spencermountain
near-DST inconsistency in PST
Hi,
I'm interested in contributing to the library but I'm unable to get the daylight savings test cases passing after pulling from master. I'm currently based in mountain time and when I update my system clock to Toronto the tests end up passing. I'm happy to take a stab at fixing the issue but would love a hint to get me started.
Thanks!
whoa! i just reproduced this. sorry - I thought these days were past. Feel-free to ignore these, i will take a look at them. Thanks for the heads-up. 😅
yeah, looks like it's #235 again - off by an hour, within an hour of a DST change:
let d = spacetime('2020-11-01T01:32:00.000-06:00')
console.log(d.time()) // 1:32am
d = d.add(30, 'minutes')
console.log(d.time() === '2:02am') // 1:02am
the tests pass in all the other locales I tried, briefly. Will take a closer look when I have some time. cheers
@spencermountainwere you able to resolve this issue? Experienced this issue in production starting yesterday (Monday, March 8th, 2021)...one year after daylight savings 2020. Any insight would be greatly appreciated.
My interpretation is, It’s basically only occurring if the user device is set to CST timezone and DST is like on that day. I can’t exactly figure out the full window where this occurs. Best I can tell is it translates the wrong time the entire day of DST except for 7-9pm EST
Literally bringing my team down to our knees lol
whoa - @ericawilson9 if you wanna join a skype call today or something, my email is [email protected]
yeah, I don't understand why the entire day would be effected. The only effect of this (known) bug is that people in 1-2am may think they're in 2-3am, which they would be soon, anyways. I've always understood it to be very minor.
It sounds like you're running on an older version - and this edge-case has improved since. It may be worth trying.
Let me know if I can help. The within-1-hour issue still exists, but you may be experiencing something else.
cheers