Character sets (GNU Coreutils 9.0)

9.1.1 Specifying sets of characters

The format of the set1 and set2 arguments resembles the format of regular expressions; however, they are not regular expressions, only lists of characters. Most characters simply represent themselves in these strings, but the strings can contain the shorthands listed below, for convenience. Some of them can be used only in set1 or set2, as noted below.

Backslash escapes

The following backslash escape sequences are recognized:

‘\a’: Control-G.
‘\b’: Control-H.
‘\f’: Control-L.
‘\n’: Control-J.
‘\r’: Control-M.
‘\t’: Control-I.
‘\v’: Control-K.
‘\ooo’: The 8-bit character with the value given by ooo, which is 1 to 3 octal digits. Note that ‘\400’ is interpreted as the two-byte sequence, ‘\040’ ‘0’.
‘\\’: A backslash.

While a backslash followed by a character not listed above is interpreted as that character, the backslash also effectively removes any special significance, so it is useful to escape ‘[’, ‘]’, ‘*’, and ‘-’.

Ranges

The notation ‘m-n’ expands to all of the characters from m through n, in ascending order. m should collate before n; if it doesn’t, an error results. As an example, ‘0-9’ is the same as ‘0123456789’.

GNU tr does not support the System V syntax that uses square brackets to enclose ranges. Translations specified in that format sometimes work as expected, since the brackets are often transliterated to themselves. However, they should be avoided because they sometimes behave unexpectedly. For example, ‘tr -d '[0-9]'’ deletes brackets as well as digits.

Many historically common and even accepted uses of ranges are not portable. For example, on EBCDIC hosts using the ‘A-Z’ range will not do what most would expect because ‘A’ through ‘Z’ are not contiguous as they are in ASCII. If you can rely on a POSIX compliant version of tr, then the best way to work around this is to use character classes (see below). Otherwise, it is most portable (and most ugly) to enumerate the members of the ranges.

Repeated characters

The notation ‘[c*n]’ in set2 expands to n copies of character c. Thus, ‘[y*6]’ is the same as ‘yyyyyy’. The notation ‘[c*]’ in string2 expands to as many copies of c as are needed to make set2 as long as set1. If n begins with ‘0’, it is interpreted in octal, otherwise in decimal.

Character classes

The notation ‘[:class:]’ expands to all of the characters in the (predefined) class class. The characters expand in no particular order, except for the upper and lower classes, which expand in ascending order. When the --delete (-d) and --squeeze-repeats (-s) options are both given, any character class can be used in set2. Otherwise, only the character classes lower and upper are accepted in set2, and then only if the corresponding character class (upper and lower, respectively) is specified in the same relative position in set1. Doing this specifies case conversion. The class names are given below; an error results when an invalid class name is given.

alnum

Letters and digits.

alpha

Letters.

blank

Horizontal whitespace.

cntrl

Control characters.

digit

Digits.

graph

Printable characters, not including space.

lower

Lowercase letters.

print

Printable characters, including space.

punct

Punctuation characters.

space

Horizontal or vertical whitespace.

upper

Uppercase letters.

xdigit

Hexadecimal digits.

Equivalence classes

The syntax ‘[=c=]’ expands to all of the characters that are equivalent to c, in no particular order. Equivalence classes are a relatively recent invention intended to support non-English alphabets. But there seems to be no standard way to define them or determine their contents. Therefore, they are not fully implemented in GNU tr; each character’s equivalence class consists only of that character, which is of no particular use.