Compare commits

...
Sign in to create a new pull request.

11 commits
main ... dev

Author SHA1 Message Date
1664694134 chore: add hls (#10)
Reviewed-on: #10
Co-authored-by: Igor Ranieri <igor@elland.me>
Co-committed-by: Igor Ranieri <igor@elland.me>
2025-10-05 13:18:49 +00:00
ebda9e1d12 nixify + npins (#9)
Reviewed-on: #9
Co-authored-by: Igor Ranieri <igor@elland.me>
Co-committed-by: Igor Ranieri <igor@elland.me>
2025-10-05 11:06:33 +00:00
82eb8435ab feat(lexer): implement numericEntity lexer (#6)
I did this yesterday actually, I just rebased it.

Reviewed-on: #6
Reviewed-by: elland <igor@elland.me>
Co-authored-by: Léana 江 <leana.jiang+git@icloud.com>
Co-committed-by: Léana 江 <leana.jiang+git@icloud.com>
2025-09-27 15:10:11 +00:00
08dc87a307 Add CI runner (#5)
Reviewed-on: #5
Co-authored-by: Igor Ranieri <igor@elland.me>
Co-committed-by: Igor Ranieri <igor@elland.me>
2025-09-27 15:03:00 +00:00
d8ba47a8b6 Applied more formatting; added Makefile (#4)
Reviewed-on: #4
Co-authored-by: Igor Ranieri <igor@elland.me>
Co-committed-by: Igor Ranieri <igor@elland.me>
2025-09-27 07:51:09 +00:00
f4912d3339 Format with fourmolu (#3)
Add the formatting config (similar to default but with indent 2)

Reviewed-on: #3
Co-authored-by: Léana 江 <leana.jiang+git@icloud.com>
Co-committed-by: Léana 江 <leana.jiang+git@icloud.com>
2025-09-26 14:52:34 +00:00
6064afd0b9 refactor lexer (#2)
Reduce manual usage of getPosition and setting it and improve some helper functions.

Reviewed-on: #2
Reviewed-by: elland <igor@elland.me>
Co-authored-by: Léana 江 <leana.jiang+git@icloud.com>
Co-committed-by: Léana 江 <leana.jiang+git@icloud.com>
2025-09-26 14:44:46 +00:00
39cfe2035d test: freeze the callstack for better failure messages (#1)
This would make the function `shouldLexTo` be transparent in the error reporting! Before all errors would point to this line.

Reviewed-on: #1
Co-authored-by: Léana 江 <leana.jiang+git@icloud.com>
Co-committed-by: Léana 江 <leana.jiang+git@icloud.com>
2025-09-24 11:01:52 +00:00
00a6e11f67 Tokenise helper 2025-09-24 10:41:09 +02:00
7ae868932d Add located combinator; improved anchor matches with tests 2025-09-24 10:13:16 +02:00
8887476626 Embed Anchor content inside token 2025-09-24 09:48:01 +02:00
17 changed files with 663 additions and 195 deletions

1
.envrc Normal file
View file

@ -0,0 +1 @@
use nix

View file

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

View file

@ -0,0 +1,99 @@
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

2
.git-blame-ignore-revs Normal file
View file

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

36
Dockerfile Normal file
View file

@ -0,0 +1,36 @@
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" ]

View file

@ -12,7 +12,7 @@ bold ::= '__' text_no_newline '__'
monospace ::= '@' text_content '@' monospace ::= '@' text_content '@'
link ::= module_link | hyperlink | markdown_link link ::= module_link | hyperlink | markdown_link
module_link ::= '"' module_name ( '#' anchor_name )? '"' module_link ::= '"' module_name ( ('#' | '\#') anchor_name )? '"'
hyperlink ::= '<' url ( ' ' link_text )? '>' hyperlink ::= '<' url ( ' ' link_text )? '>'
markdown_link ::= '[' link_text '](' ( url | module_link ) ')' markdown_link ::= '[' link_text '](' ( url | module_link ) ')'

