Show)
 
 data Token
-  = Token Text
-  | Anchor Text
-  | BirdTrack
-  | BoldOpen
-  | BoldClose
-  | Escape
-  | EmphasisOpen
-  | EmphasisClose
-  | Header Level
-  | MonospaceOpen
-  | MonospaceClose
-  | Newline
-  | LinkOpen
-  | LinkClose
-  | LabeledLinkOpen
-  | LabeledLinkClose
-  | ParenOpen
-  | ParenClose
-  | BracketOpen
-  | BracketClose
-  | MathInlineOpen
-  | MathInlineClose
-  | MathMultilineOpen
-  | MathMultilineClose
-  | NumericEntity Int
-  | Module Text
-  | QuoteOpen
-  | QuoteClose
-  | Space
-  | EOF
-  deriving (Eq, Show)
+    = Token Text
+    | Anchor Text
+    | BirdTrack
+    | BoldOpen
+    | BoldClose
+    | Escape
+    | EmphasisOpen
+    | EmphasisClose
+    | Header Level
+    | MonospaceOpen
+    | MonospaceClose
+    | Newline
+    | LinkOpen
+    | LinkClose
+    | LabeledLinkOpen
+    | LabeledLinkClose
+    | ParenOpen
+    | ParenClose
+    | BracketOpen
+    | BracketClose
+    | MathInlineOpen
+    | MathInlineClose
+    | MathMultilineOpen
+    | MathMultilineClose
+    | NumericEntity Int
+    | Module Text
+    | QuoteOpen
+    | QuoteClose
+    | Space
+    | EOF
+    deriving (Eq, Show)
 
 located :: Parser a -> Parser (SourcePos, a)
 located p = (,) <$> getPosition <*> p
 
 tokenise :: [Parser a] -> Parser [(SourcePos, a)]
-tokenise = mapM located
+tokenise = sequence . map located
 
 lexer :: String -> Either ParseError [LocatedToken]
 lexer = Parsec.runParser lexText initialParserState "input" . Text.pack
 
 lexText :: Parser [LocatedToken]
 lexText = go
- where
-  go = do
-    Parsec.optionMaybe Parsec.eof >>= \case
-      Just _ -> pure []
-      Nothing -> do
-        toks <-
-          choice $
-            Parsec.try
-              <$> [ mathMultiline
-                  , mathInline
-                  , escape -- maths go before escape to avoid mismatch
-                  , headers
-                  , newlineToken
-                  , spaceToken
-                  , link
-                  , labeledLink
-                  , module_
-                  , anchor
-                  , numericEntity
-                  , textElement
-                  , quotes
-                  , birdTrack
-                  , other
-                  ]
-        rest <- go
-        pure (toks <> rest)
+  where
+    go = do
+        Parsec.optionMaybe Parsec.eof >>= \case
+            Just _ -> pure []
+            Nothing -> do
+                toks <-
+                    choice $
+                        Parsec.try
+                            <$> [ mathMultiline
+                                , mathInline
+                                , escape -- maths go before escape to avoid mismatch
+                                , headers
+                                , newlineToken
+                                , spaceToken
+                                , link
+                                , labeledLink
+                                , module_
+                                , anchor
+                                , textElement
+                                , quotes
+                                , birdTrack
+                                , other
+                                ]
+                rest <- go
+                pure (toks <> rest)
 
 -- Tokens
 
 textElement :: Parser [LocatedToken]
 textElement =
-  choice $
-    Parsec.try
-      <$> [ emphasis
-          , bold
-          , monospace
-          ]
+    choice $
+        Parsec.try
+            <$> [ emphasis
+                , bold
+                , monospace
+                ]
 
 headers :: Parser [LocatedToken]
 headers =
-  choice $
-    Parsec.try
-      <$> [ header1
-          , header2
-          , header3
-          , header4
-          , header5
-          , header6
-          ]
+    choice $
+        Parsec.try
+            <$> [ header1
+                , header2
+                , header3
+                , header4
+                , header5
+                , header6
+                ]
 
 anyUntil :: Parser a -> Parser Text
 anyUntil p = Text.pack <$> manyTill anyChar (lookAhead p)
 
