DATESEQ(1)
β¨ NAME β¨
okay so this is dateseq(1), and the official name line is exactly: Generate a sequence of date/times from FIRST to LAST, optionally in steps of β yes, it trails off like that in the source, and we are respecting the lore. π
πΈ SYNOPSIS πΈ
dateseq [OPTION]... FIRST [[INCREMENT] LAST]
π DESCRIPTION π
Generate a sequence of date/times from FIRST to LAST, optionally in steps of INCREMENT (which defaults to 1d). slay.
If LAST is omitted it defaults to now if FIRST is a date/time, or today if FIRST is a date, or time if FIRST is a time. iconic behavior, honestly.
The values of FIRST and LAST are always inclusive and no date/times before FIRST and no date/times after LAST will be printed. periodt.
Negative INCREMENTs must be given, i.e. if FIRST is newer than LAST. no cap, thatβs the rule.
Recognized OPTIONs
-h,--helpβ Print help and exit. your little βwhat is thisβ button.-V,--versionβ Print version and exit. the classic.-q,--quietβ Suppress message about date/time and duration parser errors and fix-ups. The default is to print a warning or the fixed up value and return error code2. so if you donβt pass this, expect the tea.-f,--format=STRINGβ Output format. This can either be a specifier string (similar tostrftime()'sFMT) or the name of a calendar. choose your fighter.-i,--input-format=STRING...β Input format, can be used multiple times. Each date/time will be passed to the input format parsers in the order they are given, if a date/time can be read successfully with a given input format specifier string, that value will be used. first match wins, basically.-b,--base=DTβ For underspecified input useDTas a fallback to fill in missing fields. Also used for ambiguous format specifiers to position their range on the absolute time line. Must be a date/time in ISO8601 format. If omitted defaults to the current date/time. the safety net era.-e,--backslash-escapesβ Enable interpretation of backslash escapes in the output and input format specifier strings. backslash girlies rise.--locale=LOCALEβ Format results according toLOCALE, this would only affect month and weekday names. itβs giving localized month names, not a whole personality change.--from-locale=LOCALEβ Interpret dates on stdin or the command line as coming from the localeLOCALE, this would only affect month and weekday names as input formats have to be specified explicitly. yes, explicit input formats stay explicit.-s,--skip=STRING...β Skip weekdays specified bySTRING.STRINGcan be a single weekday (Mon,Tue, etc.), and to skip several days the--skipoption can be used multiple times.STRINGcan also be a comma-separated list of weekday names, orssto skip weekends (sat+sun) altogether.STRINGcan also contain date ranges likemo-wefor monday to wednesday. very flexible, very demure.--alt-inc=STRINGβ Alternative increment to use when a date is hit that is skipped as per--skip. This increment will be applied until a non-skipped date is reached. The special case0(default) deactivates alternative incrementing. A useful value could be1dfor increasing sequences and-1dfor decreasing sequences, so if a skipped date is encountered the next non-skipped date after/before will be used. the skip-proof backup plan.--compute-from-lastβ Compute a start value fromLASTusingINCREMENT. This option has an effect only whenINCREMENTis not a divisor of the duration betweenFIRSTandLAST. In such case, an alternativeFIRSTwill be computed by consecutively subtractingINCREMENTfromLASTuntilFIRSTis hit or crossed. math girlie moment.
ποΈ FORMAT SPECS ποΈ
Format specs in dateutils are similar to posix' strftime(). However, due to a broader range of supported calendars dateutils must employ different rules. so yeah, the list is long and we are reading all of it.
