Skip to content

Latest commit

 

History

History
488 lines (337 loc) · 19 KB

PLUGINS.md

File metadata and controls

488 lines (337 loc) · 19 KB
Raw

Plugins

This library is oriented towards minimalism, so it may seem to some developers to be lacking in features. The plugin is the most realistic solution to such dissatisfaction. By importing plugins, you can extend the functionality of this library, primarily the formatter and parser.

The formatter is used in format(), etc., the parser is used in parse(), preparse(), isValid(), etc.

Usage

  • ES Modules:
import date from 'date-and-time';
// Import the plugin named "foobar".
import foobar from 'date-and-time/plugin/foobar';

// Apply the "foobar" to the library.
date.plugin(foobar);
  • CommonJS:
const date = require('date-and-time');
// Import the plugin named "foobar".
const foobar = require('date-and-time/plugin/foobar');

// Apply the "foobar" to the library.
date.plugin(foobar);
  • ES Modules for the browser:
<script type="module">
import date from '/path/to/date-and-time.es.min.js';
// Import the plugin named "foobar".
import foobar from '/path/to/date-and-time/plugin/foobar.es.js';

// Apply the "foobar" to the library.
date.plugin(foobar);
</script>
  • Older browser:
<script src="/path/to/date-and-time.min.js"></script>
<!-- Import the plugin named "foobar". -->
<script src="/path/to/plugin/foobar.js"></script>

<script>
// Apply the "foobar" to the library.
date.plugin('foobar');
</script>

Note

  • If you want to use ES Modules in Node.js without a transpiler, you need to add "type": "module" in your package.json or change your file extension from .js to .mjs.

Plugin List

  • day-of-week

    • It adds "dummy" tokens for day of week to the parser.

  • meridiem

    • It adds various notations for AM PM.

  • microsecond

    • It adds tokens for microsecond to the parser.

  • ordinal

    • It adds ordinal notation of date to the formatter.

  • timespan

    • It adds timeSpan() function that calculates the difference of two dates to the library.

  • timezone

    • It adds formatTZ(), parseTZ(), transformTZ(), addYearsTZ(), addMonthsTZ() and addDaysTZ() that support IANA time zone names to the library.

  • two-digit-year

    • It adds two-digit year notation to the parser.

day-of-week

It adds tokens for day of week to the parser. Although day of week is not significant information for the parser to identify a date, these tokens are sometimes useful. For example, when a string to be parsed contains a day of week, and you just want to skip it.

formatter:

There is no change.

parser:

token meaning acceptable examples
dddd long Friday, Sunday
ddd short Fri, Sun
dd very short Fr, Su 
const date = require('date-and-time');
// Import "day-of-week" plugin as a named "day_of_week".
const day_of_week = require('date-and-time/plugin/day-of-week');

// Apply the "day_of_week" plugin to the library.
date.plugin(day_of_week);

// You can write like this.
date.parse('Thursday, March 05, 2020', 'dddd, MMMM, D YYYY');
// You can also write like this, but it is not versatile because length of day of week are variant.
date.parse('Thursday, March 05, 2020', '        , MMMM, D YYYY');
date.parse('Friday, March 06, 2020', '      , MMMM, D YYYY');

meridiem

It adds various notations for AM PM.

formatter:

token meaning output examples
AA uppercase with ellipsis A.M., P.M.
a lowercase am, pm
aa lowercase with ellipsis a.m., p.m.

parser:

token meaning acceptable examples
AA uppercase with ellipsis A.M., P.M.
a lowercase am, pm
aa lowercase with ellipsis a.m., p.m. 
const date = require('date-and-time');
// Import "meridiem" plugin.
const meridiem = require('date-and-time/plugin/meridiem');

// Apply "medidiem" plugin to the library.
date.plugin(meridiem);

// This is default behavior of the formatter.
date.format(new Date(), 'hh:mm A');     // => '12:34 PM'

// These are added tokens to the formatter.
date.format(new Date(), 'hh:mm AA');    // => '12:34 P.M.'
date.format(new Date(), 'hh:mm a');     // => '12:34 pm'
date.format(new Date(), 'hh:mm aa');    // => '12:34 p.m.'