 delimitedAsTuple :: Parser open -> Parser close -> Parser (Located open, LocatedToken, Located close)
 delimitedAsTuple openP closeP =
-  (,,)
-    <$> located openP
-    <*> located (Token <$> anyUntil closeP)
-    <*> located closeP
+    (,,)
+        <$> located openP
+        <*> located (Token <$> anyUntil closeP)
+        <*> located closeP
 
 delimited :: Parser open -> Parser close -> Token -> Token -> Parser [LocatedToken]
 delimited openP closeP openTok closeTok = asList <$> delimitedAsTuple (openTok <$ openP) (closeTok <$ closeP)
- where
-  asList (a, tok, b) = [a, tok, b]
+    where
+        asList (a, tok, b) = [a, tok, b]
 
 delimitedNoTrailing :: Parser open -> Parser close -> Token -> Parser [LocatedToken]
 delimitedNoTrailing openP closeP openTok = asList <$> delimitedAsTuple (openTok <$ openP) (void closeP)
- where
-  asList (a, tok, _) = [a, tok]
+    where
+        asList (a, tok, _) = [a, tok]
 
 delimitedSymmetric :: Parser a -> Token -> Token -> Parser [LocatedToken]
-delimitedSymmetric s = delimited s s
+delimitedSymmetric s t1 t2 = delimited s s t1 t2
 
 eol :: Parser ()
 eol = void "\n" <|> void "\r\n" <|> Parsec.eof
@@ -174,8 +170,9 @@ header6 = delimitedNoTrailing "====== " eol (Header Six)
 -- #anchors#
 anchor :: Lexer
 anchor = do
-  x <- located $ between "#" "#" (Anchor <$> anyUntil "#")
-  pure [x]
+    x <- located $ between "#" "#" (Anchor <$> anyUntil "#")
+    pure [x]
+
 
 moduleNames :: Parser Text
 moduleNames = intercalate "." . fmap Text.pack <$> upperId `sepBy1` char '.'
@@ -191,47 +188,47 @@ identifierChar = satisfy (\c -> isAlphaNum c || c == '_')
 -- "Module.Name\#anchor" -- known as "old anchor". this has been deprecated for 9 years, thanks Ben
 module_ :: Lexer
 module_ = between (char '"') (char '"') inner
- where
-  inner = do
-    m <- located $ Module <$> moduleNames
-    mAnchor <- optionMaybe (located $ anchorHash *> (Anchor <$> anchorText))
-    pure $ case mAnchor of
-      Just anc -> [m, anc]
-      Nothing -> [m]
+  where
+    inner = do
+        m <- located $ Module <$> moduleNames
+        mAnchor <- optionMaybe (located $ anchorHash *> (Anchor <$> anchorText))
+        pure $ case mAnchor of
+            Just anc -> [m, anc]
+            Nothing -> [m]
 
-  anchorHash :: Parser Text
-  anchorHash = "#" <|> try "\\#"
+    anchorHash :: Parser Text
+    anchorHash = "#" <|> try "\\#"
 
-  anchorText :: Parser Text
-  anchorText = Text.pack <$> many (satisfy (\c -> c /= '"' && not (isSpace c)))
+    anchorText :: Parser Text
+    anchorText = Text.pack <$> many (satisfy (\c -> c /= '"' && not (isSpace c)))
 
 linkRaw :: Lexer
 linkRaw =
-  tokenise
-    [ BracketOpen <$ "["
-    , Token <$> anyUntil "]"
-    , BracketClose <$ "]"
-    , ParenOpen <$ "("
-    , Token <$> anyUntil ")"
-    , ParenClose <$ ")"
-    ]
+    tokenise
+        [ BracketOpen <$ "["
+        , Token <$> anyUntil "]"
+        , BracketClose <$ "]"
+        , ParenOpen <$ "("
+        , Token <$> anyUntil ")"
+        , ParenClose <$ ")"
+        ]
 
 link :: Lexer
 link = do
-  pos <- getPosition
-  l <- linkRaw
-  -- register the position of the last token
-  pos' <- flip incSourceColumn (-1) <$> getPosition
-  pure $ (pos, LinkOpen) : l <> [(pos', LinkClose)]
+    pos <- getPosition
+    l <- linkRaw
+    -- register the position of the last token
+    pos' <- flip incSourceColumn (-1) <$> getPosition
+    pure $ (pos, LinkOpen) : l <> [(pos', LinkClose)]
 
