You are looking at historical revision 22599 of this page. It may differ significantly from its current revision.
Time Data Types and Procedures
- Core Procedures
- SRFI 19 Document Changes
- SRFI 18 Time
- Time Conversion
- Time Arithmetic
- Time Comparison
- Date Arithmetic
- Date Comparison
- Input/Output Procedures
- Time Period
- Core Procedures
- Bugs and Limitations
- Version history
This is a Chicken port of SRFI 19. This document only describes the extensions. For the SRFI 19 API see SRFI 19.
The core procedures are those pertaining to time, date, and timezone:
The core procedures can be separately accessed:
SRFI 19 Document Changes
The nanosecond time object element is an integer between 0 and 999,999,999 inclusive. (The SRFI 19 document mis-states the value.)
A tz-offset value follows ISO 8601; positive for east of UTC, and negative for west. This is the opposite of the POSIX TZ environment variable.
Where the SRFI 19 document states a tz-offset argument a timezone-components object is also legal.
The string->date procedure allows the template-name argument to be optional. When missing the locale's date-time-format string is used. The supplied locale bundle's strings are invertible.
make-date[procedure] (make-date NANOSECOND SECOND MINUTE HOUR DAY MONTH YEAR [ZONE-OFFSET [TZ-NAME [DST-FLAG]]]) => date
Same as SRFI 19 except for the optional parameters and allowing a timezone-components object for the ZONE-OFFSET.
The ZONE-OFFSET is an integer or timezone-components. Default is the (timezone-locale-offset), the current locale timezone offset.
The TZ-NAME is a string or #f, and is the timezone name. Default is #f.
The DST-FLAG is a boolean, and indicates whether Day Saving TIme (or Summer Time) is active. Default is #f.
When the ZONE-OFFSET is a timezone-components object the TZ-NAME and DST-FLAG are pulled from the timezone-components, unless explicitly supplied.
read-leap-second-table[procedure] (read-leap-second-table FILENAME)
Sets the leap second table from the specified FILENAME.
The file format is the same as the "tai-utc.dat" file in the distribution. Provided by the U.S. Naval Observatory.
leap-year?[procedure] (leap-year? DATE) => boolean
Does the specified DATE fall on a leap year?
The DATE may be a numeric year or a date object.
SRFI 18 Time
Note that the SRFI 18 identifiers time?, current-time, seconds->time, time->seconds, milliseconds->time, and time->milliseconds are in conflict with those of SRFI 19.
time->srfi-18-time[procedure] (time->srfi-18-time TIME) => srfi-18#time
Converts a SRFI 19 time object to a SRFI 18 time object. The conversion is really only meaningful for time-duration, but any time-type is accepted.
srfi-18-time->time[procedure] (srfi-18-time->time => srfi-18#time) => srfi-19#time
Converts a SRFI 18 time object into a SRFI 19 time-duration object.
seconds->time[procedure] (seconds->time SECONDS [TIME-TYPE time-duration]) => time
Converts a SECONDS value, may be fractional, into a TIME-TYPE time object.
seconds->date[procedure] (seconds->date SECONDS [TIMEZONE-INFO #f]) => date
Converts a SECONDS value, which may be fractional, into a date object. The TIMEZONE-INFO is #t for the local timezone, #f for the utc timezone, or a timezone-components object.
SECONDS is relative to 00:00:00 January 1, 1970 UTC.
time->nanoseconds[procedure] (time->nanoseconds TIME) => integer
Returns the TIME object value as a nanoseconds value.
nanoseconds->time[procedure] (nanoseconds->time NANOSECONDS [TIME-TYPE time-duration]) => time
Returns the NANOSECONDS value as a time TIME-TYPE object.
nanoseconds->seconds[procedure] (nanoseconds->seconds NANOSECONDS) => integer
Returns the NANOSECONDS value as an inexact seconds value.