44
Makefile Normal file
View file

@ -0,0 +1,44 @@
.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

72
fourmolu.yaml Normal file
View file

@ -0,0 +1,72 @@
# 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: []

View file

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

146
npins/default.nix Normal file
View file

@ -0,0 +1,146 @@
/*
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`"

11
npins/sources.json Normal file
View file

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

25
shell.nix Normal file
View file

@ -0,0 +1,25 @@
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;
}

View file

@ -4,18 +4,22 @@ module Lexer (
Token (..), Token (..),
lexer, lexer,
emphasis, emphasis,
) where )
where
import Control.Monad (mfilter, void) import Control.Monad (mfilter, void)
import Data.Char (ord, toLower)
import Data.Functor (($>)) import Data.Functor (($>))
import Data.Text (Text, intercalate) import Data.Text (Text, intercalate)
import Data.Text qualified as Text import Data.Text qualified as Text
import GHC.Unicode (isAlpha, isAlphaNum, isControl, isPrint, isSpace, isUpper) import GHC.Unicode (isAlphaNum, isControl, isDigit, isPrint, isSpace, isUpper)
import ParserMonad (Parser, initialParserState) import ParserMonad (Parser, initialParserState)
import Text.Parsec import Text.Parsec
import Text.Parsec qualified as Parsec import Text.Parsec qualified as Parsec
import Text.Parsec.Pos (updatePosChar) import Text.Parsec.Pos (updatePosChar)
type Located a = (SourcePos, a)
type LocatedToken = (SourcePos, Token) type LocatedToken = (SourcePos, Token)
type Lexer = Parser [LocatedToken] type Lexer = Parser [LocatedToken]
@ -31,7 +35,7 @@ data Level
data Token data Token
= Token Text = Token Text
| Anchor | Anchor Text
| BirdTrack | BirdTrack
| BoldOpen | BoldOpen
| BoldClose | BoldClose
@ -50,17 +54,24 @@ data Token
| ParenClose | ParenClose
| BracketOpen | BracketOpen
| BracketClose | BracketClose
| MathsParenOpen | MathInlineOpen
| MathsParenClose | MathInlineClose
| MathsBracketOpen | MathMultilineOpen
| MathsBracketClose | MathMultilineClose
| Module | NumericEntity Int
| Module Text
| QuoteOpen | QuoteOpen
| QuoteClose | QuoteClose
| Space | Space
| EOF | EOF
deriving (Eq, Show) deriving (Eq, Show)
located :: Parser a -> Parser (SourcePos, a)
located p = (,) <$> getPosition <*> p
tokenise :: [Parser a] -> Parser [(SourcePos, a)]
tokenise = mapM located
lexer :: String -> Either ParseError [LocatedToken] lexer :: String -> Either ParseError [LocatedToken]
lexer = Parsec.runParser lexText initialParserState "input" . Text.pack lexer = Parsec.runParser lexText initialParserState "input" . Text.pack
@ -74,16 +85,17 @@ lexText = go
toks <- toks <-
choice $ choice $
Parsec.try Parsec.try
<$> [ mathsBracket <$> [ mathMultiline
, mathsParens , mathInline
, escape -- maths go before escape to avoid mismatch , escape -- maths go before escape to avoid mismatch
, headers , headers
, newlineToken , newlineToken
, spaceToken , spaceToken
, link , link
, labeledLink , labeledLink
, modules , module_
, anchors , anchor
, numericEntity
, textElement , textElement
, quotes , quotes
, birdTrack , birdTrack
@ -115,161 +127,123 @@ headers =
, header6 , header6
] ]
delimitedMaybe :: Parser a -> Parser a -> Token -> Maybe Token -> Parser [LocatedToken]
delimitedMaybe openMark closeMark openToken closeToken = do
openPos <- getPosition
void openMark
tokenPos <- getPosition
content <- anyUntil closeMark
closePos <- getPosition
void closeMark
let openTok :: LocatedToken = (openPos, openToken)
res :: LocatedToken = (tokenPos, Token content)