 labeledLink :: Lexer
 labeledLink = do
-  open <- located $ LabeledLinkOpen <$ "<"
-  linkRes <- linkRaw
-  labelRes <- located $ Token <$> anyUntil ">"
-  close <- located $ LabeledLinkClose <$ ">"
-  pure $
-    open : linkRes <> [labelRes, close]
+    open <- located $ LabeledLinkOpen <$ "<"
+    linkRes <- linkRaw
+    labelRes <- located $ Token <$> anyUntil ">"
+    close <- located $ LabeledLinkClose <$ ">"
+    pure $
+        open : linkRes <> [ labelRes , close ]
 
 mathMultiline :: Lexer
 mathMultiline = delimited "\\[" "\\]" MathMultilineOpen MathMultilineClose
@@ -257,49 +254,25 @@ bold = delimitedSymmetric "__" BoldOpen BoldClose
 monospace :: Lexer
 monospace = delimitedSymmetric "@" MonospaceOpen MonospaceClose
 
-decimal :: Parser Int
-decimal = read . Text.unpack <$> takeWhile1_ isDigit
-
-hexadecimal :: Parser Int
-hexadecimal = "x" *> (convert 0 . fmap (normalise . toLower) <$> many1 hexDigit)
- where
-  normalise :: Char -> Int
-  normalise c
-    | ord '0' <= n && n <= ord '9' = n - ord '0'
-    | ord 'A' <= n && n <= ord 'F' = n - ord 'A' + 10
-    | ord 'a' <= n && n <= ord 'f' = n - ord 'a' + 10
-    | otherwise = error "unexpected: invalid hex number"
-   where
-    n = ord c
-
-  convert :: Int -> [Int] -> Int
-  convert acc [] = acc
-  convert acc (x : xs) = convert (acc * 16 + x) xs
-
-numericEntity :: Lexer
-numericEntity = do
-  x <- located $ between "&#" ";" (NumericEntity <$> (hexadecimal <|> decimal))
-  pure [x]
-
 other :: Lexer
 other = do
-  pos <- getPosition
-  c <- takeWhile1_ isUnicodeAlphaNum
-  pure . pure $ (pos, Token c)
- where
-  isUnicodeAlphaNum c = isPrint c && not (isControl c) && not (isSpace c)
+    pos <- getPosition
+    c <- takeWhile1_ isUnicodeAlphaNum
+    pure . pure $ (pos, Token c)
+  where
+    isUnicodeAlphaNum c = isPrint c && not (isControl c) && not (isSpace c)
 
 spaceToken :: Lexer
 spaceToken = do
-  pos <- getPosition
-  _ <- many1 (char ' ')
-  pure . pure $ (pos, Space)
+    pos <- getPosition
+    _ <- many1 (char ' ')
+    pure . pure $ (pos, Space)
 
 newlineToken :: Lexer
 newlineToken = do
-  pos <- getPosition
-  _ <- newline
-  pure . pure $ (pos, Newline)
+    pos <- getPosition
+    _ <- newline
+    pure . pure $ (pos, Newline)
 
 -------
 -- Helpers
@@ -308,11 +281,11 @@ newlineToken = do
 -- | Like `takeWhile`, but unconditionally take escaped characters.
 takeWhile_ :: (Char -> Bool) -> Parser Text
 takeWhile_ p = scan p_ False
- where
-  p_ escaped c
-    | escaped = Just False
-    | not $ p c = Nothing
-    | otherwise = Just (c == '\\')
+  where
+    p_ escaped c
+        | escaped = Just False
+        | not $ p c = Nothing
+        | otherwise = Just (c == '\\')
 