time->milliseconds[procedure] (time->milliseconds TIME) => integer
Returns the TIME object value as a milliseconds value.
milliseconds->time[procedure] (milliseconds->time MILLISECONDS [TIME-TYPE time-duration]) => time
Returns the MILLISECONDS value as a time TIME-TYPE object.
milliseconds->seconds[procedure] (milliseconds->seconds MILLISECONDS) => integer
Returns the MILLISECONDS value as an inexact seconds value.
time->date[procedure] (time->date TIME) => date
Returns the TIME object value as a date. A shorthand for the (time-*->date...) procedures.
time->julian-day[procedure] (time->julian-day TIME) => rational
Returns the julian day for the TIME object.
time->modified-julian-day[procedure] (time->modified-julian-day TIME) => rational
Returns the modified julian day for the TIME object.
make-duration[procedure] (make-duration [#:days 0] [#:hours 0] [#:minutes 0] [#:seconds 0] [#:milliseconds 0] [#:microseconds 0] [#:nanoseconds 0]) => time
Returns a time-object of clock-type time-duration where the seconds and nanoseconds values are calculated by summing the keyword arguments.
ONE-SECOND-DURATION and ONE-NANOSECOND-DURATION are pre-defined.
divide-duration[procedure] (divide-duration DURATION NUMBER) => time
Returns a duration, from DURATION, divided by NUMBER, without remainder.
divide-duration![procedure] (divide-duration! DURATION NUMBER) => time
Returns DURATION, divided by NUMBER, without remainder.
multiply-duration[procedure] (multiply-duration DURATION NUMBER) => time
Returns a duration, from DURATION, multiplied by NUMBER, truncated.
multiply-duration![procedure] (multiply-duration! DURATION NUMBER) => time
Returns DURATION, multiplied by NUMBER, truncated.
time-negative?[procedure] (time-negative? TIME) => boolean
Is TIME negative?
A time object will never have a negative nanoseconds value.
time-positve?[procedure] (time-positve? TIME) => boolean
Is TIME positive?
time-zero?[procedure] (time-zero? TIME) => boolean
Is TIME zero?
time-abs[procedure] (time-abs TIME) => time
Returns the absolute time value, from TIME.
time-abs![procedure] (time-abs! TIME) => time
Returns the absolute TIME value.
time-negate[procedure] (time-negate TIME) => time
Returns the sign inverted time value, from TIME.
time-negate![procedure] (time-negate! TIME) => time
Returns the TIME sign inverted value.
time-compare[procedure] (time-compare TIME1 TIME2) => integer
Returns -1, 0, or 1.
time-max[procedure] (time-max TIME1 [TIME2...]) => time
Returns the maximum time object from TIME1 TIME2....
time-min[procedure] (time-min TIME1 [TIME2...]) => time
Returns the minimum time object from TIME1 TIME2....
default-date-clock-type[parameter] (default-date-clock-type [CLOCK-TYPE time-utc]) => symbol
Sets or gets the clock-type used by default for conversion of a date to a time.
copy-date[procedure] (copy-date DATE) => date
Returns an exact copy of the specified DATE object.
date->time[procedure] (date->time DATE [CLOCK-TYPE (default-date-clock-type)]) => time
Returns the specified DATE as a time-object of type CLOCK-TYPE.
date-zone-name[procedure] (date-zone-name DATE) => (union boolean string)
Returns the timezone abbreviation of the specified DATE object. The result is either a string or #f.
date-dst?[procedure] (date-dst? DATE) => boolean
Returns the daylight saving time flag of the specified DATE object.