closeToks :: [LocatedToken] = case closeToken of
Just close -> [(closePos, close)]
Nothing -> []
pure $ [openTok, res] <> closeToks
anyUntil :: Parser a -> Parser Text anyUntil :: Parser a -> Parser Text
anyUntil p = Text.pack <$> manyTill anyChar (lookAhead p) anyUntil p = Text.pack <$> manyTill anyChar (lookAhead p)
delimited :: Parser a -> Parser a -> Token -> Token -> Parser [LocatedToken] delimitedAsTuple :: Parser open -> Parser close -> Parser (Located open, LocatedToken, Located close)
delimited a b c d = delimitedMaybe a b c (Just d) delimitedAsTuple openP 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]
delimitedNoTrailing :: Parser open -> Parser close -> Token -> Parser [LocatedToken]
delimitedNoTrailing openP closeP openTok = asList <$> delimitedAsTuple (openTok <$ openP) (void closeP)
where
asList (a, tok, _) = [a, tok]
delimitedSymmetric :: Parser a -> Token -> Token -> Parser [LocatedToken] delimitedSymmetric :: Parser a -> Token -> Token -> Parser [LocatedToken]
delimitedSymmetric s t1 t2 = delimited s s t1 t2 delimitedSymmetric s = delimited s s
eol :: Parser () eol :: Parser ()
eol = void "\n" <|> Parsec.eof eol = void "\n" <|> void "\r\n" <|> Parsec.eof
header1 :: Lexer header1 :: Lexer
header1 = delimitedMaybe (void $ "= ") eol (Header One) Nothing header1 = delimitedNoTrailing "= " eol (Header One)
header2 :: Lexer header2 :: Lexer
header2 = delimitedMaybe (void $ "== ") eol (Header Two) Nothing header2 = delimitedNoTrailing "== " eol (Header Two)
header3 :: Lexer header3 :: Lexer
header3 = delimitedMaybe (void $ "=== ") eol (Header Three) Nothing header3 = delimitedNoTrailing "=== " eol (Header Three)
header4 :: Lexer header4 :: Lexer
header4 = delimitedMaybe (void $ "==== ") eol (Header Four) Nothing header4 = delimitedNoTrailing "==== " eol (Header Four)
header5 :: Lexer header5 :: Lexer
header5 = delimitedMaybe (void $ "===== ") eol (Header Five) Nothing header5 = delimitedNoTrailing "===== " eol (Header Five)
header6 :: Lexer header6 :: Lexer
header6 = delimitedMaybe (void $ "====== ") eol (Header Six) Nothing header6 = delimitedNoTrailing "====== " eol (Header Six)
-- #anchors# -- #anchors#
anchors :: Lexer anchor :: Lexer
anchors = do anchor = do
pos <- getPosition x <- located $ between "#" "#" (Anchor <$> anyUntil "#")
void $ try anchor' pure [x]
pos' <- getPosition
txt <- anyUntil anchor'
pos'' <- getPosition
void $ try anchor'
pure moduleNames :: Parser Text
[ (pos, Anchor) moduleNames = intercalate "." . fmap Text.pack <$> upperId `sepBy1` char '.'
, (pos', Token txt)
, (pos'', Anchor) upperId :: Parser String
] upperId = (:) <$> satisfy isUpper <*> many1 identifierChar
where
anchor' = (string "#" <|> string "\\#") identifierChar :: Parser Char
identifierChar = satisfy (\c -> isAlphaNum c || c == '_')
-- "Module.Name" -- "Module.Name"
-- "Module.Name#anchor" -- "Module.Name#anchor"
-- "Module.Name#anchor" -- "Module.Name\#anchor" -- known as "old anchor". this has been deprecated for 9 years, thanks Ben
modules :: Lexer module_ :: Lexer
modules = do module_ = between (char '"') (char '"') inner
pos <- getPosition
void $ char '"'
pos' <- getPosition
modName <- modId
anch <- option [] do
pos'' <- getPosition
void $ try (string "#" <|> string "\\#")
pos''' <- getPosition
a <- Text.pack <$> many (satisfy (\c -> c /= '"' && not (isSpace c)))
pure [(pos'', Anchor), (pos''', Token a)]
void $ char '"'
pure $ [(pos, Module), (pos', Token modName)] <> anch
where where
modId = intercalate "." <$> (fmap Text.pack <$> (conId `sepBy1` (char '.'))) inner = do
m <- located $ Module <$> moduleNames
mAnchor <- optionMaybe (located $ anchorHash *> (Anchor <$> anchorText))