// This is default behavior of the parser.
date.parse('12:34 PM', 'hh:mm A');      // => Jan 1 1970 12:34:00

// These are added tokens to the parser.
date.parse('12:34 P.M.', 'hh:mm AA');   // => Jan 1 1970 12:34:00
date.parse('12:34 pm', 'hh:mm a');      // => Jan 1 1970 12:34:00
date.parse('12:34 p.m.', 'hh:mm aa');   // => Jan 1 1970 12:34:00

This plugin has a breaking change. In previous versions, the A token for the parser could parse various notations for AM PM, but in the new version, it can only parse AM and PM. For other notations, a dedicated token is now provided for each.

microsecond

It adds tokens for microsecond to the parser. If a time string to be parsed contains microsecond, these tokens are useful. In JS, however, it is not supported microsecond accuracy, a parsed value is rounded to millisecond accuracy.

formatter:

There is no change.

parser:

token meaning acceptable examples
SSSSSS high accuracy 753123, 022113
SSSSS middle accuracy 75312, 02211
SSSS low accuracy 7531, 0221 
const date = require('date-and-time');
// Import "microsecond" plugin.
const microsecond = require('date-and-time/plugin/microsecond');

// Apply "microsecond" plugin to the library.
date.plugin(microsecond);

// A date object in JavaScript supports `millisecond` (ms) like this:
date.parse('12:34:56.123', 'HH:mm:ss.SSS');

// 4 or more digits number sometimes seen is not `millisecond`, probably `microsecond` (μs):
date.parse('12:34:56.123456', 'HH:mm:ss.SSSSSS');

// 123456µs will be rounded to 123ms.

ordinal

It adds DDD token that output ordinal notation of date to the formatter.

formatter:

token meaning output examples
DDD ordinal notation of date 1st, 2nd, 3rd, 31th

parser:

There is no change.

const date = require('date-and-time');
// Import "ordinal" plugin.
const ordinal = require('date-and-time/plugin/ordinal');

// Apply "ordinal" plugin to the library.
date.plugin(ordinal);

// These are default behavior of the formatter.
date.format(new Date(), 'MMM D YYYY');    // => Jan 1 2019
date.format(new Date(), 'MMM DD YYYY');   // => Jan 01 2019

// `DDD` token outputs ordinal number of date.
date.format(new Date(), 'MMM DDD YYYY');  // => Jan 1st 2019

timespan

It adds timeSpan() function that calculates the difference of two dates to the library. This function is similar to subtract(), the difference is that it can format the calculation results.

timeSpan(date1, date2)

  • @param {Date} date1 - A Date object
  • @param {Date} date2 - A Date object
  • @returns {Object} The result object of subtracting date2 from date1
const date = require('date-and-time');
// Import "timespan" plugin.
const timespan = require('date-and-time/plugin/timespan');

// Apply "timespan" plugin to the library.
date.plugin(timespan);

const now = new Date(2020, 2, 5, 1, 2, 3, 4);
const new_years_day = new Date(2020, 0, 1);

date.timeSpan(now, new_years_day).toDays('D HH:mm:ss.SSS'); // => '64 01:02:03.004'
date.timeSpan(now, new_years_day).toHours('H [hours] m [minutes] s [seconds]');  // => '1537 hours 2 minutes 3 seconds'
date.timeSpan(now, new_years_day).toMinutes('mmmmmmmmmm [minutes]');  // => '0000092222 minutes'

Like subtract(), timeSpan() returns an object with functions like this:

function description
toDays Outputs in dates
toHours Outputs in hours
toMinutes Outputs in minutes
toSeconds Outputs in seconds
toMilliseconds Outputs in milliseconds

In these functions can be available some tokens to format the calculation result. Here are the tokens and their meanings:

function available tokens
toDays D, H, m, s, S
toHours H, m, s, S
toMinutes m, s, S
toSeconds s, S
toMilliseconds S
token meaning
D date
H 24-hour
m minute
s second
S millisecond

timezone

It adds formatTZ(), parseTZ(), transformTZ(), addYearsTZ(), addMonthsTZ() and addDaysTZ() that support IANA time zone names (America/Los_Angeles, Asia/Tokyo, and so on) to the library.

formatTZ(dateObj, arg[, timeZone])

  • @param {Date} dateObj - A Date object
  • @param {string|Array.<string>} arg - A format string or its compiled object
  • @param {string} [timeZone] - Output as this time zone
  • @returns {string} A formatted string

formatTZ() is upward compatible with format(). Tokens available for arg are the same as those for format(). If timeZone is omitted, this function assumes timeZone to be the local time zone and outputs a string. This means that the result is the same as when format() is used.

parseTZ(dateString, arg[, timeZone])

  • @param {string} dateString - A date and time string
  • @param {string|Array.<string>} arg - A format string or its compiled object
  • @param {string} [timeZone] - Input as this time zone
  • @returns {Date} A Date object

parseTZ() is upward compatible with parse(). Tokens available for arg are the same as those for parse(). timeZone in this function is used as supplemental information. if dateString contains a time zone offset value (i.e. -0800, +0900), timeZone is not be used. If dateString doesn't contain a time zone offset value and timeZone is omitted, this function assumes timeZone to be the local time zone. This means that the result is the same as when parse() is used.

