Detect timezone abbreviation using luxon

Question

Moment timezone results with short timezone abbreviations, e.g

moment.tz([2012, 0], 'America/New_York').format('z');    // EST
moment.tz([2012, 5], 'America/New_York').format('z');    // EDT

Is there a similar way we can achieve that using luxon

I tried offsetNameShort, but, it results to GMT+5:30 for a date like "2020-05-23T13:30:00+05:30"

Something like DateTime.fromISO(""2020-05-23T13:30:00+05:30"").toFormat('z') doesn't work either

Is there a way we can remove the +5:30 timezone from the format?

Answer 1

Review Luxon's table of formatting tokens. You want ZZZZ for the abbreviated named offset.

Examples:

DateTime.fromObject({year: 2012, month: 1, zone: 'America/New_York'})
  .toFormat('ZZZZ') //=> "EST"

DateTime.fromObject({year: 2012, month: 6, zone: 'America/New_York'})
  .toFormat('ZZZZ') //=> "EDT"

DateTime.local()
  .toFormat('ZZZZ') //=> "PDT"  (on my computer)

DateTime.fromISO("2020-05-23T13:30:00+05:30", {zone: 'Asia/Kolkata', locale: 'en-IN'})
  .toFormat('ZZZZ') //=> "IST"

Note in that last one, you also have to specify en-IN as the locale to get IST. Otherwise you will get GMT+05:30, unless the system locale is already en-IN. That is because Luxon relies upon the browser's internationalization APIs, which in turn takes its data from CLDR.

In CLDR, many names and abbreviations are designated as being specific to a given locale, rather than being used worldwide. The same thing happens with Europe/London getting GMT+1 instead of BST unless the locale is en-GB. (I personally disagree with this, but that is how it is currently implemented.)