A Unicode Set is a representation of a set of Unicode characters or character strings. The contents of that set are specified by patterns or by building them programmatically. This library implements parsing of unicode sets, resolving them to a list of codepoints and matching a given codepoint to that list. This expansion supports the following public API:

• Unicode.Set.match?/2 which is a macro that matches a codepoint to a unicode set.
• Unicode.Regex.compile/2 which pre-processes a regex string expanding unicode sets into a regex executable by the Regex module.
• Unicode.Set.to_utf8_char/1 that converts a unicode set into a form usable with nimble_parsec
• Unicode.Set.compile_pattern/1 which converts a unicode set into a string that is then compiled with :binary.compile_pattern/1.

The implementation conforms closely to the Unicode Set specification but currently omits support for the \N{codepoint_name} syntax.

This is helpful in defining function guards. For example:

defmodule Guards do
  require Unicode.Set

  # Define a guard that checks if a codepoint is a unicode digit
  defguard digit?(x) when Unicode.Set.match?(x, "[[:Nd:]]")
end

defmodule MyModule do
  require Guards

  # Define a function using the previously defined guard
  def my_function(<< x :: utf8, _rest :: binary>>) when Guards.digit?(x) do
    IO.puts "Its a digit!"
  end

  # Define a guard directly on the function
  def my_other_function_(<< x :: utf8, _rest :: binary>>) when Unicode.Set.match?(x, "[[:Nd:]]") do
    IO.puts "Its also a digit!"
  end
end

String.split/3 and String.replace/3 allow for patterns and compiled patterns to be used with compiled patterns being the more performant approach. Unicode Set supports the generation of patterns and compiled patterns:

iex> pattern = Unicode.Set.compile_pattern!("[[:digit:]]")
iex> list = String.split("abc1def2ghi3jkl", pattern)
["abc", "def", "ghi", "jkl"]

The parser generator nimble_parsec allows a list of codepoint ranges as parameters to several combinators. Unicode Set can generate such ranges:

iex> Unicode.Set.to_utf8_char!("[[^abcd][mnb]]")
[98, 109..110, {:not, 97..100}]

This can be used as shown in the following example:

defmodule MyCombinators do
  import NimbleParsec

  @digit_list = Unicode.Set.to_utf8_char!("[[:digit:]]")
  def unicode_digit do
    utf8_char(@digit_list)
    |> label("a digit in any Unicode script")
  end
end

The Regex module supports a limited set of Unicode Sets. The Unicode.Regex module provides compile/2 and compile!/2 functions that have the same arguments and compatible functionality with Regexp.compile/2 other that they pre-process the regular expression, expanding any Unicode Sets. This makes it simple to incorporate Unicode Sets in regular expressions.

All Unicode Sets are expanded, even those that are known to Regex.compile/2 since the erlang :re module upon Regex is based does not always keep pace with Unicode releases.

For example:

iex> Unicode.Regex.compile("\\p{Zs}")
{:ok, ~r/[\x{20}\x{A0}\x{1680}\x{2000}-\x{200A}\x{202F}\x{205F}\x{3000}]/u}

iex> Unicode.Regex.compile("[:graphic:]")
{:ok,
 ~r/[\x{20}-\x{7E}\x{A0}-\x{AC}\x{AE}-\x{377}\x{37A}-\x{37F}...]/u}

These examples show how to combine sets (union, difference and intersection) to deliver a flexible targeting of the required match.

# The character "๓" is the thai digit `1`
iex> Unicode.Set.match? ?๓, "[[:digit:]]"
true

# Set operations allow union, insersection and difference
# This example matches on digits, but not the Thai script
iex> Unicode.Set.match? ?๓, "[[:digit:]-[:thai:]]"
false

As much work as possible is done at compile time in order to deliver good performance. The macro Unicode.Set.match?/2 parses the unicode set, expands the require codepoints and generates guard clauses at compile time. The resulting code is a simple set of boolean operators that executes quickly at runtime.

This version of Unicode Set supports the following enumerable unicode properties in unicode sets:

• script such as [:script=arabic:], \p{script=arabic} or [:arabic:]
• block such as [:block=sudanese:], \p{block=sudanese}, \p{IsSudanese} or [:IsSudanese:]
• general category such as [:Lu:], \p{Lu}, [:gc=Lu:] or [:general category=Lu:]
• combining class such as [:ccc=230:]

In addition, the following boolean properties are supported. These are expressed as [:white space:] or \p{White Space}.

In all cases, property names and property values may include whitespace and mixed case notation.

In addition to the Unicode properties, some additional properties are also defined for convenience. These properties related to quote marks and are:

• quote_mark
• quote_mark_left
• quote_mark_right
• quote_mark_ambidextrous
• quote_mark_single
• quote_mark_double

As above these properties can be expressed in mixed case with spaces and underscores inserted for readability. They can be used in the same way as any Unicode property name.

Here are a few examples of sets. Although elements of the syntax appear similar to regular expressions, unicode sets only expresses one or more ranges of unicode codepoints.

In addition to being a set of characters (of Unicode code points), a UnicodeSet may also contain string values. Conceptually, the UnicodeSet is always a set of strings, not a set of characters, although in many common use cases the strings are all of length one, which reduces to being a set of characters.

