File: coreutils.info,  Node: Version-sort ordering rules,  Next: Version sort is not the same as numeric sort,  Up: Version sort implementation

30.2.1 Version-sort ordering rules
----------------------------------

The version sort ordering rules are:

  1. The strings are compared from left to right.

  2. First the initial part of each string consisting entirely of
     non-digit bytes is determined.

       A. These two parts (either of which may be empty) are compared
          lexically.  If a difference is found it is returned.

       B. The lexical comparison is a lexicographic comparison of byte
          strings, except that:

            a. ASCII letters sort before other bytes.
            b. A tilde sorts before anything, even an empty string.

  3. Then the initial part of the remainder of each string that contains
     all the leading digits is determined.  The numerical values
     represented by these two parts are compared, and any difference
     found is returned as the result of the comparison.

       A. For these purposes an empty string (which can only occur at
          the end of one or both version strings being compared) counts
          as zero.

       B. Because the numerical value is used, non-identical strings can
          compare equal.  For example, ‘123’ compares equal to ‘00123’,
          and the empty string compares equal to ‘0’.

  4. These two steps (comparing and removing initial non-digit strings
     and initial digit strings) are repeated until a difference is found
     or both strings are exhausted.

   Consider the version-sort comparison of two file names: ‘foo07.7z’
and ‘foo7a.7z’.  The two strings will be broken down to the following
parts, and the parts compared respectively from each string:

     foo  vs  foo   (rule 2, non-digits)
     07   vs  7     (rule 3, digits)
     .    vs  a.    (rule 2)
     7    vs  7     (rule 3)
     z    vs  z     (rule 2)

   Comparison flow based on above algorithm:

  1. The first parts (‘foo’) are identical.

  2. The second parts (‘07’ and ‘7’) are compared numerically, and
     compare equal.

  3. The third parts (‘.’ vs ‘a.’) are compared lexically by ASCII value
     (rule 2.B).

  4. The first byte of the first string (‘.’) is compared to the first
     byte of the second string (‘a’).

  5. Rule 2.B.a says letters sorts before non-letters.  Hence, ‘a’ comes
     before ‘.’.

  6. The returned result is that ‘foo7a.7z’ comes before ‘foo07.7z’.

   Result when using sort:

     $ cat input3
     foo07.7z
     foo7a.7z
     $ sort -V input3
     foo7a.7z
     foo07.7z

   See *note Differences from Debian version sort:: for additional rules
that extend the Debian algorithm in Coreutils.