MarkupParse

A Markup parser and printer of strict bytestrings. Markup is a representation of data such as HTML, SVG or XML but the parsing is sub-standard.

Documentation

import MarkupParse
import Data.ByteString qualified as B

bs <- B.readFile "other/line.svg"
m = markup_ bs

markdown . 'markup_ is an approximate round trip from ByteString to Markup back to ByteString'. The underscores represent versions of main functions that throw an exception on warnings encountered along the way.

At a lower level, a round trip pipeline might look something like:

:t tokenize Html >=> gather Html >>> fmap (Markup Html >>> normalize) >=> degather >>> fmap (fmap (detokenize Html) >>> mconcat)
ByteString -> These [MarkupWarning] ByteString

From left to right:

• tokenize converts a ByteString to a Token list,
• gather takes the tokens and gathers them into Trees of tokens
• this is then wrapped into the Markup data type.
• normalize concatenates content, and normalizes attributes,
• degather turns the markup tree back into a token list. Finally,
• detokenize turns a token back into a bytestring.

Along the way, the kleisi fishies and compose forward usage accumulates any warnings via the These monad instance.

Markup

data Markup

A Tree list of markup Tokens

>>> markup Html "<foo class=\"bar\">baz</foo>"
That (Markup {standard = Html, markupTree = [Node {rootLabel = StartTag "foo" [Attr "class" "bar"], subForest = [Node {rootLabel = Content "baz", subForest = []}]}]})

Constructors

Markup

Fields

• standard :: Standard
• markupTree :: [Tree Token]

Instances

Generic Markup

Defined in MarkupParse

Associated Types

type Rep Markup :: Type -> Type

Methods

from :: Markup -> Rep Markup x

to :: Rep Markup x -> Markup

Show Markup

Defined in MarkupParse

Methods

showsPrec :: Int -> Markup -> ShowS

show :: Markup -> String

showList :: [Markup] -> ShowS

NFData Markup

Defined in MarkupParse

Methods

rnf :: Markup -> ()

Eq Markup

Defined in MarkupParse

Methods

(==) :: Markup -> Markup -> Bool

(/=) :: Markup -> Markup -> Bool

Ord Markup

Defined in MarkupParse

Methods

compare :: Markup -> Markup -> Ordering

(<) :: Markup -> Markup -> Bool

(<=) :: Markup -> Markup -> Bool

(>) :: Markup -> Markup -> Bool

(>=) :: Markup -> Markup -> Bool

max :: Markup -> Markup -> Markup

min :: Markup -> Markup -> Markup

ToExpr Markup

Defined in MarkupParse

Methods

toExpr :: Markup -> Expr

listToExpr :: [Markup] -> Expr

type Rep Markup

Defined in MarkupParse

type Rep Markup = D1 ('MetaData "Markup" "MarkupParse" "markup-parse-0.0.0.2-HnQ2lGo9uJcI0zQ43aTniK" 'False) (C1 ('MetaCons "Markup" 'PrefixI 'True) (S1 ('MetaSel ('Just "standard") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedLazy) (Rec0 Standard) :*: S1 ('MetaSel ('Just "markupTree") 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedLazy) (Rec0 [Tree Token])))

data Standard

From a parsing pov, Html & Xml (& Svg) are close enough that they share a lot of parsing logic, so that parsing and printing just need some tweaking.

The xml parsing logic is based on the XML productions found in https://www.w3.org/TR/xml/

The html parsing was based on a reading of html-parse, but ignores the various 'x00' to 'xfffd' & eof directives that form part of the html standards.