transformTZ(dateString, arg1, arg2[, timeZone])

  • @param {string} dateString - A date and time string
  • @param {string|Array.<string>} arg1 - A format string before transformation or its compiled object
  • @param {string|Array.<string>} arg2 - A format string after transformation or its compiled object
  • @param {string} [timeZone] - Output as this time zone
  • @returns {string} A formatted string

transformTZ() is upward compatible with transform(). dateString must itself contain a time zone offset value (i.e. -0800, +0900), otherwise this function assumes it is the local time zone. Tokens available for arg1 are the same as those for parse(), also tokens available for arg2 are the same as those for format(). timeZone is a IANA time zone names, which is required to output a new formatted string. If it is omitted, this function assumes timeZone to be the local time zone. This means that the result is the same as when transform() is used.

addYearsTZ(dateObj, years[, timeZone])

  • @param {Date} dateObj - A Date object
  • @param {number} years - The number of years to add
  • @param {string} [timeZone] - The time zone to use for the calculation
  • @returns {Date} The Date object after adding the specified number of years

addYearsTZ() can calculate adding years in the specified time zone regardless of the execution environment.

addMonthsTZ(dateObj, months[, timeZone])

  • @param {Date} dateObj - A Date object
  • @param {number} months - The number of months to add
  • @param {string} [timeZone] - The time zone to use for the calculation
  • @returns {Date} The Date object after adding the specified number of months

addMonthsTZ() can calculate adding months in the specified time zone regardless of the execution environment.

addDaysTZ(dateObj, days[, timeZone])

  • @param {Date} dateObj - A Date object
  • @param {number} days - The number of days to add
  • @param {string} [timeZone] - The time zone to use for the calculation
  • @returns {Date} The Date object after adding the specified number of days

addDaysTZ() can calculate adding days in the specified time zone regardless of the execution environment.

const date = require('date-and-time');
// Import "timezone" plugin.
const timezone = require('date-and-time/plugin/timezone');

// Apply "timezone" plugin to the library.
date.plugin(timezone);

const d1 = new Date(Date.UTC(2021, 2, 14, 9, 59, 59, 999)); // 2021-03-14T09:59:59.999Z
date.formatTZ(d1, 'MMMM DD YYYY H:mm:ss.SSS [UTC]Z', 'America/Los_Angeles'); // March 14 2021 1:59:59.999 UTC-0800

const d2 = new Date(Date.UTC(2021, 2, 14, 10, 0, 0, 0));  // 2021-03-14T10:00:00.000Z
date.formatTZ(d2, 'MMMM DD YYYY H:mm:ss.SSS [UTC]Z', 'America/Los_Angeles'); // March 14 2021 3:00:00.000 UTC-0700

// Parses the date string assuming that the time zone is "Pacific/Honolulu" (UTC-1000).
date.parseTZ('Sep 25 2021 4:00:00', 'MMM D YYYY H:mm:ss', 'Pacific/Honolulu');  // 2021-09-25T14:00:00.000Z

// Parses the date string assuming that the time zone is "Europe/London" (UTC+0100).
date.parseTZ('Sep 25 2021 4:00:00', 'MMM D YYYY H:mm:ss', 'Europe/London');  // 2021-09-25T03:00:00.000Z

// Transforms the date string from EST (Eastern Standard Time) to PDT (Pacific Daylight Time).
date.transformTZ('2021-11-07T03:59:59 UTC-0500', 'YYYY-MM-DD[T]HH:mm:ss [UTC]Z', 'MMMM D YYYY H:mm:ss UTC[Z]', 'America/Los_Angeles'); // November 7 2021 1:59:59 UTC-0700

// Transforms the date string from PDT(Pacific Daylight Time) to JST (Japan Standard Time).
date.transformTZ('2021-03-14T03:00:00 UTC-0700', 'YYYY-MM-DD[T]HH:mm:ss [UTC]Z', 'MMMM D YYYY H:mm:ss UTC[Z]', 'Asia/Tokyo');   // March 14 2021 19:00:00 UTC+0900

Caveats

  • This plugin uses the Intl object to parse IANA time zone names. Note that if you use this plugin in older browsers, this may NOT be supported there. At least it does not work in IE.
  • If you don't need to use IANA time zone names, you should not use this plugin for performance reasons. Recommended to use format() and parse().