pure $ case mAnchor of
Just anc -> [m, anc]
Nothing -> [m]
conId :: Parser String anchorHash :: Parser Text
conId = anchorHash = "#" <|> try "\\#"
(:)
<$> satisfy (\c -> isAlpha c && isUpper c)
<*> many1 conChar
conChar :: Parser Char anchorText :: Parser Text
conChar = satisfy (\c -> isAlphaNum c || c == '_') anchorText = Text.pack <$> many (satisfy (\c -> c /= '"' && not (isSpace c)))
linkRaw :: Lexer linkRaw :: Lexer
linkRaw = do linkRaw =
pos1 <- getPosition tokenise
void $ string "[" [ BracketOpen <$ "["
pos2 <- getPosition , Token <$> anyUntil "]"
text <- anyUntil $ Text.pack <$> string "]" , BracketClose <$ "]"
pos3 <- getPosition , ParenOpen <$ "("
void $ "]" , Token <$> anyUntil ")"
pos4 <- getPosition , ParenClose <$ ")"
void $ "("
pos5 <- getPosition
link' <- anyUntil $ Text.pack <$> string ")"
pos6 <- getPosition
void $ ")"
pure $
[ (pos1, BracketOpen)
, (pos2, Token text)
, (pos3, BracketClose)
, (pos4, ParenOpen)
, (pos5, Token link')
, (pos6, ParenClose)
] ]
link :: Lexer link :: Lexer
link = do link = do
pos <- getPosition pos <- getPosition
l <- linkRaw l <- linkRaw
-- "unconsume" the last token -- register the position of the last token
pos' <- flip incSourceColumn (-1) <$> getPosition pos' <- flip incSourceColumn (-1) <$> getPosition
pure $ (pos, LinkOpen) : l <> [(pos', LinkClose)] pure $ (pos, LinkOpen) : l <> [(pos', LinkClose)]
labeledLink :: Lexer labeledLink :: Lexer
labeledLink = do labeledLink = do
pos <- getPosition open <- located $ LabeledLinkOpen <$ "<"
void $ string "<" linkRes <- linkRaw
link' <- linkRaw labelRes <- located $ Token <$> anyUntil ">"
pos7 <- getPosition close <- located $ LabeledLinkClose <$ ">"
label' <- anyUntil $ string ">"
pos8 <- getPosition
void $ ">"
pure $ pure $
(pos, LabeledLinkOpen) open : linkRes <> [labelRes, close]
: link'
<> [ (pos7, Token label')
, (pos8, LabeledLinkClose)
]
mathsBracket :: Lexer mathMultiline :: Lexer
mathsBracket = delimited (void $ "\\[") (void "\\]") MathsBracketOpen MathsBracketClose mathMultiline = delimited "\\[" "\\]" MathMultilineOpen MathMultilineClose
mathsParens :: Lexer mathInline :: Lexer
mathsParens = delimited (void $ "\\(") (void "\\)") MathsParenOpen MathsParenClose mathInline = delimited "\\(" "\\)" MathInlineOpen MathInlineClose
birdTrack :: Lexer birdTrack :: Lexer
birdTrack = delimitedMaybe (void ">> ") eol BirdTrack Nothing birdTrack = delimitedNoTrailing ">> " eol BirdTrack
escape :: Lexer escape :: Lexer
escape = delimitedMaybe (void "\\") eol Escape Nothing escape = delimitedNoTrailing "\\" eol Escape
quotes :: Lexer quotes :: Lexer
quotes = delimitedSymmetric "\"" QuoteOpen QuoteClose quotes = delimitedSymmetric "\"" QuoteOpen QuoteClose
@ -283,6 +257,30 @@ bold = delimitedSymmetric "__" BoldOpen BoldClose
monospace :: Lexer monospace :: Lexer
monospace = delimitedSymmetric "@" MonospaceOpen MonospaceClose 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 :: Lexer
other = do other = do
pos <- getPosition pos <- getPosition
@ -331,6 +329,7 @@ scan ::
Parser Text Parser Text
scan f initState = do scan f initState = do
parserState@State{stateInput = input, statePos = pos} <- Parsec.getParserState parserState@State{stateInput = input, statePos = pos} <- Parsec.getParserState
(remaining, finalPos, ct) <- go input initState pos 0 (remaining, finalPos, ct) <- go input initState pos 0
let newState = parserState{stateInput = remaining, statePos = finalPos} let newState = parserState{stateInput = remaining, statePos = finalPos}
Parsec.setParserState newState $> Text.take ct input Parsec.setParserState newState $> Text.take ct input