Constructors

Html

Xml

Instances

Generic Standard

Defined in MarkupParse

Associated Types

type Rep Standard :: Type -> Type

Methods

from :: Standard -> Rep Standard x

to :: Rep Standard x -> Standard

Show Standard

Defined in MarkupParse

Methods

showsPrec :: Int -> Standard -> ShowS

show :: Standard -> String

showList :: [Standard] -> ShowS

NFData Standard

Defined in MarkupParse

Methods

rnf :: Standard -> ()

Eq Standard

Defined in MarkupParse

Methods

(==) :: Standard -> Standard -> Bool

(/=) :: Standard -> Standard -> Bool

Ord Standard

Defined in MarkupParse

Methods

compare :: Standard -> Standard -> Ordering

(<) :: Standard -> Standard -> Bool

(<=) :: Standard -> Standard -> Bool

(>) :: Standard -> Standard -> Bool

(>=) :: Standard -> Standard -> Bool

max :: Standard -> Standard -> Standard

min :: Standard -> Standard -> Standard

ToExpr Standard

Defined in MarkupParse

Methods

toExpr :: Standard -> Expr

listToExpr :: [Standard] -> Expr

type Rep Standard

Defined in MarkupParse

type Rep Standard = D1 ('MetaData "Standard" "MarkupParse" "markup-parse-0.0.0.2-HnQ2lGo9uJcI0zQ43aTniK" 'False) (C1 ('MetaCons "Html" 'PrefixI 'False) (U1 :: Type -> Type) :+: C1 ('MetaCons "Xml" 'PrefixI 'False) (U1 :: Type -> Type))

markup :: Standard -> ByteString -> These [MarkupWarning] Markup

Convert bytestrings to Markup

>>> markup Html "<foo><br></foo><baz"
These [MarkupParser (ParserLeftover "<baz")] (Markup {standard = Html, markupTree = [Node {rootLabel = StartTag "foo" [], subForest = [Node {rootLabel = StartTag "br" [], subForest = []}]}]})

markup_ :: Standard -> ByteString -> Markup

markup but errors on warnings.

data RenderStyle

Indented 0 puts newlines in between the tags.

Constructors

Compact

Indented Int

Instances

Generic RenderStyle

Defined in MarkupParse

Associated Types

type Rep RenderStyle :: Type -> Type

Methods

from :: RenderStyle -> Rep RenderStyle x

to :: Rep RenderStyle x -> RenderStyle

Show RenderStyle

Defined in MarkupParse

Methods

showsPrec :: Int -> RenderStyle -> ShowS

show :: RenderStyle -> String

showList :: [RenderStyle] -> ShowS

Eq RenderStyle

Defined in MarkupParse

Methods

(==) :: RenderStyle -> RenderStyle -> Bool

(/=) :: RenderStyle -> RenderStyle -> Bool

type Rep RenderStyle

Defined in MarkupParse

type Rep RenderStyle = D1 ('MetaData "RenderStyle" "MarkupParse" "markup-parse-0.0.0.2-HnQ2lGo9uJcI0zQ43aTniK" 'False) (C1 ('MetaCons "Compact" 'PrefixI 'False) (U1 :: Type -> Type) :+: C1 ('MetaCons "Indented" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedLazy) (Rec0 Int)))

markdown :: RenderStyle -> Markup -> ByteString

Convert Markup to bytestrings

>>> B.putStr $ markdown (Indented 4) (markup_ Html [i|<foo><br></foo>|])
<foo>
    <br>
</foo>

normalize :: Markup -> Markup

concatenate sequential content, and normalize attributes; unwording class values and removing duplicate attributes (taking last).

>>> B.putStr $ markdown Compact $ normalize (markup_ Xml [i|<foo class="a" class="b" bar="first" bar="last"/>|])
<foo bar="last" class="a b"/>

wellFormed :: Markup -> [MarkupWarning]

Check for well-formedness and rerturn warnings encountered.

>>> wellFormed $ Markup Html [Node (Comment "") [], Node (EndTag "foo") [], Node (EmptyElemTag "foo" []) [Node (Content "bar") []], Node (EmptyElemTag "foo" []) []]
[EmptyContent,EndTagInTree,LeafWithChildren,BadEmptyElemTag]

isWellFormed :: Markup -> Bool

Are the trees in the markup well-formed?

Warnings

data MarkupWarning

markup-parse generally tries to continue on parse errors, and return what has/can still be parsed, together with any warnings.

