Characters

Creates and returns a new parser that does not produce a string result, but instead only the number of characters successfully parsed as an Int.

Creates and returns a new parser that consumes at most the specified maximum number of characters. Always succeeds, unless a minimum number of matches is also specified.

Creates and returns a new parser that fails if it does not consume the specified minimum number of characters. It may still consume more characters in case of further matches.

Creates and returns a new parser that consumes exactly the specified number of characters. Fails if there are less matches, but succeeds in case there are more matches, simply ignoring them. Calling take 3 for example is equivalent to calling min 3 max 3.

Creates and returns a new parser that does not produce a result, but instead only consumes the number of characters successfully parsed.

Returns a parser that repeatedly applies this parser. It will always succeed, potentially with an empty list as the result.

Returns a parser that repeatedly applies this parser (at least once).

Applies the specified parser to the input left over by this parser, but only keeps the left result.

a <~ b only succeeds if both parsers succeed.

Attempts to parse the specified literal string from the input left over by this parser, but only keeps the left result.

a <~ b only succeeds if both parsers succeed.

Operator synonym for flatMap.

Returns a parser that optionally parses what this parser parses.

A synonym for map, allowing the grammar to be declared in a concise way.

Returns a parser that ignores the result of this parser (if it succeeds) and returns the specified result instead.

Subclasses may override this method to avoid any expensive result processing.

Returns a parser that applies a partial function to the result of this parser.

p.collect(f) succeeds if p succeeds and f is defined at the result of p, In that case it returns f applied to the result of p.

Provides a cursor over the input consumed by this parser while discarding the actual result. Use withCursor if you also need access to the result.

This is required for parsers that create AST nodes that need to be resolved in a rewrite step and need to report the source location in case of failure. It is also required when passing a result of a first-pass parser to a recursive parser to preserve line positions.

Returns a parser that applies a function to the result of this parser producing an Either where Left is interpreted as failure. It is an alternative to ^? for scenarios where the conditional check cannot be easily performed in a pattern match.

p.evalMap(f) succeeds if p succeeds and f returns a Right when applied to the result of p.

Builds a new parser by applying the specified function to the result of this parser and subsequently applying the parser returned by that function to the input left over by this parser.

Handle any error, potentially recovering from it, by mapping it to a new parser that will be applied at the same starting position than the failing parser.

This is similar to the orElse or | method, but allows the alternative parser to inspect the error of the preceding one.

Builds a new parser by applying the specified function to the result of this parser.

Applies the specified parser when this parser fails.

a orElse b succeeds if either of the parsers succeeds.

In case both parsers fail, the Failure instance will be from the parser with the most successfully read characters. In the case of multiple failures having the same number of characters, the one with the highest precedence (this parser) will be chosen.

Implementation note: The parameter is by-name to allow the definition of recursive parsers. In contrast to the former SDK parser combinators this is the only place where a parser with a by-name parameter is used whereas in all other places the additional cost is avoided.

Parses the specified string and returns the result.

Handle certain errors, potentially recovering from it, by mapping them to a new parser that will be applied at the same starting position than the failing parser.

Returns a parser that repeatedly applies this parser with the specified separator string between those invocations.

p.rep(sep).min(1) is equivalent to (p ~ (sep ~> p).rep).concat.

The returned parser offers an API to specify further constraints like min or max.

Returns a parser that repeatedly applies this parser with the specified separator parser between those invocations.

p.rep(sep).min(1) is equivalent to (p ~ (sep ~> p).rep).concat.

The returned parser offers an API to specify further constraints like min or max.

Returns a parser that repeatedly applies this parser. The returned parser offers an API to specify further constraints like min or max.

Returns a parser that repeatedly applies this parser until either this parser fails or the specified end condition is met. The end condition will be applied after each successful invocation of this parser.

The result of the returned parser is a tuple consisting of the list containing the result of the invocations of this parser plus the result of the end condition. The latter is returned as an Option as it might be empty when the parsing finished because of this parser failing.

Note that it is more convenient to include the end condition in the repeating parser itself and use the simpler rep method. This combinator is an alternative if you need to know the result of the end condition.

Returns a parser that invokes the specified function repeatedly, passing the result of this parser if it succeeds, to produce new parsers that get applied until one of them fails.

The result of the returned parser is a list containing the result of this parser (if it succeeds) plus the results of successful invocations of the parsers returned by the specified function.

Retrieves the part of the input consumed by this parser while discarding the result.

This is useful in scenarios where many string-based parsers are combined and produce a deeply nested result like String ~ Option[String] ~ List[String] where it would require some boilerplate to concatenate the results. Using the source method, the entire text consumed by this combination of parsers will be returned.

If you also need the position within the input or need to pass the result to a recursive parser manually, use the cursor method instead.

Provides the result of this parser together with a cursor over the input, capturing the consumed source string and its position within the root input. Use cursor if you do not need access to the actual result.

This is required for parsers that create AST nodes that need to be resolved in a rewrite step and need to report the source location in case of failure. It is also required when passing a result of a first-pass parser to a recursive parser to preserve line positions.

Changes the failure message produced by a parser.

Attempts to parse the specified literal string when this parser fails.

a | b succeeds if either of the parsers succeeds.

Applies the specified parser when this parser fails.

a | b succeeds if either of the parsers succeeds.

Implementation note: The parameter is by-name to allow the definition of recursive parsers. In contrast to the former SDK parser combinators this is the only place where a parser with a by-name parameter is used whereas in all other places the additional cost is avoided.

Applies the specified parser to the input left over by this parser and combines the two results.

a ~ b only succeeds if both parsers succeed, with the results in a wrapper class named ~ for convenient pattern matching:

  a ~ b ~ c ^^ {
    case a ~ b ~ c => processResult(a, b, c)
  }

Attempts to parse the specified literal string from the input left over by this parser and combines the two results.

a ~ b only succeeds if both parsers succeed, with the results in a wrapper class named ~ for convenient pattern matching:

  a ~ b ~ c ^^ {
    case a ~ b ~ c => processResult(a, b, c)
  }

Applies the specified parser to the input left over by this parser, but only keeps the right result.

a ~> b only succeeds if both parsers succeed.

Attempts to parse the specified literal string from the input left over by this parser, but only keeps the right result.

a ~> b only succeeds if both parsers succeed.