Replies: 8 comments 6 replies
-
Hey! Thank you for the proposal. For starters, we would like to understand what exactly you're trying to fix, could you please provide a test case when date-fns fails to respect DST? |
Beta Was this translation helpful? Give feedback.
-
I looked into tests, and it seems you respect DST changes. That was not the case when I studied this library like 3 years ago. I can't be asked to open a PR without an advance, do you want a better test coverage since not every function is tested at dst transitions? When I emailed you I pretty much meant to create date-fns-tz with dst sensitivity when your tz change results in dst transition (My dst started 2 hours ago, and you're 3 hours behind changing time to my tz). |
Beta Was this translation helpful? Give feedback.
-
Intl api has like 2 methods which means you need to start doing tricks to
convert tz accurately because there’s no api for it.
…On Sat, 11 Jun 2022 at 09:59 Sasha Koss ***@***.***> wrote:
You're not really adding or subtracting time, but changing timezone.
Can you please elaborate, I really struggle to get the point across?
—
Reply to this email directly, view it on GitHub
<#3082 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACWP5AKVQZI7I2AVA6CH4CTVOQFIHANCNFSM5YFZIWCA>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
I don’t actually remember details myself, despite being interested in work.
…On Sat, 11 Jun 2022 at 09:59 Sasha Koss ***@***.***> wrote:
You're not really adding or subtracting time, but changing timezone.
Can you please elaborate, I really struggle to get the point across?
—
Reply to this email directly, view it on GitHub
<#3082 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACWP5AKVQZI7I2AVA6CH4CTVOQFIHANCNFSM5YFZIWCA>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
I remember what pissed me of about the API. You can't tell when tsd occurs
using API, so when you want to change locale you don't know what tz offset
you actually need. Thats different to subtracting in a sense that intl API
handles it or something.
On Sat, Jun 11, 2022 at 11:18 AM zhan ishzhanov ***@***.***>
wrote:
… I don’t actually remember details myself, despite being interested in work.
On Sat, 11 Jun 2022 at 09:59 Sasha Koss ***@***.***> wrote:
> You're not really adding or subtracting time, but changing timezone.
>
> Can you please elaborate, I really struggle to get the point across?
>
> —
> Reply to this email directly, view it on GitHub
> <#3082 (reply in thread)>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/ACWP5AKVQZI7I2AVA6CH4CTVOQFIHANCNFSM5YFZIWCA>
> .
> You are receiving this because you authored the thread.Message ID:
> ***@***.***>
>
|
Beta Was this translation helpful? Give feedback.
-
Anyway in the issue that I linked people seem to be very annoyed when
discovering that these edge cases are not covered. The most voted reply
there is about having date packages to handle edge cases and not being
agile or simple to read (having less code).
On Sat, Jun 11, 2022 at 1:41 PM zhan ishzhanov ***@***.***>
wrote:
… I remember what pissed me of about the API. You can't tell when tsd occurs
using API, so when you want to change locale you don't know what tz offset
you actually need. Thats different to subtracting in a sense that intl API
handles it or something.
On Sat, Jun 11, 2022 at 11:18 AM zhan ishzhanov ***@***.***>
wrote:
> I don’t actually remember details myself, despite being interested in
> work.
>
> On Sat, 11 Jun 2022 at 09:59 Sasha Koss ***@***.***> wrote:
>
>> You're not really adding or subtracting time, but changing timezone.
>>
>> Can you please elaborate, I really struggle to get the point across?
>>
>> —
>> Reply to this email directly, view it on GitHub
>> <#3082 (reply in thread)>,
>> or unsubscribe
>> <https://github.com/notifications/unsubscribe-auth/ACWP5AKVQZI7I2AVA6CH4CTVOQFIHANCNFSM5YFZIWCA>
>> .
>> You are receiving this because you authored the thread.Message ID:
>> ***@***.***>
>>
>
|
Beta Was this translation helpful? Give feedback.
-
Sorry if I wasn't clear enough. I want to see clear and concrete plans of what you want to do. Linking old issues won't do. Sorry about that. I want to see a test case so I can understand the exact problem you want to solve or step-by-step plans, so I can understand what's your proposal. You're touching on so many topics I'm not willing to discuss right now, from timezones support that we don't have to the library architecture and the tradeoffs we make. I would like to focus on one thing only - your proposal. Also, please make an effort and post it in a single comment. The previous conversation is impossible to follow. |
Beta Was this translation helpful? Give feedback.
-
The idea is to add intl api based timezone support even if a package for it
exists, because it doesn't appear to cover all use cases. The primary
motivation for the proposal was that I felt like I managed to manipulate
intl api to cover some of important use cases in date library previously
and that you can afford upfront payment with like 80k in open collective.
Although I should be familiar with problem domain I'm unsure myself that
using a date object is sufficient like with the current plugin.
…On Mon, 13 Jun 2022 at 10:54 Sasha Koss ***@***.***> wrote:
Sorry if I wasn't clear enough. I want to see clear and concrete plans of
what you want to do. Linking old issues won't do. Sorry about that.
I want to see a test case so I can understand the exact problem you want
to solve or step-by-step plans, so I can understand what's your proposal.
You're touching on so many topics I'm not willing to discuss right now,
from timezones support that we don't have to the library architecture and
the tradeoffs we make. I would like to focus on one thing only - your
proposal.
Also, please make an effort and post it in a single comment. The previous
conversation is impossible to follow.
—
Reply to this email directly, view it on GitHub
<#3082 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACWP5AKDAKLJXHDJJNDS6ETVO25ILANCNFSM5YFZIWCA>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
I wrote a feature that handles dst stuff in another library like 3 years ago, I thought I could implement it here. There are a bunch of issues about dst and I’m guessing your algorithm isn’t as accurate. I was hoping I could apply for open collective payment, with a advance because the repo seems inactive.
Beta Was this translation helpful? Give feedback.
All reactions