Constructors

BadEmptyElemTag

A tag ending with "/>" that is not an element of selfClosers (Html only).

SelfCloserWithChildren

A tag ending with "/>" that has children. Cannot happen in the parsing phase.

LeafWithChildren

Only a StartTag can have child tokens.

TagMismatch TagName TagName

A CloseTag with a different name to the currently open StartTag.

UnmatchedEndTag

An EndTag with no corresponding StartTag.

UnclosedTag

An EndTag with corresponding StartTag.

EndTagInTree

An EndTag should never appear in Markup

EmptyContent

Empty Content, Comment, Decl or Doctype

MarkupParser ParserWarning

Instances

Generic MarkupWarning

Defined in MarkupParse

Associated Types

type Rep MarkupWarning :: Type -> Type

Methods

from :: MarkupWarning -> Rep MarkupWarning x

to :: Rep MarkupWarning x -> MarkupWarning

Show MarkupWarning

Defined in MarkupParse

Methods

showsPrec :: Int -> MarkupWarning -> ShowS

show :: MarkupWarning -> String

showList :: [MarkupWarning] -> ShowS

NFData MarkupWarning

Defined in MarkupParse

Methods

rnf :: MarkupWarning -> ()

Eq MarkupWarning

Defined in MarkupParse

Methods

(==) :: MarkupWarning -> MarkupWarning -> Bool

(/=) :: MarkupWarning -> MarkupWarning -> Bool

Ord MarkupWarning

Defined in MarkupParse

Methods

compare :: MarkupWarning -> MarkupWarning -> Ordering

(<) :: MarkupWarning -> MarkupWarning -> Bool

(<=) :: MarkupWarning -> MarkupWarning -> Bool

(>) :: MarkupWarning -> MarkupWarning -> Bool

(>=) :: MarkupWarning -> MarkupWarning -> Bool

max :: MarkupWarning -> MarkupWarning -> MarkupWarning

min :: MarkupWarning -> MarkupWarning -> MarkupWarning

type Rep MarkupWarning

Defined in MarkupParse

type Rep MarkupWarning = D1 ('MetaData "MarkupWarning" "MarkupParse" "markup-parse-0.0.0.2-HnQ2lGo9uJcI0zQ43aTniK" 'False) (((C1 ('MetaCons "BadEmptyElemTag" 'PrefixI 'False) (U1 :: Type -> Type) :+: C1 ('MetaCons "SelfCloserWithChildren" 'PrefixI 'False) (U1 :: Type -> Type)) :+: (C1 ('MetaCons "LeafWithChildren" 'PrefixI 'False) (U1 :: Type -> Type) :+: C1 ('MetaCons "TagMismatch" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedLazy) (Rec0 TagName) :*: S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedLazy) (Rec0 TagName)))) :+: ((C1 ('MetaCons "UnmatchedEndTag" 'PrefixI 'False) (U1 :: Type -> Type) :+: C1 ('MetaCons "UnclosedTag" 'PrefixI 'False) (U1 :: Type -> Type)) :+: (C1 ('MetaCons "EndTagInTree" 'PrefixI 'False) (U1 :: Type -> Type) :+: (C1 ('MetaCons "EmptyContent" 'PrefixI 'False) (U1 :: Type -> Type) :+: C1 ('MetaCons "MarkupParser" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'NoSourceStrictness 'DecidedLazy) (Rec0 ParserWarning))))))

type Result a = These [MarkupWarning] a

The structure of many returning functions.

A common computation pipeline is to take advantage of the These Monad instance eg

markup s bs = bs & (tokenize s >=> gather s) & second (Markup s)

resultError :: Result a -> a

Convert any warnings to an error

>>> resultError $ (tokenize Html) "<foo"
*** Exception: MarkupParser (ParserLeftover "<foo")
...

resultEither :: Result a -> Either [MarkupWarning] a

Returns Left on any warnings

>>> resultEither $ (tokenize Html) "<foo><baz"
Left [MarkupParser (ParserLeftover "<baz")]

resultMaybe :: Result a -> Maybe a

Returns results if any, ignoring warnings.

