Agda-2.6.2.2: A dependently typed functional programming language and proof assistant
Safe HaskellNone
LanguageHaskell2010

Agda.Syntax.Concrete.Operators.Parser

Synopsis

Documentation

satNoPlaceholder :: (e -> Maybe a) -> Parser e a Source #

data ExprView e Source #

Constructors

LocalV QName 
WildV e 
OtherV e 
AppV e (NamedArg e) 
OpAppV QName (Set Name) (OpAppArgs' e)

The QName is possibly ambiguous, but it must correspond to one of the names in the set.

HiddenArgV (Named_ e) 
InstanceArgV (Named_ e) 
LamV (List1 LamBinding) e 
ParenV e 

Instances

Instances details
IsExpr e => HasRange (ExprView e) Source # 
Instance details

Defined in Agda.Syntax.Concrete.Operators.Parser

Methods

getRange :: ExprView e -> Range Source #

data ParseSections Source #

Should sections be parsed?

parse :: IsExpr e => (ParseSections, Parser e a) -> [e] -> [a] Source #

Runs a parser. If sections should be parsed, then identifiers with at least two name parts are split up into multiple tokens, using PositionInName to record the tokens' original positions within their respective identifiers.

Parser combinators

partP :: IsExpr e => [Name] -> RawName -> Parser e Range Source #

Parse a specific identifier as a NamePart

atLeastTwoParts :: IsExpr e => Parser e Name Source #

Parses a split-up, unqualified name consisting of at least two name parts.

The parser does not check that underscores and other name parts alternate. The range of the resulting name is the range of the first name part that is not an underscore.

patternBinder :: IsExpr e => Parser e Binder Source #

Parses a potentially pattern-matching binder

type family OperatorType (k :: NotationKind) (e :: Type) :: Type Source #

Used to define the return type of opP.

data NK (k :: NotationKind) :: Type where Source #

A singleton type for NotationKind (except for the constructor NoNotation).

opP :: forall e k. IsExpr e => ParseSections -> Parser e e -> NewNotation -> NK k -> Parser e (OperatorType k e) Source #

Parse the "operator part" of the given notation.

Normal holes (but not binders) at the beginning and end are ignored.

If the notation does not contain any binders, then a section notation is allowed.

argsP :: IsExpr e => Parser e e -> Parser e [NamedArg e] Source #

appP :: IsExpr e => Parser e e -> Parser e [NamedArg e] -> Parser e e Source #

atomP :: IsExpr e => (QName -> Bool) -> Parser e e Source #