 -- | Like 'takeWhile1', but unconditionally take escaped characters.
 takeWhile1_ :: (Char -> Bool) -> Parser Text
@@ -322,20 +295,19 @@ takeWhile1_ = mfilter (not . Text.null) . takeWhile_
 function returns true.
 -}
 scan ::
-  -- | scan function
-  (state -> Char -> Maybe state) ->
-  -- | initial state
-  state ->
-  Parser Text
+    -- | scan function
+    (state -> Char -> Maybe state) ->
+    -- | initial state
+    state ->
+    Parser Text
 scan f initState = do
-  parserState@State{stateInput = input, statePos = pos} <- Parsec.getParserState
-
-  (remaining, finalPos, ct) <- go input initState pos 0
-  let newState = parserState{stateInput = remaining, statePos = finalPos}
-  Parsec.setParserState newState $> Text.take ct input
- where
-  go !input' !st !posAccum !count' = case Text.uncons input' of
-    Nothing -> pure (input', posAccum, count')
-    Just (char', input'') -> case f st char' of
-      Nothing -> pure (input', posAccum, count')
-      Just st' -> go input'' st' (updatePosChar posAccum char') (count' + 1)
+    parserState@State{stateInput = input, statePos = pos} <- Parsec.getParserState
+    (remaining, finalPos, ct) <- go input initState pos 0
+    let newState = parserState{stateInput = remaining, statePos = finalPos}
+    Parsec.setParserState newState $> Text.take ct input
+  where
+    go !input' !st !posAccum !count' = case Text.uncons input' of
+        Nothing -> pure (input', posAccum, count')
+        Just (char', input'') -> case f st char' of
+            Nothing -> pure (input', posAccum, count')
+            Just st' -> go input'' st' (updatePosChar posAccum char') (count' + 1)

src/Parser/Util.hs

@@ -13,9 +13,7 @@ import Text.Parsec.Pos (updatePosChar)
 Return everything consumed except for the end pattern itself.
 -}
 takeUntil :: Text -> Parser Text
-takeUntil end_ =
-  requireEnd (scan p (False, end))
-    >>= gotSome . Text.dropEnd (Text.length end_)
+takeUntil end_ = Text.dropEnd (Text.length end_) <$> requireEnd (scan p (False, end)) >>= gotSome
  where
   end = Text.unpack end_
 

src/Types.hs

@@ -9,8 +9,6 @@ module Types (
 )
 where
 
-import Data.Foldable (fold)
-
 newtype Document = Document
   { meta :: Meta
   }
@@ -30,7 +28,6 @@ data Since = Since
 
 -- Could have a better type?
 type Version = [Int]
-
 type Package = String
 
 data DocMarkup mod id
@@ -54,78 +51,78 @@ data DocMarkup mod id
   | -- | Bold __bold text__
     DocBold (DocMarkup mod id)
   | {- | Unordered lists
-     * this
-     or
-     - this
+      * this
+      or
+      - this
     -}
     DocUnorderedList [DocMarkup mod id]
   | {- | Ordered lists
-     1. this
-     or
-     (1) this
+      1. this
+      or
+      (1) this
     -}
     DocOrderedList [(Int, DocMarkup mod id)]
   | {- | Definition lists
-     [term] a term
-     [another term] another definition
+      [term] a term
+      [another term] another definition
     -}
     DocDefinitionList [(DocMarkup mod id, DocMarkup mod id)]
   | {- | Code blocks
-     @
-     a code block in here
-     with multiple lines
-     @
+      @
+      a code block in here
+      with multiple lines
+      @
 
-     Or with bird tracks:
-     > some code
-     > goes here
+      Or with bird tracks:
+      > some code
+      > goes here
     -}
     DocCodeBlock (DocMarkup mod id)
   | {- | Hyperlinks
-     __marked__:
-     <http://example.com>
-     <http://example.com label text>
-     __Auto-detected URLs__:
-     http://example.com
-     https://example.com
-     ftp://example.com
-     __Markdown style__
-     [link text](http://example.com)
-     [link text]("Module.Name")
+      __marked__:
+      <http://example.com>
+      <http://example.com label text>
+      __Auto-detected URLs__:
+      http://example.com
+      https://example.com
+      ftp://example.com
+      __Markdown style__
+      [link text](http://example.com)
+      [link text]("Module.Name")
     -}
     DocHyperlink (Hyperlink (DocMarkup mod id))
   | {- | Pictures
-       <<image.png>>
-       <<image.png title text>>
+        <<image.png>>
+        <<image.png title text>>
 