Start of DST (Daylight Saving Time)

For example, in the US, when local standard time is about to reach 02:00:00 on Sunday, 14 March 2021, the clocks are set forward by 1 hour to 03:00:00 local daylight time instead. As a result, the time from 02:00:00 to 02:59:59 on 14 March 2021 does not exist. In such edge cases, parseTZ() will handle the case in the following way:

date.parseTZ('Mar 14 2021 1:59:59', 'MMM D YYYY H:mm:ss', 'America/Los_Angeles'); // => 2021-03-14T09:59:59Z
date.parseTZ('Mar 14 2021 3:00:00', 'MMM D YYYY H:mm:ss', 'America/Los_Angeles'); // => 2021-03-14T10:00:00Z

// These times don't actually exist, but parseTZ() will handle as follows:
date.parseTZ('Mar 14 2021 2:00:00', 'MMM D YYYY H:mm:ss', 'America/Los_Angeles'); // => 2021-03-14T10:00:00Z
date.parseTZ('Mar 14 2021 2:59:59', 'MMM D YYYY H:mm:ss', 'America/Los_Angeles'); // => 2021-03-14T10:59:59Z

End of DST

Also, when local daylight time is about to reach 02:00:00 on Sunday, 7 November 2021, the clocks are set back by 1 hour to 01:00:00 local standard time instead. As a result, the time from 01:00:00 to 01:59:59 on 7 November 2021 occurs twice. Because this time period happens twice, parseTZ() assumes that the time is the earlier one (during DST) in order to make the result unique:

// This time is DST or PST? The parseTZ() always assumes that it is DST.
date.parseTZ('Nov 7 2021 1:59:59', 'MMM D YYYY H:mm:ss', 'America/Los_Angeles');  // => 2021-11-07T08:59:59Z
// This time is already PST.
date.parseTZ('Nov 7 2021 2:00:00', 'MMM D YYYY H:mm:ss', 'America/Los_Angeles');  // => 2021-11-07T10:00:00Z

In the first example above, if you want parseTZ() to parse the time as PST, you need to pass a date string containing the time zone offset value. In this case, it is better to use parse() instead:

date.parse('Nov 7 2021 1:59:59 -0800', 'MMM D YYYY H:mm:ss Z'); // => 2021-11-07T09:59:59Z

Token Extension

This plugin also adds tokens for time zone name to the formatter.

formatter:

token meaning output examples
z time zone name abbreviation PST, EST
zz time zone name Pacific Standard Time

The z and zz are lowercase. Also, currently it does not support output other than English.

parser:

There is no change.

const date = require('date-and-time');
// Import "timezone" plugin.
const timezone = require('date-and-time/plugin/timezone');

// Apply "timezone" plugin to the library.
date.plugin(timezone);

const d1 = new Date(Date.UTC(2021, 2, 14, 9, 59, 59, 999));
date.format(d1, 'MMMM DD YYYY H:mm:ss.SSS zz');
// March 14 2021 1:59:59.999 Pacific Standard Time

date.format(d1, 'MMMM DD YYYY H:mm:ss.SSS zz', true);
// March 14 2021 9:59:59.999 Coordinated Universal Time

date.formatTZ(d1, 'MMMM DD YYYY H:mm:ss.SSS z', 'Asia/Tokyo');
// March 14 2021 18:59:59.999 JST

// Transforms the date string from EST (Eastern Standard Time) to PDT (Pacific Daylight Time).
date.transform('2021-11-07T03:59:59 UTC-0500', 'YYYY-MM-DD[T]HH:mm:ss [UTC]Z', 'MMMM D YYYY H:mm:ss z');
// November 7 2021 1:59:59 PDT

two-digit-year

It adds YY token to the parser. This token will convert the year 69 or earlier to 2000s, the year 70 or later to 1900s. In brief:

examples result
00, 01, 02, ..., 68, 69 2000s
70, 71, 72, ..., 98, 99 1900s

formatter:

There is no change.

parser:

token meaning acceptable examples
YY two-digit year 90, 00, 08, 19 
const date = require('date-and-time');
// Import "two-digit-year" plugin as a named "two_digit_year".
const two_digit_year = require('date-and-time/plugin/two-digit-year');

// This is the default behavior of the parser.
date.parse('Dec 25 69', 'MMM D YY');      // => Invalid Date

// Apply the "two_digit_year" plugin to the library.
date.plugin(two_digit_year);

// The `YY` token convert the year 69 or earlier to 2000s, the year 70 or later to 1900s.
date.parse('Dec 25 69', 'MMM D YY');      // => Dec 25 2069
date.parse('Dec 25 70', 'MMM D YY');      // => Dec 25 1970

This plugin has a breaking change. In previous versions, this plugin overrode the default behavior of the Y token, but this has been obsolete.