Re: POSIX TS spec reverses the meaning of TZ offset compared to ISO (was: [FEATURE REQUEST] Timezone support in org-mode datestamps and org-agenda)

[Ihor Radchenko]

Max Nikulin  writes:

>> Because `encode-time' accepts POSIX TZ as argument. So, may as well.
>> Note that Etc/GMT-8 is also accepted as POSIX TZ.
>
> "May" does not assume "should".

It will be easier to implement POSIX TZ compared to limiting it to TZDB.
We have no way to know if a given string is a valid TZDB entry to not.

> The disappointing issue with POSIX syntax is the case of decision of 
> authorities to change time zone offset. It is neither regular DST rule 
> nor constant offset. The result is incorrect mapping between UTC and 
> local time.

We cannot do much about it, AFAIU. Maximum is trying to do feature
request in glibc, but I assume that they know about this issue.

-- 
Ihor Radchenko // yantar92,
Org mode contributor,
Learn more about Org mode at .
Support Org development at ,
or support my work at 

Re: POSIX TS spec reverses the meaning of TZ offset compared to ISO (was: [FEATURE REQUEST] Timezone support in org-mode datestamps and org-agenda)

[Max Nikulin]

On 02/02/2023 15:34, Ihor Radchenko wrote:

Max Nikulin writes:

Is there a strong reason to support POSIX TZ? Time zones with fixed
offsets are available as e.g. Etc/GMT-8 (unfortunately inverted sign as
well).


Because `encode-time' accepts POSIX TZ as argument. So, may as well.
Note that Etc/GMT-8 is also accepted as POSIX TZ.


"May" does not assume "should". I am leaving final decision to the 
person who will implement it though. I hope, there is no problem with 
such syntax on windows.


The disappointing issue with POSIX syntax is the case of decision of 
authorities to change time zone offset. It is neither regular DST rule 
nor constant offset. The result is incorrect mapping between UTC and 
local time.

Re: POSIX TS spec reverses the meaning of TZ offset compared to ISO (was: [FEATURE REQUEST] Timezone support in org-mode datestamps and org-agenda)

[Ihor Radchenko]

Max Nikulin  writes:

> On 01/02/2023 20:26, Ihor Radchenko wrote:
>> 
>> There is potential confusion coming from the different interpretations
>> of the time zone offsets used in ISO8601 and POSIX TZ specs:
>
> Ihor, I am sorry that I did not reply to your early question.
>
> Is there a strong reason to support POSIX TZ? Time zones with fixed 
> offsets are available as e.g. Etc/GMT-8 (unfortunately inverted sign as 
> well).

Because `encode-time' accepts POSIX TZ as argument. So, may as well.
Note that Etc/GMT-8 is also accepted as POSIX TZ.

-- 
Ihor Radchenko // yantar92,
Org mode contributor,
Learn more about Org mode at .
Support Org development at ,
or support my work at