-       __Markdown Images__
+        __Markdown Images__
 
-       ![alt text](image.png)
+        ![alt text](image.png)
     -}
     DocPicture Picture
   | {- | Inline math expressions
-     \(mathematical expression\)
+      \(mathematical expression\)
     -}
     DocMathInline String
   | {- | Math multiline display
-     \[
-     mathematical expression
-     in multiple lines
-     \]
+    \[
+      mathematical expression
+      in multiple lines
+    \]
     -}
     DocMathDisplay String
   | {- | Anchors, no spaces allowed
-     #anchor-name#
+      #anchor-name#
     -}
     DocAnchor String
   | {- | Property descriptions
-     prop> property description
+      prop> property description
     -}
     DocProperty String
   | {- | Examples
-     >>> expression
-     result line 1
-     result line 2
+      >>> expression
+      result line 1
+      result line 2
     -}
     DocExamples [Example]
   | -- | Header
@@ -139,7 +136,7 @@ instance Semigroup (DocMarkup mod id) where
 
 instance Monoid (DocMarkup mod id) where
   mempty = DocEmpty
-  mconcat = fold
+  mconcat = foldr (<>) mempty
 
 data ModuleLink id = ModuleLink
   { name :: String

test/Spec.hs

@@ -1,42 +1,42 @@
 {-# LANGUAGE OverloadedStrings #-}
 {-# OPTIONS_GHC -Wno-orphans #-}
 
-import Data.String (IsString (..))
-import Data.Text (Text)
-import GHC.Stack
+import Test.Hspec
+
 import Identifier (Identifier)
 import Lexer
 import Parser
 import Types
 
-import Test.Hspec
+import Data.String (IsString (..))
+import Data.Text (Text)
 import Text.Parsec.Pos
+import GHC.Stack
 
 main :: IO ()
 main = hspec $ do
-  describe "Lexer" do
-    describe "minimal" do
-      it "handles unicode" unicode
-      it "escapes" escaping
-      it "maths" math
-      it "anchors" anchor
-      it "space chars" space
-      it "bare string" someString
-      it "emphasis" emphatic
-      it "monospace" monospace
-      it "labeled link" labeledLink
-      it "markdown link" link
-      it "bird tracks" birdTracks
-      it "module names" modules
-      it "quotes" quotes
-      it "numeric entity" numericEntity
-      it "ignores nesting" ignoreNesting
+    describe "Lexer" do
+        describe "minimal" do
+            it "handles unicode" unicode
+            it "escapes" escaping
+            it "maths" math
+            it "anchors" anchor
+            it "space chars" space
+            it "bare string" someString
+            it "emphasis" emphatic
+            it "monospace" monospace
+            it "labeled link" labeledLink
+            it "markdown link" link
+            it "bird tracks" birdTracks
+            it "module names" modules
+            it "quotes" quotes
+            it "ignores nesting" ignoreNesting
 
-  describe "Parser" do
-    it "Bold" do
-      "__bold__" `shouldParseTo` DocBold (DocString "bold")
-    it "Emphasis" do
-      "/emphasis/" `shouldParseTo` DocEmphasis (DocString "emphasis")
+    describe "Parser" do
+        it "Bold" do
+            "__bold__" `shouldParseTo` (DocBold (DocString "bold"))
+        it "Emphasis" do
+            "/emphasis/" `shouldParseTo` (DocEmphasis (DocString "emphasis"))
 
 ------------
 -- Tests
@@ -44,150 +44,137 @@ main = hspec $ do
 
 modules :: Expectation
 modules = do
-  "\"MyModule.Name\""
-    `shouldLexTo` [ (1, 2, Module "MyModule.Name")
-                  ]
+    "\"MyModule.Name\""
+        `shouldLexTo` [ (1, 2, Module "MyModule.Name")
+                      ]
 
-  "\"OtherModule.Name#myAnchor\""
-    `shouldLexTo` [ (1, 2, Module "OtherModule.Name")
-                  , (1, 18, Anchor "myAnchor")
-                  ]
-
-  "\"OtherModule.Name\\#myAnchor\""
-    `shouldLexTo` [ (1, 2, Module "OtherModule.Name")
-                  , (1, 18, Anchor "myAnchor")
-                  ]
+    "\"OtherModule.Name#myAnchor\""
+        `shouldLexTo` [ (1, 2, Module "OtherModule.Name")
+                      , (1, 18, Anchor "myAnchor")
+                      ]
 
