numeric

This is a collection of parsers concerned with handling signed real numbers (like floats and doubles).

These literals consist of a (possibly optional) integer prefix, with at least one of a fractional component (with .) or an exponential component.

Real numbers are an extension of signed integers with the following additional configuration:

• desc.numericDesc.leadingDotAllowed: determines whether a literal like .0 would be considered legal

• desc.numericDesc.trailingDotAllowed: determines whether a literal like 0. would be considered legal

• desc.numericDesc.realNumbersCanBe{Hexadecimal/Octal/Binary}: these flags control what kind of literals can appear within the number parser. Each type of literal may still be individually parsed with its corresponding parser, regardless of the value of the flag

• desc.numericDesc.{decimal/hexadecimal/octal/binary}ExponentDesc: describes how the exponential syntax works for each kind of base. If the syntax is legal, then this describes: which characters start it (classically, this would be e or E for decimals); whether or not it is compulsory for the literal (in Java and C, hexadecimal floats are only valid when they have an exponent attached); and whether or not a + sign is mandatory, optional, or illegal for positive exponents

Additional to the parsing of decimal, hexadecimal, octal, and binary floating literals, each parser can be given a precision of IEEE 754 float or double. This can either be achieved by rounding to the nearest representable value, or by ensuring that the literal must be precisely representable as one of these numbers (which is defined as being one of binary, decimal or exact float and double values as described by Java)

This is a collection of parsers concerned with handling signed integer literals.

Signed integer literals are an extension of unsigned integer literals with the following extra configuration:

• desc.numericDesc.positiveSign: describes whether or not literals are allowed to omit + for positive literals, must write a +, or can never write a +.

This is a collection of parsers concerned with handling unsigned (positive) integer literals.

Natural numbers are described generally as follows:

• desc.numericDesc.literalBreakChar: determines whether or not it is legal to "break up" the digits within a literal, for example: is 1_000_000 allowed? If this is legal, describes what the break character is, and whether it can appear after a hexadecimal/octal/binary prefix

• desc.numericDesc.leadingZerosAllowed: determines whether or not it is possible to add extraneous zero digits onto the front of a number or not. In some languages, like C, this is disallowed, as numbers starting with 0 are octal numbers.

• desc.numericDesc.integerNumbersCanBe{Hexadecimal/Octal/Binary}: these flags control what kind of literals can appear within the number parser. Each type of literal can be individually parsed with its corresponding parser, regardless of the value of the flag

• desc.numericDesc.{hexadecimal/octal/binary}Leads: controls what character must follow a 0 when starting a number to change it from decimal into another base. This set may be empty, in which case the literal is described purely with leading zero (C style octals would set octalLeads to Set.empty)

Additional to the parsing of decimal, hexadecimal, octal, and binary literals, each parser can be given a bit-width from 8- to 64-bit: this will check the parsed literal to ensure it is a legal literal of that size.

This is a collection of parsers concerned with handling signed real numbers (like floats and doubles).

These literals consist of a (possibly optional) integer prefix, with at least one of a fractional component (with .) or an exponential component.

Real numbers are an extension of signed integers with the following additional configuration:

• desc.numericDesc.leadingDotAllowed: determines whether a literal like .0 would be considered legal

• desc.numericDesc.trailingDotAllowed: determines whether a literal like 0. would be considered legal

• desc.numericDesc.realNumbersCanBe{Hexadecimal/Octal/Binary}: these flags control what kind of literals can appear within the number parser. Each type of literal may still be individually parsed with its corresponding parser, regardless of the value of the flag

• desc.numericDesc.{decimal/hexadecimal/octal/binary}ExponentDesc: describes how the exponential syntax works for each kind of base. If the syntax is legal, then this describes: which characters start it (classically, this would be e or E for decimals); whether or not it is compulsory for the literal (in Java and C, hexadecimal floats are only valid when they have an exponent attached); and whether or not a + sign is mandatory, optional, or illegal for positive exponents

Additional to the parsing of decimal, hexadecimal, octal, and binary floating literals, each parser can be given a precision of IEEE 754 float or double. This can either be achieved by rounding to the nearest representable value, or by ensuring that the literal must be precisely representable as one of these numbers (which is defined as being one of binary, decimal or exact float and double values as described by Java)

This is a collection of parsers concerned with handling signed integer literals.

Signed integer literals are an extension of unsigned integer literals with the following extra configuration:

• desc.numericDesc.positiveSign: describes whether or not literals are allowed to omit + for positive literals, must write a +, or can never write a +.

This is a collection of parsers concerned with handling numeric literals that may either be signed integers or signed reals.

There is no additional configuration offered over that found in integer or real.

the bit-bounds and precision of the integer or real parts of the result can be specified in any pairing.

This is a collection of parsers concerned with handling unsigned (positive) integer literals.

Natural numbers are described generally as follows:

• desc.numericDesc.literalBreakChar: determines whether or not it is legal to "break up" the digits within a literal, for example: is 1_000_000 allowed? If this is legal, describes what the break character is, and whether it can appear after a hexadecimal/octal/binary prefix

• desc.numericDesc.leadingZerosAllowed: determines whether or not it is possible to add extraneous zero digits onto the front of a number or not. In some languages, like C, this is disallowed, as numbers starting with 0 are octal numbers.

• desc.numericDesc.integerNumbersCanBe{Hexadecimal/Octal/Binary}: these flags control what kind of literals can appear within the number parser. Each type of literal can be individually parsed with its corresponding parser, regardless of the value of the flag

• desc.numericDesc.{hexadecimal/octal/binary}Leads: controls what character must follow a 0 when starting a number to change it from decimal into another base. This set may be empty, in which case the literal is described purely with leading zero (C style octals would set octalLeads to Set.empty)

Additional to the parsing of decimal, hexadecimal, octal, and binary literals, each parser can be given a bit-width from 8- to 64-bit: this will check the parsed literal to ensure it is a legal literal of that size.

This is a collection of parsers concerned with handling numeric literals that may either be unsigned integers or unsigned reals.

There is no additional configuration offered over that found in natural or real.

the bit-bounds and precision of the integer or real parts of the result can be specified in any pairing.