Only valid for "current" dates. Historical dates will not have a correct setting. Future dates cannot have a correct setting.
date-difference[procedure] (date-difference DATE1 DATE2 [CLOCK-TYPE]) => time
Returns the duration between DATE1 and DATE2.
date-add-duration[procedure] (date-add-duration DATE DURATION [CLOCK-TYPE]) => time
Returns the DATE plus the DURATION.
date-subtract-duration[procedure] (date-subtract-duration DATE DURATION [CLOCK-TYPE]) => time
Returns the DATE minus the DURATION.
date-compare[procedure] (date-compare DATE1 DATE2) => integer
Returns -1, 0, or 1.
date=?[procedure] (date=? DATE1 DATE2) => boolean
Is DATE1 on DATE2?
date>?[procedure] (date>? DATE1 DATE2) => boolean
Is DATE1 after DATE2?
date<?[procedure] (date<? DATE1 DATE2) => boolean
Is DATE1 before DATE2?
date>=?[procedure] (date>=? DATE1 DATE2) => boolean
Is DATE1 after or on DATE2?
date<=?[procedure] (date<=? DATE1 DATE2) => boolean
Is DATE1 before or on DATE2?
- Note that the daylight saving time (summer time) flag is always taken from the system, unless supplied. Any summer time rule component of a timezone-components object is not processed.
Remember that SRFI 19 timezone offset follows ISO 8601.
local-timezone-locale[parameter] (local-timezone-locale [TZ-COMPONENTS])
Gets or sets the local timezone-locale object.
utc-timezone-locale[parameter] (utc-timezone-locale [TZ-COMPONENTS])
Gets or sets the utc timezone-locale object.
Probably not a good idea to change the value.
timezone-locale-name[procedure] (timezone-locale-name [TZ-COMPONENTS]) => symbol
Returns the timezone-locale name of the supplied TZ-COMPONENTS, or the (local-timezone-locale) if missing.
timezone-locale-offset[procedure] (timezone-locale-offset [TZ-COMPONENTS]) => integer
Returns the timezone-locale offset of the supplied TZ-COMPONENTS, or the (local-timezone-locale) if missing.
timezone-locale-dst?[procedure] (timezone-locale-dst? [TZ-COMPONENTS]) => boolean
Returns the timezone-locale daylight saving time flag of the supplied TZ-COMPONENTS, or the (local-timezone-locale) if missing.
DATE->STRING conversion specifiers
The SRFI 19 document does not mention the padding character override feature for the normally zero-padded conversions f, H, I, j, m, M, N, S, y. If the the tilde is followed by a - then padding is suppressed. If followed by a _ the space character is used for padding. If followed by the conversion character then zero-padding is perfomed, the defualt.
format-date[procedure] (format-date DESTINATION DATE-FORMAT-STRING DATE)
Displays a text form of the DATE on the DESTINATION using the DATE-FORMAT-STRING.
When the destination is #t the (current-output-port) is used, and the date object must be specified.
When the destination is a port it must be an output-port, and the date object must be specified. When the destination is a number the (current-error-port) is the destination, and the DATE object must be specified.
When the destination is #f the result is returned as a string, and the DATE object must be specified.[procedure] (format-date DATE-FORMAT-STRING DATE)
Result is returned as a string.
scan-date[procedure] (scan-date SOURCE TEMPLATE-STRING)
Reads a text form of a date from the SOURCE, following the TEMPLATE-STRING, and returns a date object.
When the source is #t the (current-input-port) is used.
When the source is a port it must be an input-port.
When the source is string it should be a date text form.
A time-period is an interval, [begin end), where begin and end are time objects of the same clock type. When end <= begin the interval is null.
make-null-time-period[procedure] (make-null-time-period [CLOCK-TYPE (default-date-clock-type)]) => time-period
Returns a null interval for the specified CLOCK-TYPE.
make-time-period[procedure] (make-time-period BEGIN END [CLOCK-TYPE (default-date-clock-type)]) => time-period
Returns a new time-period object. The clock types must be compatible.
BEGIN maybe a seconds value, a date, or a time (except time-duration). A seconds value or date are converted to CLOCK-TYPE.