π Date specs
%aβ The abbreviated weekday name%Aβ The full weekday name%_aβ The weekday name shortened to a single character (MTWRFAS)%bβ The abbreviated month name%Bβ The full month name%_bβ The month name shortened to a single character (FGHJKMNQUVXZ)%cβ The count of the weekday within the month (range00to05)%Cβ The count of the weekday within the year (range00to53)%dβ The day of the month, 2 digits (range00to31)%Dβ The day of the year, 3 digits (range000to366)%Fβ Equivalent to%Y-%m-%d(ymd's canonical format)%gβ ISO week date year without the century (range00to99)%Gβ ISO week date year including the century%jβ Equivalent to%D%mβ The month in the current calendar (range00to12)%Qβ The quarter of the year (rangeQ1toQ4)%qβ The number of the quarter (range01to04)%sβ The number of seconds since the Epoch.%uβ The weekday as number (range01to07, Sunday being07)%Uβ The week count, day of week is Sun (range00to53)%Vβ The ISO week count, day of week is Mon (range01to53)%wβ The weekday as number (range00to06, Sunday being00)%Wβ The week count, day of week is Mon (range00to53)%yβ The year without a century (range00to99)%Yβ The year including the century%_yβ The year shortened to a single digit%Zβ The zone offset in hours and minutes (HH:MM) with a preceding sign (+for offsets east of UTC,-for offsets west of UTC)%Odβ The day as roman numerals%Omβ The month as roman numerals%Oyβ The two digit year as roman numerals%OYβ The year including the century as roman numerals%rsβ In time systems whose Epoch is different from the unix Epoch, this selects the number of seconds since then.%rYβ In calendars with years that don't coincide with the Gregorian years, this selects the calendar's year.%dthβ The day of the month as an ordinal number,1st,2nd,3rd, etc.%mthβ The month of the year as an ordinal number,1st,2nd,3rd, etc.%dbβ The business day of the month (since last month's ultimo)%dBβ Number of business days until this month's ultimo
β° Time specs
%Hβ The hour of the day using a 24h clock, 2 digits (range00to23)%Iβ The hour of the day using a 12h clock, 2 digits (range01to12)%Mβ The minute (range00to59)%Nβ The nanoseconds (range000000000to999999999)%pβ The stringAMorPM, noon isPMand midnight isAM.%Pβ Like%pbut in lowercase%Sβ βThe (range00to60,60is for leap seconds)β β yes, the source leaves the noun out here; the range and leap-second fact are the important bits, bestie.%Tβ Equivalent to%H:%M:%S
β¨ General specs
%nβ A newline character%tβ A tab character%%β A literal%character
π Modifiers
%Oβ Modifier to turn decimal numbers into Roman numerals%rβ Modifier to turn units into real units%0β Modifier to turn on zero prefixes%SPCβ Modifier to turn on space prefixes%-β Modifier to turn off prefixes altogetherthβ Suffix, read and print ordinal numbersbβ Suffix, treat days as business days
β οΈ By design dates before 1601-01-01 are not supported.
For conformity here is a list of calendar designators and their corresponding format string:
ymdβ%Y-%m-%dymcwβ%Y-%m-%c-%wywdβ%rY-W%V-%ubizdaβ%Y-%m-%dblilianβn/aldnβn/ajulianβn/ajdnβn/amatlabβn/amdnβn/a
These designators can be used as output format string, moreover, @code{lilian}/@code{ldn} and @code{julian}/@code{jdn} can also be used as input format string. and yes, those raw @code{...} bits are literally in the source β iykyk π
β³ SPECIFYING DURATIONS β³
Some tools ("dadd", "dseq") need durations as their input. Durations are generally incompatible with input formats as specified by "-i|--input-format" and (at the moment) the input syntax is fixed. so donβt try to freestyle it too hard.
The general format is "+-Nunit" where "+" or "-" is the sign, "N" a number, and "unit" the unit as discussed below.
