For certain operations and transformations (especially in Unicode Sets) there is an expectation that certain derived general categories exists even though they are not defined in the unicode character database.

These categories are:

• :any which is the full unicode character range 0x0..0x10ffff

• :assigned which is the set of codepoints that are assigned and is therefore equivalent to [:any]-[:Cn]. In fact that is exactly how it is calculated using unicode_set and the results are copied here so that there is no mutual dependency.

• :ascii which is the range for the US ASCII character set of 0x0..0x7f

In addition there are derived categories not part of the Unicode specification that support additional use cases. These include:

• Categories related to recognising quotation marks. See the module Unicode.Category.QuoteMarks.

• :printable which implements the same semantics as String.printable?/1. This is a very broad definition of printable characters.

• :graph which includes characters from the [^\p{space}\p{gc=Control}\p{gc=Surrogate}\p{gc=Unassigned}] set defined by Unicode Regular Expressions.