END maybe a seconds value, a date, or a time. A seconds value or date are converted to the same clock type as BEGIN. A time-duration is treated as an offset from BEGIN.
copy-time-period[procedure] (copy-time-period TIME-PERIOD) => time-period
Returns a copy of TIME-PERIOD.
time-period-begin[procedure] (time-period-begin TIME-PERIOD)
Returns the start time for the TIME-PERIOD.
time-period-end[procedure] (time-period-end TIME-PERIOD) => time
Returns the end time for the TIME-PERIOD.
time-period-last[procedure] (time-period-last TIME-PERIOD) => time
Returns the last time for the TIME-PERIOD; (time-period-end - 1ns).
time-period-type[procedure] (time-period-type TIME-PERIOD) => symbol
Returns the clock-type of the TIME-PERIOD.
time-period?[procedure] (time-period? OBJECT) => boolean
Is OBJECT a time-period?
time-period-null?[procedure] (time-period-null? TIME-PERIOD) => boolean
Is the TIME-PERIOD null?
time-period-length[procedure] (time-period-length TIME-PERIOD) => time
Returns the time-duration of the TIME-PERIOD.
time-period-compare[procedure] (time-period-compare TIME-PERIOD-1 TIME-PERIOD-2) => integer
Returns -1 when TIME-PERIOD-1 < TIME-PERIOD-2, 0 when TIME-PERIOD-1 = TIME-PERIOD-2 and 1 TIME-PERIOD-1 > TIME-PERIOD-2.
time-period=?[procedure] (time-period=? TIME-PERIOD-1 TIME-PERIOD-2) => boolean
Does TIME-PERIOD-1 begin & end with TIME-PERIOD-2?
time-period<?[procedure] (time-period<? TIME-PERIOD-1 TIME-PERIOD-2) => boolean
Does TIME-PERIOD-1 end before TIME-PERIOD-2 begins?
time-period>?[procedure] (time-period>? TIME-PERIOD-1 TIME-PERIOD-2) => boolean
Does TIME-PERIOD-1 begin after TIME-PERIOD-2 ends?
time-period<=?[procedure] (time-period<=? TIME-PERIOD-1 TIME-PERIOD-2) => boolean
Does TIME-PERIOD-1 end on or before TIME-PERIOD-2 begins?
time-period>=?[procedure] (time-period>=? TIME-PERIOD-1 TIME-PERIOD-2) => boolean
Does TIME-PERIOD-1 begin on or after TIME-PERIOD-2 ends?
time-period-preceding[procedure] (time-period-preceding TIME-PERIOD-1 TIME-PERIOD-2) => (union boolean time-period)
Return the portion of TIME-PERIOD-1 before TIME-PERIOD-2 or #f when it doesn't precede.
time-period-succeeding[procedure] (time-period-succeeding TIME-PERIOD-1 TIME-PERIOD-2) => (union boolean time-period)
Return the portion of TIME-PERIOD-1 after TIME-PERIOD-2 or #f when it doesn't succeed.
time-period-contains/period?[procedure] (time-period-contains/period? TIME-PERIOD-1 TIME-PERIOD-2) => boolean
Is TIME-PERIOD-2 within TIME-PERIOD-1?
time-period-contains/time?[procedure] (time-period-contains/time? TIME-PERIOD TIME) => boolean
Is TIME within TIME-PERIOD?
TIME is converted to a compatible clock-type if possible.
time-period-contains/date?[procedure] (time-period-contains/date? TIME-PERIOD DATE) => boolean
Is DATE within TIME-PERIOD?
DATE is converted to a compatible time if possible.
time-period-contains?[procedure] (time-period-contains? TIME-PERIOD OBJECT) => boolean
Is OBJECT within TIME-PERIOD?