>>> resultMaybe $ (tokenize Html) "<foo><baz"
Just [StartTag "foo" []]

Token components

type TagName = ByteString

Name of token

name :: Standard -> Parser e ByteString

Parse a tag name. Each standard is slightly different.

selfClosers :: [TagName]

Html tags that self-close

type AttrName = ByteString

Name of an attribute.

type AttrValue = ByteString

Value of an attribute. "" is equivalent to true with respect to boolean attributes.

data Attr

An attribute of a tag

In parsing, boolean attributes, which are not required to have a value in HTML, will be set a value of "", which is ok. But this will then be rendered.

>>> detokenize Html <$> tokenize_ Html [i|<input checked>|]
["<input checked=\"\">"]

Constructors

Attr !AttrName !AttrValue

Instances

Generic Attr

Defined in MarkupParse

Associated Types

type Rep Attr :: Type -> Type

Methods

from :: Attr -> Rep Attr x

to :: Rep Attr x -> Attr

Show Attr

Defined in MarkupParse

Methods

showsPrec :: Int -> Attr -> ShowS

show :: Attr -> String

showList :: [Attr] -> ShowS

NFData Attr

Defined in MarkupParse

Methods

rnf :: Attr -> ()

Eq Attr

Defined in MarkupParse

Methods

(==) :: Attr -> Attr -> Bool

(/=) :: Attr -> Attr -> Bool

Ord Attr

Defined in MarkupParse

Methods

compare :: Attr -> Attr -> Ordering

(<) :: Attr -> Attr -> Bool

(<=) :: Attr -> Attr -> Bool

(>) :: Attr -> Attr -> Bool

(>=) :: Attr -> Attr -> Bool

max :: Attr -> Attr -> Attr

min :: Attr -> Attr -> Attr

ToExpr Attr

Defined in MarkupParse

Methods

toExpr :: Attr -> Expr

listToExpr :: [Attr] -> Expr

type Rep Attr

Defined in MarkupParse

type Rep Attr = D1 ('MetaData "Attr" "MarkupParse" "markup-parse-0.0.0.2-HnQ2lGo9uJcI0zQ43aTniK" 'False) (C1 ('MetaCons "Attr" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 AttrName) :*: S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 AttrValue)))

attrs :: Standard -> Parser a [Attr]

Parse attributions

Tokens

data Token

A Markup token

>>> runParser_ (many (token Html)) [i|<foo>content</foo>|]
[StartTag "foo" [],Content "content",EndTag "foo"]

>>> runParser_ (token Xml) [i|<foo/>|]
EmptyElemTag "foo" []

>>> runParser_ (token Html) "<!-- Comment -->"
Comment " Comment "

>>> runParser_ (token Xml) [i|<?xml version="1.0" encoding="UTF-8"?>|]
Decl "xml version=\"1.0\" encoding=\"UTF-8\""

>>> runParser_ (token Html) "<!DOCTYPE html>"
Doctype "DOCTYPE html"

>>> runParser_ (token Xml) "<!DOCTYPE foo [ declarations ]>"
Doctype "DOCTYPE foo [ declarations ]"

>>> runParser (token Html) [i|<foo a="a" b="b" c=c check>|]
OK (StartTag "foo" [Attr "a" "a",Attr "b" "b",Attr "c" "c",Attr "check" ""]) ""

>>> runParser (token Xml) [i|<foo a="a" b="b" c=c check>|]
Fail

Constructors

StartTag !TagName ![Attr]

A start tag. https://developer.mozilla.org/en-US/docs/Glossary/Tag

EmptyElemTag !TagName ![Attr]

An empty element tag. Optional for XML and kind of not allowed in HTML.

EndTag !TagName

A closing tag.

Content !ByteString

The content between tags.

Comment !ByteString

Contents of a comment.