This concept can be confusing when first encountered, probably because similar set constructs from other environments (regular expressions) can only contain characters.

Patterns are a series of characters bounded by square brackets that contain lists of characters and Unicode property sets. Lists are a sequence of characters that may have ranges indicated by a '-' between two characters, as in "a-z". The sequence specifies the range of all characters from the left to the right, in Unicode order. For example, [a c d-f m] is equivalent to [a c d e f m]. Whitespace can be freely used for clarity as [a c d-f m] means the same as [acd-fm].

Unicode property sets are specified by a Unicode property, such as [:Letter:]. For a list of supported properties, see the Properties section. For details on the use of short vs. long property and property value names, see the end of this section. The syntax for specifying the property names is an extension of either POSIX or Perl syntax with the addition of =value. For example, you can match letters by using the POSIX syntax [:Letter:], or by using the Perl-style syntax \p{Letter}. The type can be omitted for the Category and Script properties, but is required for other properties.

The table below shows the two kinds of syntax: POSIX and Perl style. Also, the table shows the "Negative", which is a property that excludes all characters of a given kind. For example, [:^Letter:] matches all characters that are not [:Letter:].

These following low-level lists or properties then can be freely combined with the normal set operations (union, inverse, difference, and intersection):

The binary operators of union, intersection, and set-difference have equal precedence and bind left-to-right. Thus the following are equivalent:

• [[:letter:] - [a-z] [:number:] & [\u0100-\u01FF]]
• [[[[[:letter:] - [a-z]] [:number:]] & [\u0100-\u01FF]]

Another example is that the set [[ace][bdf] - [abc][def]] is not the empty set, but instead the set [def]. That is because the syntax corresponds to the following UnicodeSet operations:

1. start with [ace]
2. union [bdf] -- we now have [abcdef]
3. subtract [abc] -- we now have [def]
4. union [def] -- no effect, we still have [def]

This only really matters where there are the difference and intersection operations, as the union operation is commutative. To make sure that the - is the main operator, add brackets to group the operations as desired, such as [[ace][bdf] - [[abc][def]]].

Another caveat with the & and - operators is that they operate between sets. That is, they must be immediately preceded and immediately followed by a set. For example, the pattern [[:Lu:]-A] is illegal, since it is interpreted as the set [:Lu:] followed by the incomplete range -A. To specify the set of uppercase letters except for A, enclose the A in a set: [[:Lu:]-[A]].

• [a] The set containing 'a'
• [a-z] The set containing 'a' through 'z' and all letters in between, in Unicode order
• [^a-z] The set containing all characters but 'a' through 'z', that is, U+0000 through 'a'-1 and 'z'+1 through U+FFFF
• [[pat1][pat2]] The union of sets specified by pat1 and pat2
• [[pat1]&[pat2]] The intersection of sets specified by pat1 and pat2
• [[pat1]-[pat2]] The asymmetric difference of sets specified by pat1 and pat2
• [:Lu:] The set of characters belonging to the given Unicode category; in this case, Unicode uppercase letters. The long form for this is [:UppercaseLetter:].
• [:L:] The set of characters belonging to all Unicode categories starting with 'L', that is, [[:Lu:][:Ll:][:Lt:][:Lm:][:Lo:]]. The long form for this is [:Letter:].

String values are enclosed in {curly brackets}.

Two single quotes represents a single quote, either inside or outside single quotes.

Text within single quotes is not interpreted in any way (except for two adjacent single quotes). It is taken as literal text (special characters become non-special).

These quoting conventions for ICU UnicodeSets differ from those of regular expression character set expressions. In regular expressions, single quotes have no special meaning and are treated like any other literal character.

Outside of single quotes, certain backslashed characters have special meaning. Note that these are escapes processed by Unicode Set (this library) and therefore require \\\\ to be entered as a prefix. Elixir also provides similar escapes as native part of its string processing and Elixir's escapes are to be preferred where possible.

Certain other escapes are native to Elixir and are applicable in Unicode Sets they are in any Elixir string:

Anything else following a backslash is mapped to itself, except in an environment where it is defined to have some special meaning. For example, \p{Lu} is the set of uppercase letters in a Unicode Set.

Any character formed as the result of a backslash escape loses any special meaning and is treated as a literal. In particular, note that \u and \U escapes create literal characters.

Whitespace (as defined by the specification) is ignored unless it is quoted or backslashed.

The following property value variants are recognized:

If the type or value is omitted, then the equals sign is also omitted. The short style is only used for Category and Script properties because these properties are very common and their omission is unambiguous.

In actual practice, you can mix type names and values that are omitted, abbreviated, or full. For example, if Category=Unassigned you could use what is in the table explicitly, \p{gc=Unassigned}, \p{Category=Cn}, or \p{Unassigned}.

When these are processed, case and whitespace are ignored so you may use them for clarity, if desired. For example, \p{Category = Uppercase Letter} or \p{Category = uppercase letter}.

To install, add the package unicode_set to your list of dependencies in mix.exs:

def deps do
  [
    {:unicode_set, "~> 1.0"}
  ]
end

Documentation can be found at https://hexdocs.pm/unicode_set.