+    "\"OtherModule.Name\\#myAnchor\""
+        `shouldLexTo` [ (1, 2, Module "OtherModule.Name")
+                      , (1, 18, Anchor "myAnchor")
+                      ]
 link :: Expectation
 link =
-  "[link to](http://some.website)"
-    `shouldLexTo` [ (1, 1, LinkOpen)
-                  , (1, 1, BracketOpen)
-                  , (1, 2, Token "link to")
-                  , (1, 9, BracketClose)
-                  , (1, 10, ParenOpen)
-                  , (1, 11, Token "http://some.website")
-                  , (1, 30, ParenClose)
-                  , (1, 30, LinkClose)
-                  ]
+    "[link to](http://some.website)"
+        `shouldLexTo` [ (1, 1, LinkOpen)
+                      , (1, 1, BracketOpen)
+                      , (1, 2, Token "link to")
+                      , (1, 9, BracketClose)
+                      , (1, 10, ParenOpen)
+                      , (1, 11, Token "http://some.website")
+                      , (1, 30, ParenClose)
+                      , (1, 30, LinkClose)
+                      ]
 
 labeledLink :: Expectation
 labeledLink =
-  "<[link here](http://to.here) label>"
-    `shouldLexTo` [ (1, 1, LabeledLinkOpen)
-                  , (1, 2, BracketOpen)
-                  , (1, 3, Token "link here")
-                  , (1, 12, BracketClose)
-                  , (1, 13, ParenOpen)
-                  , (1, 14, Token "http://to.here")
-                  , (1, 28, ParenClose)
-                  , (1, 29, Token " label")
-                  , (1, 35, LabeledLinkClose)
-                  ]
+    "<[link here](http://to.here) label>"
+        `shouldLexTo` [ (1, 1, LabeledLinkOpen)
+                      , (1, 2, BracketOpen)
+                      , (1, 3, Token "link here")
+                      , (1, 12, BracketClose)
+                      , (1, 13, ParenOpen)
+                      , (1, 14, Token "http://to.here")
+                      , (1, 28, ParenClose)
+                      , (1, 29, Token " label")
+                      , (1, 35, LabeledLinkClose)
+                      ]
 
 anchor :: Expectation
 anchor =
-  "#myAnchor#"
-    `shouldLexTo` [ (1, 1, Anchor "myAnchor")
-                  ]
+    "#myAnchor#"
+        `shouldLexTo` [ (1, 1, Anchor "myAnchor")
+                      ]
 
 math :: IO ()
 math = do
-  "\\[some math\\]"
-    `shouldLexTo` [ (1, 1, MathMultilineOpen)
-                  , (1, 3, Token "some math")
-                  , (1, 12, MathMultilineClose)
-                  ]
-  "\\(other maths\\)"
-    `shouldLexTo` [ (1, 1, MathInlineOpen)
-                  , (1, 3, Token "other maths")
-                  , (1, 14, MathInlineClose)
-                  ]
+    "\\[some math\\]"
+        `shouldLexTo` [ (1, 1, MathMultilineOpen)
+                      , (1, 3, Token "some math")
+                      , (1, 12, MathMultilineClose)
+                      ]
+    "\\(other maths\\)"
+        `shouldLexTo` [ (1, 1, MathInlineOpen)
+                      , (1, 3, Token "other maths")
+                      , (1, 14, MathInlineClose)
+                      ]
 
 escaping :: Expectation
 escaping = do
-  "\\("
-    `shouldLexTo` [ (1, 1, Escape)
-                  , (1, 2, Token "(")
-                  ]
-  "\\(\r\n"
-    `shouldLexTo` [ (1, 1, Escape)
-                  , (1, 2, Token "(")
-                  ]
+    "\\("
+        `shouldLexTo` [ (1, 1, Escape)
+                      , (1, 2, Token "(")
+                      ]
+    "\\(\r\n"
+        `shouldLexTo` [ (1, 1, Escape)
+                      , (1, 2, Token "(")
+                      ]
 
 unicode :: Expectation
 unicode =
-  "ドラゴンクエストの冒険者🐉"
-    `shouldLexTo` [ (1, 1, Token "ドラゴンクエストの冒険者🐉")
-                  ]
+    "ドラゴンクエストの冒険者🐉"
+        `shouldLexTo` [ (1, 1, Token "ドラゴンクエストの冒険者🐉")
+                      ]
 