Units:
sβ secondsmβ minuteshβ hoursrsβ real-life seconds, as in including leap transitionsdβ daysbβ business daysmoβ monthsyβ years
For historical reasons, we used to accept "m" in the context of date-only input as a qualifier for months. As of 0.4.4, this is no longer the case. the more you know π
β¨ EXAMPLES β¨
okay bestie, this is the part where it all clicks π these are real, actual outputs β copy them, run them, live your truth. Watch how the same little date-range gets completely different answers depending on which flags you throw at it π
ποΈ the classic: just walk the days
$ dseq 2012-02-01 2012-03-01
2012-02-01
2012-02-02
2012-02-03
2012-02-04
2012-02-05
2012-02-06
2012-02-07
2012-02-08
2012-02-09
2012-02-10
2012-02-11
2012-02-12
2012-02-13
2012-02-14
2012-02-15
2012-02-16
2012-02-17
2012-02-18
2012-02-19
2012-02-20
2012-02-21
2012-02-22
2012-02-23
2012-02-24
2012-02-25
2012-02-26
2012-02-27
2012-02-28
2012-02-29
2012-03-01
$
no increment given? it just gives you every single day, one per line, no notes. and yeah β it sailed right through 2012-02-29 because 2012 was a leap year, she ate that π¦
π« skipping saturdays like it's nothing
$ dseq 2001-02-03 2001-03-03 --skip sat -f "%F %a"
2001-02-04 Sun
2001-02-05 Mon
2001-02-06 Tue
2001-02-07 Wed
2001-02-08 Thu
2001-02-09 Fri
2001-02-11 Sun
2001-02-12 Mon
2001-02-13 Tue
2001-02-14 Wed
2001-02-15 Thu
2001-02-16 Fri
2001-02-18 Sun
2001-02-19 Mon
2001-02-20 Tue
2001-02-21 Wed
2001-02-22 Thu
2001-02-23 Fri
2001-02-25 Sun
2001-02-26 Mon
2001-02-27 Tue
2001-02-28 Wed
2001-03-01 Thu
2001-03-02 Fri
$
look at that β 2001-02-10, 2001-02-17, 2001-02-24 are gone, not a single Sat in sight π and the -f "%F %a" is showing you the weekday so you can literally verify it yourself. trust but verify, bestie π
πͺ same thing, but counting backwards from the end
$ dseq --compute-from-last 2001-02-03 1 2001-03-03 --skip sat -f "%F %a"
2001-02-04 Sun
2001-02-05 Mon
2001-02-06 Tue
2001-02-07 Wed
2001-02-08 Thu
2001-02-09 Fri
2001-02-11 Sun
2001-02-12 Mon
2001-02-13 Tue
2001-02-14 Wed
2001-02-15 Thu
2001-02-16 Fri
2001-02-18 Sun
2001-02-19 Mon
2001-02-20 Tue
2001-02-21 Wed
2001-02-22 Thu
2001-02-23 Fri
2001-02-25 Sun
2001-02-26 Mon
2001-02-27 Tue
2001-02-28 Wed
2001-03-01 Thu
2001-03-02 Fri
$
same output as above β because with an increment of 1 and only saturdays skipped, forward and backward land in the exact same spot. cute coincidence, not a guarantee β¨
π₯ now THIS is where it gets spicy
$ dseq 2001-02-03 3 2001-03-03 --skip sat,fri -f "%F %a"
2001-02-06 Tue
2001-02-12 Mon
2001-02-15 Thu
2001-02-18 Sun
2001-02-21 Wed
2001-02-27 Tue
$
$ dseq --compute-from-last 2001-02-03 3 2001-03-03 --skip sat,fri -f "%F %a"
2001-02-04 Sun
2001-02-07 Wed
2001-02-13 Tue
2001-02-19 Mon
2001-02-22 Thu
2001-02-25 Sun
2001-02-28 Wed
$
see how these two are completely different?? π same args, same skips, only --compute-from-last flipped β and the whole list shifts. that's the vibe check: with a step of 3 and two skipped days, direction genuinely matters. no cap.