View file

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

View file

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

View file

@ -1,15 +1,15 @@
{-# LANGUAGE OverloadedStrings #-} {-# LANGUAGE OverloadedStrings #-}
{-# OPTIONS_GHC -Wno-orphans #-} {-# OPTIONS_GHC -Wno-orphans #-}
import Test.Hspec import Data.String (IsString (..))
import Data.Text (Text)
import GHC.Stack
import Identifier (Identifier) import Identifier (Identifier)
import Lexer import Lexer
import Parser import Parser
import Types import Types
import Data.String (IsString (..)) import Test.Hspec
import Data.Text (Text)
import Text.Parsec.Pos import Text.Parsec.Pos
main :: IO () main :: IO ()
@ -18,8 +18,8 @@ main = hspec $ do
describe "minimal" do describe "minimal" do
it "handles unicode" unicode it "handles unicode" unicode
it "escapes" escaping it "escapes" escaping
it "maths" maths it "maths" math
it "anchors" anchors it "anchors" anchor
it "space chars" space it "space chars" space
it "bare string" someString it "bare string" someString
it "emphasis" emphatic it "emphasis" emphatic
@ -29,13 +29,14 @@ main = hspec $ do
it "bird tracks" birdTracks it "bird tracks" birdTracks
it "module names" modules it "module names" modules
it "quotes" quotes it "quotes" quotes
it "numeric entity" numericEntity
it "ignores nesting" ignoreNesting it "ignores nesting" ignoreNesting
describe "Parser" do describe "Parser" do
it "Bold" do it "Bold" do
"__bold__" `shouldParseTo` (DocBold (DocString "bold")) "__bold__" `shouldParseTo` DocBold (DocString "bold")
it "Emphasis" do it "Emphasis" do
"/emphasis/" `shouldParseTo` (DocEmphasis (DocString "emphasis")) "/emphasis/" `shouldParseTo` DocEmphasis (DocString "emphasis")
------------ ------------
-- Tests -- Tests
@ -44,15 +45,17 @@ main = hspec $ do
modules :: Expectation modules :: Expectation
modules = do modules = do
"\"MyModule.Name\"" "\"MyModule.Name\""
`shouldLexTo` [ (1, 1, Module) `shouldLexTo` [ (1, 2, Module "MyModule.Name")
, (1, 2, Token "MyModule.Name")
] ]
"\"OtherModule.Name#myAnchor\"" "\"OtherModule.Name#myAnchor\""
`shouldLexTo` [ (1, 1, Module) `shouldLexTo` [ (1, 2, Module "OtherModule.Name")
, (1, 2, Token "OtherModule.Name") , (1, 18, Anchor "myAnchor")
, (1, 18, Anchor) ]
, (1, 19, Token "myAnchor")
"\"OtherModule.Name\\#myAnchor\""
`shouldLexTo` [ (1, 2, Module "OtherModule.Name")
, (1, 18, Anchor "myAnchor")
] ]
link :: Expectation link :: Expectation
@ -82,33 +85,35 @@ labeledLink =
, (1, 35, LabeledLinkClose) , (1, 35, LabeledLinkClose)