 ignoreNesting :: Expectation
 ignoreNesting =
-  ">/foo/"
-    `shouldLexTo` [ (1, 1, Token ">/foo/")
-                  ]
+    ">/foo/"
+        `shouldLexTo` [ (1, 1, Token ">/foo/")
+                      ]
 
 birdTracks :: Expectation
 birdTracks =
-  ">> code"
-    `shouldLexTo` [ (1, 1, BirdTrack)
-                  , (1, 4, Token "code")
-                  ]
+    ">> code"
+        `shouldLexTo` [ (1, 1, BirdTrack)
+                      , (1, 4, Token "code")
+                      ]
 
 quotes :: Expectation
 quotes =
-  "\"quoted\""
-    `shouldLexTo` [ (1, 1, QuoteOpen)
-                  , (1, 2, Token "quoted")
-                  , (1, 8, QuoteClose)
-                  ]
+    "\"quoted\""
+        `shouldLexTo` [ (1, 1, QuoteOpen)
+                      , (1, 2, Token "quoted")
+                      , (1, 8, QuoteClose)
+                      ]
 
 space :: Expectation
 space = do
-  "\n "
-    `shouldLexTo` [ (1, 1, Newline)
-                  , (2, 1, Space)
-                  ]
-  " \n"
-    `shouldLexTo` [ (1, 1, Space)
-                  , (1, 2, Newline)
-                  ]
-
-numericEntity :: Expectation
-numericEntity = do
-  "&#65; &#955;"
-    `shouldLexTo` [ (1, 1, NumericEntity 65)
-                  , (1, 6, Space)
-                  , (1, 7, NumericEntity 955) -- lambda
-                  ]
-  -- Hex
-  "&#x65;"
-    `shouldLexTo` [ (1, 1, NumericEntity 101)
-                  ]
+    "\n "
+        `shouldLexTo` [ (1, 1, Newline)
+                      , (2, 1, Space)
+                      ]
+    " \n"
+        `shouldLexTo` [ (1, 1, Space)
+                      , (1, 2, Newline)
+                      ]
 
 monospace :: Expectation
 monospace =
-  "@mono@"
-    `shouldLexTo` [ (1, 1, MonospaceOpen)
-                  , (1, 2, Token "mono")
-                  , (1, 6, MonospaceClose)
-                  ]
+    "@mono@"
+        `shouldLexTo` [ (1, 1, MonospaceOpen)
+                      , (1, 2, Token "mono")
+                      , (1, 6, MonospaceClose)
+                      ]
 
 emphatic :: Expectation
 emphatic =
-  "/emphatic/"
-    `shouldLexTo` [ (1, 1, EmphasisOpen)
-                  , (1, 2, Token "emphatic")
-                  , (1, 10, EmphasisClose)
-                  ]
+    "/emphatic/"
+        `shouldLexTo` [ (1, 1, EmphasisOpen)
+                      , (1, 2, Token "emphatic")
+                      , (1, 10, EmphasisClose)
+                      ]
 
 someString :: Expectation
 someString =
-  "some string"
-    `shouldLexTo` [ (1, 1, Token "some")
-                  , (1, 5, Space)
-                  , (1, 6, Token "string")
-                  ]
+    "some string"
+        `shouldLexTo` [ (1, 1, Token "some")
+                      , (1, 5, Space)
+                      , (1, 6, Token "string")
+                      ]
 
 --------------
 -- Helpers
@@ -196,16 +183,16 @@ someString =
 type Doc id = DocMarkup () id
 
 instance IsString (Doc String) where
-  fromString = DocString
+    fromString = DocString
 
 shouldLexTo :: String -> [(Int, Int, Token)] -> Expectation
 shouldLexTo input expected =
-  withFrozenCallStack $
+   withFrozenCallStack $
     case lexer input of
-      Right tokens -> do
-        let actual = map (\(pos, tok) -> (sourceLine pos, sourceColumn pos, tok)) tokens
-        actual `shouldBe` expected
-      Left err -> expectationFailure $ "Parse error: " <> show err
+        Right tokens -> do
+            let actual = map (\(pos, tok) -> (sourceLine pos, sourceColumn pos, tok)) tokens
+            actual `shouldBe` expected
+        Left err -> expectationFailure $ "Parse error: " <> show err
 
 shouldParseTo :: Text -> DocMarkup mod Identifier -> Expectation
 shouldParseTo input ast = parseText input `shouldBe` ast