The options are as follows:
To handle calendars in your national code table you can specify ``LANG=<locale_name>'' in the calendar file as early as possible. To handle national Easter names in the calendars, ``Easter=<national_name>'' (for Catholic Easter) or ``Paskha=<national_name>'' (for Orthodox Easter) can be used.
A special locale name exists: `utf-8' Specifying ``LANG=utf-8'' indicates that the dates will be read using the C locale, and the descriptions will be encoded in UTF-8. This is usually used for the distributed calendar files. The ``CALENDAR'' variable can be used to specify the style. Only `Julian' and `Gregorian' styles are currently supported. Use ``CALENDAR='' to return to the default (Gregorian).
To enforce special date calculation mode for Cyrillic calendars you should specify ``LANG=<local_name>'' and ``BODUN=<bodun_prefix>'' where <local_name> can be ru_RU.KOI8-R, uk_UA.KOI8-U or by_BY.KOI8-B.
Note that the locale is reset to the user's default for each new file that is read. This is so that locales from one file do not accidentally carry over into another file.
Other lines should begin with a month and day. They may be entered in almost any format, either numeric or as character strings. If proper locale is set, national months and weekdays names can be used. A single asterisk (`*') matches every month. A day without a month matches that day of every week. A month without a day matches the first of that month. Two numbers default to the month followed by the day. Lines with leading tabs default to the last entered date, allowing multiple line specifications for a single date. ``Easter'' (may be followed by a positive or negative integer) is Easter for this year. ``Paskha'' (may be followed by a positive or negative integer) is Orthodox Easter for this year. Weekdays may be followed by ``-4'' ... ``+5'' (aliases last, first, second, third, fourth) for moving events like ``the last Monday in April''
By convention, dates followed by an asterisk (`*' ) are not fixed, i.e., change from year to year.
Day descriptions start after the first <tab> character in the line; if the line does not contain a <tab> character, it isn't printed out. If the first character in the line is a <tab> character, it is treated as the continuation of the previous description.
The calendar file is preprocessed by cpp(1), allowing the inclusion of shared files such as company holidays or meetings. If the shared file is not referenced by a full pathname, cpp(1) searches in the current (or home) directory first, and then in the directory directory /etc/calendar and finally in /usr/share/calendar Empty lines and lines protected by the C commenting syntax (/* ... */ ) are ignored.
Some possible calendar entries (a \t sequence denotes a <tab> character):
LANG=C Easter=Ostern #include <calendar.usholiday> #include <calendar.birthday> 6/15\tJune 15 (if ambiguous, will default to month/day). Jun. 15\tJune 15. 15 June\tJune 15. Thursday\tEvery Thursday. June\tEvery June 1st. 15 *\t15th of every month. May Sun+2\tsecond Sunday in May (Muttertag) 04/SunLast\tlast Sunday in April, \tsummer time in Europe Easter\tEaster Ostern-2\tGood Friday (2 days before Easter) Paskha\tOrthodox Easter
The -t flag argument syntax is from the original FreeBSD calendar program.
The -l and -w flags are Debian-specific enhancements. Also, the original calendar program did not accept 0 as an argument to the -A flag.
Using `utf-8' as a locale name is a Debian-specific enhancement.