] ]
anchors :: Expectation anchor :: Expectation
anchors = anchor =
"#myAnchor#" "#myAnchor#"
`shouldLexTo` [ (1, 1, Anchor) `shouldLexTo` [ (1, 1, Anchor "myAnchor")
, (1, 2, Token "myAnchor")
, (1, 10, Anchor)
] ]
maths :: IO () math :: IO ()
maths = do math = do
"\\[some math\\]" "\\[some math\\]"
`shouldLexTo` [ (1, 1, MathsBracketOpen) `shouldLexTo` [ (1, 1, MathMultilineOpen)
, (1, 3, Token "some math") , (1, 3, Token "some math")
, (1, 12, MathsBracketClose) , (1, 12, MathMultilineClose)
] ]
"\\(other maths\\)" "\\(other maths\\)"
`shouldLexTo` [ (1, 1, MathsParenOpen) `shouldLexTo` [ (1, 1, MathInlineOpen)
, (1, 3, Token "other maths") , (1, 3, Token "other maths")
, (1, 14, MathsParenClose) , (1, 14, MathInlineClose)
] ]
escaping :: Expectation escaping :: Expectation
escaping = escaping = do
"\\(" "\\("
`shouldLexTo` [ (1, 1, Escape) `shouldLexTo` [ (1, 1, Escape)
, (1, 2, Token "(") , (1, 2, Token "(")
] ]
"\\(\r\n"
`shouldLexTo` [ (1, 1, Escape)
, (1, 2, Token "(")
]
unicode :: Expectation unicode :: Expectation
unicode = unicode =
@ -148,6 +153,18 @@ space = do
, (1, 2, Newline) , (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)
]
monospace :: Expectation monospace :: Expectation
monospace = monospace =
"@mono@" "@mono@"
@ -183,11 +200,12 @@ instance IsString (Doc String) where
shouldLexTo :: String -> [(Int, Int, Token)] -> Expectation shouldLexTo :: String -> [(Int, Int, Token)] -> Expectation
shouldLexTo input expected = shouldLexTo input expected =
case lexer input of withFrozenCallStack $
Right tokens -> do case lexer input of
let actual = map (\(pos, tok) -> (sourceLine pos, sourceColumn pos, tok)) tokens Right tokens -> do
actual `shouldBe` expected let actual = map (\(pos, tok) -> (sourceLine pos, sourceColumn pos, tok)) tokens
Left err -> expectationFailure $ "Parse error: " <> show err actual `shouldBe` expected
Left err -> expectationFailure $ "Parse error: " <> show err
shouldParseTo :: Text -> DocMarkup mod Identifier -> Expectation shouldParseTo :: Text -> DocMarkup mod Identifier -> Expectation
shouldParseTo input ast = parseText input `shouldBe` ast shouldParseTo input ast = parseText input `shouldBe` ast

View file

@ -27,8 +27,16 @@ ftp\://example.com
![alt text](image.png) ![alt text](image.png)
\(mathematical expression\) \(mathematical 1+3 expression\)
\[mathematical expression\]
\[mathematical
expression
accross lines with + addition and such
\]
&#123
&#x65
&#165
@ @
code block content code block content