Decl !ByteString

Contents of a declaration

Doctype !ByteString

Contents of a doctype declaration.

Instances

Generic Token

Defined in MarkupParse

Associated Types

type Rep Token :: Type -> Type

Methods

from :: Token -> Rep Token x

to :: Rep Token x -> Token

Show Token

Defined in MarkupParse

Methods

showsPrec :: Int -> Token -> ShowS

show :: Token -> String

showList :: [Token] -> ShowS

NFData Token

Defined in MarkupParse

Methods

rnf :: Token -> ()

Eq Token

Defined in MarkupParse

Methods

(==) :: Token -> Token -> Bool

(/=) :: Token -> Token -> Bool

Ord Token

Defined in MarkupParse

Methods

compare :: Token -> Token -> Ordering

(<) :: Token -> Token -> Bool

(<=) :: Token -> Token -> Bool

(>) :: Token -> Token -> Bool

(>=) :: Token -> Token -> Bool

max :: Token -> Token -> Token

min :: Token -> Token -> Token

ToExpr Token

Defined in MarkupParse

Methods

toExpr :: Token -> Expr

listToExpr :: [Token] -> Expr

type Rep Token

Defined in MarkupParse

type Rep Token = D1 ('MetaData "Token" "MarkupParse" "markup-parse-0.0.0.2-HnQ2lGo9uJcI0zQ43aTniK" 'False) ((C1 ('MetaCons "StartTag" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 TagName) :*: S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 [Attr])) :+: (C1 ('MetaCons "EmptyElemTag" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 TagName) :*: S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 [Attr])) :+: C1 ('MetaCons "EndTag" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 TagName)))) :+: ((C1 ('MetaCons "Content" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 ByteString)) :+: C1 ('MetaCons "Comment" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 ByteString))) :+: (C1 ('MetaCons "Decl" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 ByteString)) :+: C1 ('MetaCons "Doctype" 'PrefixI 'False) (S1 ('MetaSel ('Nothing :: Maybe Symbol) 'NoSourceUnpackedness 'SourceStrict 'DecidedStrict) (Rec0 ByteString)))))

tokenize :: Standard -> ByteString -> These [MarkupWarning] [Token]

Parse a bytestring into tokens

>>> tokenize Html [i|<foo>content</foo>|]
That [StartTag "foo" [],Content "content",EndTag "foo"]

tokenize_ :: Standard -> ByteString -> [Token]

tokenize but errors on warnings.

token :: Standard -> Parser String Token

A flatparse Token parser.

>>> runParser (token Html) "<foo>content</foo>"
OK (StartTag "foo" []) "content</foo>"

detokenize :: Standard -> Token -> ByteString

bytestring representation of Token.

>>> detokenize Html (StartTag "foo" [])
"<foo>"

gather :: Standard -> [Token] -> These [MarkupWarning] [Tree Token]

Gather together token trees from a token list, placing child elements in nodes and removing EndTags.

>>> gather Html =<< tokenize Html "<foo class=\"bar\">baz</foo>"
That [Node {rootLabel = StartTag "foo" [Attr "class" "bar"], subForest = [Node {rootLabel = Content "baz", subForest = []}]}]

gather_ :: Standard -> [Token] -> [Tree Token]

gather but errors on warnings.

degather :: Markup -> These [MarkupWarning] [Token]

Convert a markup into a token list, adding end tags.

>>> degather =<< markup Html "<foo class=\"bar\">baz</foo>"
That [StartTag "foo" [Attr "class" "bar"],Content "baz",EndTag "foo"]

degather_ :: Markup -> [Token]

degather but errors on warning

XML specific Parsers

xmlVersionInfo :: Parser e ByteString

xml production [24]

xmlEncodingDecl :: Parser e ByteString

xml production [80]

xmlStandalone :: Parser e ByteString

Xml production [32]

xmlVersionNum :: Parser e ByteString

xml production [26]

xmlEncName :: Parser e ByteString

xml production [81]

xmlYesNo :: Parser e ByteString

Xml yes/no