OBJECT maybe a time, date, or time-period.
time-period-intersects?[procedure] (time-period-intersects? TIME-PERIOD-1 TIME-PERIOD-2) => boolean
Does TIME-PERIOD-2 overlap TIME-PERIOD-1?
time-period-intersection[procedure] (time-period-intersection TIME-PERIOD-1 TIME-PERIOD-2) => (union boolean time-period)
The overlapping time-period of TIME-PERIOD-2 and TIME-PERIOD-1, or #f when no overlap.
time-period-union[procedure] (time-period-union TIME-PERIOD-1 TIME-PERIOD-2) => (union boolean time-period)
Returns the time-period spanned by TIME-PERIOD-1 and TIME-PERIOD-2, or #f when they do not intersect.
time-period-span[procedure] (time-period-span TIME-PERIOD-1 TIME-PERIOD-2) => time-period
Returns the time-period spanned by TIME-PERIOD-1 and TIME-PERIOD-2, including any gaps.
time-period-shift[procedure] (time-period-shift TIME-PERIOD DURATION) => time-period
Returns a copy of TIME-PERIOD shifted by DURATION.
time-period-shift![procedure] (time-period-shift! TIME-PERIOD DURATION) => time-period
Returns TIME-PERIOD shifted by DURATION.
- This module exports the time, date, timezone, and io APIs.
- The string->date and scan-date procedures will not create an incomplete date. At a minimum the input must include day, month and year components; the time and timezone components default to 0 and the locale, respectively.
- 31 December 1 BCE + 1 day => 1 January 1 CE. There is no year 0. Unlike the ISO 8601 convention do not subtract 1 when converting a year BCE to a SRFI 19 year, just negate the year.
- The SRFI 18 current-time and time? bindings conflict with SRFI 19 bindings.
- A SRFI 18 time object is not accepted except by the conversion procedures.
- The expression (time=? (seconds->time (nanoseconds->seconds (time->nanoseconds <time-duration>))) <time-duration>) might be #f, due to the use of inexact arithmetic.
- Be careful using the procedures that return some form of 'julian-day'. These are implemented using the full numeric tower and will return rational numbers. Performing arithmetic with such a result will require the "numbers" egg. See the file "srfi-19-test.scm" in this egg for an example.
This will be a problem with code that assumes fixnum and/or flonum only numbers. Perhaps an intermediate file that wraps any 'julian-day' calls and coerces to an inexact number. Use the wrapped 'julian-day' call in the problematic code.
Bugs and Limitations
- Local timezone information is not necessarily valid for historic dates and problematic for future dates. Daylight saving time is especially an issue. Conversion of a time or seconds value to a local date will use the current timezone offset value. The current offset will reflect the daylight saving time status. So target dates outside of the DST period will be converted incorrectly!
- Will not read years less than 1 properly. The ISO 8601 year convention for years 1 BCE and before and years 10000 CE and after is not supported.
- Cannot swap SRFI 29 bundle. Fixed at load time.
- Fix for flonum (current-milliseconds). Fix for date-add/subtract-duration (actually time->date support for date-timezone-info).
- Bug fix for non-unique month name key for May. Added padding character override information. The ZONE-OFFSET argument of make-date is optional. Added German bundle by Moritz Heidkamp.
- Use of record-variants extension.
- Bug fix for missing seconds->time & seconds->date in srfi-19. (Reported by Alex Suraci)
- Bug fix for format-date output port check, format-date and scan-date argument type checks.
- Bug fix for ~y input conversion.
- Initial Chicken 4 release
Copyright (C) 2009-2010 Kon Lovett. All rights reserved.
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the Software), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED ASIS, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Copyright (C) I/NET, Inc. (2000, 2002, 2003). All Rights Reserved. Copyright (C) Neodesic Corporation (2000). All Rights Reserved.
This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to the Scheme Request For Implementation process or editors, except as needed for the purpose of developing SRFIs in which case the procedures for copyrights defined in the SRFI process must be followed, or as required to translate it into languages other than English.
The limited permissions granted above are perpetual and will not be revoked by the authors or their successors or assigns.
This document and the information contained herein is provided on an "AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.