Streamly.Internal.Unicode.Parser

To parse a text input, use the decode routines from Streamly.Unicode.Stream module to convert an input byte stream to a Unicode Char stream and then use these parsers on the Char stream.

Setup

To execute the code examples provided in this module in ghci, please run the following commands first.

>>> :m
>>> import qualified Streamly.Data.Stream as Stream
>>> import qualified Streamly.Unicode.Parser as Unicode

For APIs that have not been released yet.

>>> import qualified Streamly.Internal.Unicode.Parser as Unicode (number, mkDouble)

Generic

char :: Monad m => Char -> Parser Char m Char

Match a specific character.

charIgnoreCase :: Monad m => Char -> Parser Char m Char

Match a specific character ignoring case.

Sequences

string :: Monad m => String -> Parser Char m String

Match the input with the supplied string and return it if successful.

stringIgnoreCase :: Monad m => String -> Parser Char m String

Match the input with the supplied string and return it if successful.

dropSpace :: Monad m => Parser Char m ()

Drop zero or more white space characters.

dropSpace1 :: Monad m => Parser Char m ()

Drop one or more white space characters.

Classes

alpha :: Monad m => Parser Char m Char

Match any character that satisfies isAlpha

alphaNum :: Monad m => Parser Char m Char

Match any character that satisfies isAlphaNum

letter :: Monad m => Parser Char m Char

Match any character that satisfies isLetter

ascii :: Monad m => Parser Char m Char

Match any character that satisfies isAscii

asciiLower :: Monad m => Parser Char m Char

Match any character that satisfies isAsciiLower

asciiUpper :: Monad m => Parser Char m Char

Match any character that satisfies isAsciiUpper

latin1 :: Monad m => Parser Char m Char

Match any character that satisfies isLatin1

lower :: Monad m => Parser Char m Char

Match any character that satisfies isLower

upper :: Monad m => Parser Char m Char

Match any character that satisfies isUpper

mark :: Monad m => Parser Char m Char

Match any character that satisfies isMark

printable :: Monad m => Parser Char m Char

Match any character that satisfies isPrint

punctuation :: Monad m => Parser Char m Char

Match any character that satisfies isPunctuation

separator :: Monad m => Parser Char m Char

Match any character that satisfies isSeparator

space :: Monad m => Parser Char m Char

Match any character that satisfies isSpace

symbol :: Monad m => Parser Char m Char

Match any character that satisfies isSymbol

digit :: Monad m => Parser Char m Char

Match any character that satisfies isDigit

octDigit :: Monad m => Parser Char m Char

Match any character that satisfies isOctDigit

hexDigit :: Monad m => Parser Char m Char

Match any character that satisfies isHexDigit

numeric :: Monad m => Parser Char m Char

Match any character that satisfies isNumber

Numeric

signed :: (Num a, Monad m) => Parser Char m a -> Parser Char m a

Allow an optional leading '+' or '-' sign character before any parser.

number :: Monad m => Parser Char m (Integer, Int)

A generic parser for scientific notation of numbers. Returns (mantissa, exponent) tuple. The result can be mapped to Double or any other number representation e.g. Scientific.

For example, using the scientific package: >> parserScientific = uncurry Data.Scientific.scientific $ number

doubleParser :: Monad m => Parser Char m (Int, Int)

A fast, custom parser for double precision flaoting point numbers. Returns (mantissa, exponent) tuple. This is much faster than number because it assumes the number will fit in a Double type and uses Int representation to store mantissa.

Number larger than Double may overflow. Int overflow is not checked in the exponent.

double :: Monad m => Parser Char m Double

Parse a decimal Double value. This parser accepts an optional sign (+ or -) followed by at least one decimal digit. Decimal digits are optionally followed by a decimal point and at least one decimal digit after the point. This parser accepts the maximal valid input as long as it gives a valid number. Specifcally a trailing decimal point is allowed but not consumed. This function does not accept "NaN" or "Infinity" string representations of double values.

Definition:

>>> double = uncurry Unicode.mkDouble <$> Unicode.number

Examples:

>>> p = Stream.parse Unicode.double . Stream.fromList

>>> p "-1.23e-123"
Right (-1.23e-123)

Trailing input examples:

>>> p "1."
Right 1.0

>>> p "1.2.3"
Right 1.2

>>> p "1e"
Right 1.0

>>> p "1e2.3"
Right 100.0

>>> p "1+2"
Right 1.0

Error cases:

>>> p ""
Left (ParseError "number: expecting sign or decimal digit, got end of input")

>>> p ".1"
Left (ParseError "number: expecting sign or decimal digit, got '.'")

>>> p "+"
Left (ParseError "number: expecting decimal digit, got end of input")

decimal :: (Monad m, Integral a) => Parser Char m a

Parse and decode an unsigned integral decimal number.

hexadecimal :: (Monad m, Integral a, Bits a) => Parser Char m a

Parse and decode an unsigned integral hexadecimal number. The hex digits 'a' through 'f' may be upper or lower case.

Note: This parser does not accept a leading "0x" string.

Utilities

mkDouble :: Integer -> Int -> Double

mkDouble mantissa exponent converts a mantissa and exponent to a Double value equivalent to mantissa * 10^exponent. It does not check for overflow, powers more than 308 will overflow.