π§ plain stepping, no skips at all
$ dseq 2001-02-05 4 2001-03-04 -f "%F %a"
2001-02-05 Mon
2001-02-09 Fri
2001-02-13 Tue
2001-02-17 Sat
2001-02-21 Wed
2001-02-25 Sun
2001-03-01 Thu
$
$ dseq --compute-from-last 2001-02-05 4 2001-03-04 -f "%F %a"
2001-02-08 Thu
2001-02-12 Mon
2001-02-16 Fri
2001-02-20 Tue
2001-02-24 Sat
2001-02-28 Wed
2001-03-04 Sun
$
forward lands on the start date, backward lands on the end date. both valid, both slaying, pick your fighter π
πΈ --alt-inc, the main character
$ dseq --alt-inc 1d 2001-02-03 3 2001-03-03 --skip sat,fri -f "%F %a"
2001-02-04 Sun
2001-02-07 Wed
2001-02-11 Sun
2001-02-14 Wed
2001-02-18 Sun
2001-02-21 Wed
2001-02-25 Sun
2001-02-28 Wed
$
$ dseq --compute-from-last --alt-inc 1d 2001-02-03 3 2001-03-03 --skip sat,fri -f "%F %a"
2001-02-04 Sun
2001-02-07 Wed
2001-02-11 Sun
2001-02-14 Wed
2001-02-18 Sun
2001-02-21 Wed
2001-02-25 Sun
2001-02-28 Wed
$
and here they converge again π with --alt-inc 1d the alternative increment takes over the stepping, so the two directions give you the identical list. iconic behavior, honestly.
π duration increments hitting different
$ dseq 2001-01-01 2d 2001-01-08
2001-01-01
2001-01-03
2001-01-05
2001-01-07
$
$ dseq --compute-from-last 2001-01-01 2d 2001-01-08
2001-01-02
2001-01-04
2001-01-06
2001-01-08
$
see it? forward starts at 2001-01-01; backward lands on 2001-01-08. every other day, both times. simple, clean, obsessed π«Ά
π negative increments, because why not
$ dseq 2001-01-08 -2d 2001-01-01
2001-01-08
2001-01-06
2001-01-04
2001-01-02
$
$ dseq --compute-from-last 2001-01-08 -2d 2001-01-01
2001-01-07
2001-01-05
2001-01-03
2001-01-01
$
walking backwards through time, but which end you anchor to still flips with --compute-from-last π
β° times, not just dates
$ dseq 10:00:00 12m 11:20:00
10:00:00
10:12:00
10:24:00
10:36:00
10:48:00
11:00:00
11:12:00
$
$ dseq --compute-from-last 10:00:00 12m 11:20:00
10:08:00
10:20:00
10:32:00
10:44:00
10:56:00
11:08:00
11:20:00
$
12m means twelve minutes here, and yes it works on bare clock times, not just full dates π
π and going backward in time
$ dseq 11:20:00 -12m 10:00:00
11:20:00
11:08:00
10:56:00
10:44:00
10:32:00
10:20:00
10:08:00
$
$ dseq --compute-from-last 11:20:00 -12m 10:00:00
11:12:00
11:00:00
10:48:00
10:36:00
10:24:00
10:12:00
10:00:00
$
negative increment = walk downhill on the clock. and --compute-from-last still does its little anchor flip. periodt. π
π AUTHOR
Written by Sebastian Freundt <freundt@fresse.org>
π REPORTING BUGS
Report bugs to: https://github.com/hroptatyr/dateutils/issues
bestie if something's broken, tell the maintainers, not me β I'm just the messenger π
π SEE ALSO
The full documentation for dateseq is maintained as a Texinfo manual. If the info and dateseq programs are properly installed at your site, the command
info (dateutils)dateseq
should give you access to the complete manual.
okay that's the whole page, you're officially a dseq girlie now πΈ this gem is from dateutils 0.4.11, dated January 2024 β go forth and generate dates, bestie, and remember: when in doubt, --compute-from-last changes everything π¦β¨
π¦ where this page comes from
- alpine 3.24: dateutils-doc 0.4.11-r0 Β· BSD-3-Clause
- nixos 26.05: dateutils 0.4.11
- nixpkgs unstable: dateutils 0.4.11
- opensuse-leap 16.0: dateutils 0.4.11-bp160.1.17 Β· BSD-3-Clause
- opensuse-leap 16.0: dateutils 0.4.11-bp160.1.17 Β· BSD-3-Clause
- opensuse-leap 16.0: dateutils 0.4.11-bp160.1.17 Β· BSD-3-Clause
- opensuse-leap 16.0: dateutils 0.4.11-bp160.1.17 Β· BSD-3-Clause
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms