Emacs Lisp date/time functions and RFC 3339?

Emacs Lisp date/time functions and RFC 3339?

[T. V. Raman]

RFC 3339 
http://www.ietf.org/rfc/rfc3339.txt

defines a simplified/unambiguous  date/time format 
that I'm running into more and more on the Web.

The date/time handling functions in Emacs ---
calendar/time-date.el, together with the primitives for
encode-time and decode-time do not handle this format well, nor
does format-time-string.

Is there interest other than just me in getting some of this into
the core date/time libraries for Emacs? 

Note: this is not something that needs to be checked in before
releasing Emacs 22; on the other hand, if implemented, it would
be benign with respect to the rest of Emacs as it stands at present.

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs

Re: Emacs Lisp date/time functions and RFC 3339?

[Edward O'Connor]

> RFC 3339[...] defines a simplified/unambiguous date/time format that
> I'm running into more and more on the Web.
[...]]
> Is there interest other than just me in getting some of this into
> the core date/time libraries for Emacs? 

Count me in! This would be very handy for the Atom-, APP-, and
microformat-related elisp coding I've been doing lately.


Ted

-- 
Edward O'Connor
hober0@gmail.com

Ense petit placidam sub libertate quietem.

Re: Emacs Lisp date/time functions and RFC 3339?

[Richard Stallman]

    defines a simplified/unambiguous  date/time format 
    that I'm running into more and more on the Web.

What does that format look like?

    Is there interest other than just me in getting some of this into
    the core date/time libraries for Emacs? 

If people are using it, we certainly want to support it.

Re: Emacs Lisp date/time functions and RFC 3339?

[T. V. Raman]

RFC 3339 is used by the Atom Publishing Protocol and friends.

Its primary characteristics gleened from the RFC:

A)  Simple, Unambiguous 
B)  Sorting values with standard lexical sort does the write
    thing 

Examples:

2006-11-09T18:52:00-08:00
The same with an optional fractional seconds part:
2006-11-09T18:52:00.000-08:00

The only optional part in the format is the fractional seconds.

You can also write just a date, i.e. no time part, or just a
time. TZ can also be specified as [zZ] -

>>>>> "Richard" == Richard Stallman <rms@gnu.org> writes:
    Richard>     defines a simplified/unambiguous date/time
    Richard> format that I'm running into more and more on the
    Richard> Web.
    Richard> 
    Richard> What does that format look like?
    Richard> 
    Richard>     Is there interest other than just me in getting
    Richard> some of this into the core date/time libraries for
    Richard> Emacs?
    Richard> 
    Richard> If people are using it, we certainly want to support
    Richard> 

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs

Re: Emacs Lisp date/time functions and RFC 3339?

[Andreas Schwab]

"T. V. Raman" <raman@users.sf.net> writes:

> The date/time handling functions in Emacs ---
> calendar/time-date.el, together with the primitives for
> encode-time and decode-time do not handle this format well, nor
> does format-time-string.

In which way does format-time-string not handle that format?

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Emacs Lisp date/time functions and RFC 3339?

[Kim F. Storm]

Andreas Schwab <schwab@suse.de> writes:

> "T. V. Raman" <raman@users.sf.net> writes:
>
>> The date/time handling functions in Emacs ---
>> calendar/time-date.el, together with the primitives for
>> encode-time and decode-time do not handle this format well, nor
>> does format-time-string.
>
> In which way does format-time-string not handle that format?

E.g.  %z => -0800   whereas the rfc requests -08:00

Also for UTC, %z => 0000 whereas the rfc requests Z


Also, we need a function which parses this date/time format.

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Emacs Lisp date/time functions and RFC 3339?

[Andreas Schwab]

storm@cua.dk (Kim F. Storm) writes:

> Andreas Schwab <schwab@suse.de> writes:
>
>> "T. V. Raman" <raman@users.sf.net> writes:
>>
>>> The date/time handling functions in Emacs ---
>>> calendar/time-date.el, together with the primitives for
>>> encode-time and decode-time do not handle this format well, nor
>>> does format-time-string.
>>
>> In which way does format-time-string not handle that format?
>
> E.g.  %z => -0800   whereas the rfc requests -08:00

GNU date supports %:z, %::z and %:::z, should be easy to add that to
format-time-string.

> Also for UTC, %z => 0000 whereas the rfc requests Z

If you know that you are formatting UTC you can just put the Z in the
format string.  The use of Z for UTC is not mandated.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Emacs Lisp date/time functions and RFC 3339?

[Kim F. Storm]

Andreas Schwab <schwab@suse.de> writes:

> storm@cua.dk (Kim F. Storm) writes:
>
>> Andreas Schwab <schwab@suse.de> writes:
>>
>>> "T. V. Raman" <raman@users.sf.net> writes:
>>>
>>>> The date/time handling functions in Emacs ---
>>>> calendar/time-date.el, together with the primitives for
>>>> encode-time and decode-time do not handle this format well, nor
>>>> does format-time-string.
>>>
>>> In which way does format-time-string not handle that format?
>>
>> E.g.  %z => -0800   whereas the rfc requests -08:00
>
> GNU date supports %:z, %::z and %:::z, should be easy to add that to
> format-time-string.

So we should add at least %:z .

>
>> Also for UTC, %z => 0000 whereas the rfc requests Z
>
> If you know that you are formatting UTC you can just put the Z in the
> format string.  The use of Z for UTC is not mandated.

The timezone is not optional.

   time-offset     = "Z" / time-numoffset

   partial-time    = time-hour ":" time-minute ":" time-second
                     [time-secfrac]

   full-time       = partial-time time-offset

   date-time       = full-date "T" full-time


-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Emacs Lisp date/time functions and RFC 3339?

[Andreas Schwab]

storm@cua.dk (Kim F. Storm) writes:

>> If you know that you are formatting UTC you can just put the Z in the
>> format string.  The use of Z for UTC is not mandated.
>
> The timezone is not optional.

I didn't say that.  You can always use +00:00 instead of Z, so %:z would
work fine here.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Emacs Lisp date/time functions and RFC 3339?

[Kim F. Storm]

Andreas Schwab <schwab@suse.de> writes:

> storm@cua.dk (Kim F. Storm) writes:
>
>>> If you know that you are formatting UTC you can just put the Z in the
>>> format string.  The use of Z for UTC is not mandated.
>>
>> The timezone is not optional.
>
> I didn't say that.  You can always use +00:00 instead of Z, so %:z would
> work fine here.

Right.

But if we are going to change this, we could just make %:Z output "Z" for
UTC and work as %:z otherwise.

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Emacs Lisp date/time functions and RFC 3339?

[T. V. Raman]

format-time-string cannot at present output the timezone in a
form that is rfc 3339 compliant -- except for UTC.

The parsing routines do not handle dates formatted using RFC 3339 

>>>>> "Andreas" == Andreas Schwab <schwab@suse.de> writes:
    Andreas> "T. V. Raman" <raman@users.sf.net> writes:
    >> The date/time handling functions in Emacs ---
    >> calendar/time-date.el, together with the primitives for
    >> encode-time and decode-time do not handle this format
    >> well, nor does format-time-string.
    Andreas> 
    Andreas> In which way does format-time-string not handle that
    Andreas> format?
    Andreas> 
    Andreas> Andreas.
    Andreas> 
    Andreas> -- Andreas Schwab, SuSE Labs, schwab@suse.de SuSE
    Andreas> Linux Products GmbH, Maxfeldstraße 5, 90409
    Andreas> Nürnberg, Germany PGP key fingerprint = 58CA 54C7
    Andreas> 6D53 942B 1756 01D3 44D5 214B 8276 4ED5 "And now for
    Andreas> something completely different."

--
Best Regards,
--raman


Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman
IRC:    irc://irc.freenode.net/#emacs

Re: Emacs Lisp date/time functions and RFC 3339?

[T. V. Raman]

The -08:00 vs emacs' -0800 for local timezones is a bigger problem.

>>>>> "Kim" == Kim F Storm <storm@cua.dk> writes:
    Kim> Andreas Schwab <schwab@suse.de> writes:
    >> storm@cua.dk (Kim F. Storm) writes:
    >> 
    >>> Andreas Schwab <schwab@suse.de> writes:
    >>> 
    >>>> "T. V. Raman" <raman@users.sf.net> writes:
    >>>> 
    >>>>> The date/time handling functions in Emacs ---
    >>>>> calendar/time-date.el, together with the primitives for
    >>>>> encode-time and decode-time do not handle this format
    >>>>> well, nor does format-time-string.
    >>>> 
    >>>> In which way does format-time-string not handle that
    >>>> format?
    >>> 
    >>> E.g.  %z => -0800 whereas the rfc requests -08:00
    >> 
    >> GNU date supports %:z, %::z and %:::z, should be easy to
    >> add that to format-time-string.
    Kim> 
    Kim> So we should add at least %:z .
    Kim> 
    >> 
>> Also for UTC, %z => 0000 whereas the rfc requests Z
    >> 
    >> If you know that you are formatting UTC you can just put
    >> the Z in the format string.  The use of Z for UTC is not
    >> mandated.
    Kim> 
    Kim> The timezone is not optional.
    Kim> 
    Kim>    time-offset = "Z" / time-numoffset
    Kim> 
    Kim>    partial-time = time-hour ":" time-minute ":"
    Kim> time-second [time-secfrac]
    Kim> 
    Kim>    full-time = partial-time time-offset
    Kim> 
    Kim>    date-time = full-date "T" full-time
    Kim> 
    Kim> 
    Kim> -- Kim F. Storm <storm@cua.dk> http://www.cua.dk
    Kim> 
    Kim> 
    Kim> 
    Kim> _______________________________________________
    Kim> Emacs-devel mailing list Emacs-devel@gnu.org
    Kim> http://lists.gnu.org/mailman/listinfo/emacs-devel

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs

Re: Emacs Lisp date/time functions and RFC 3339?

[Miles Bader]

"T. V. Raman" <raman@users.sf.net> writes:
> The only optional part in the format is the fractional seconds.
>
> You can also write just a date, i.e. no time part, or just a
> time. TZ can also be specified as [zZ] -

It seems that nobody actually uses the "T" separator though, they just
use space (which makes it far more readable).

-Miles
-- 
o The existentialist, not having a pillow, goes everywhere with the book by
  Sullivan, _I am going to spit on your graves_.

Re: Emacs Lisp date/time functions and RFC 3339?

[Nic James Ferrier]

Miles Bader <miles@gnu.org> writes:

> "T. V. Raman" <raman@users.sf.net> writes:
>> The only optional part in the format is the fractional seconds.
>>
>> You can also write just a date, i.e. no time part, or just a
>> time. TZ can also be specified as [zZ] -
>
> It seems that nobody actually uses the "T" separator though, they just
> use space (which makes it far more readable).

Not so... some of the subscription formats (various versions of RSS)
use the T.

The T is really annoying.


-- 
Nic Ferrier
http://www.tapsellferrier.co.uk   for all your tapsell ferrier needs

Re: Emacs Lisp date/time functions and RFC 3339?

[Richard Stallman]

    2006-11-09T18:52:00-08:00
    The same with an optional fractional seconds part:
    2006-11-09T18:52:00.000-08:00

I am all in favor of making Emacs support that format
(assuming it doesn't cause some surprising harmful side effect).

Re: Emacs Lisp date/time functions and RFC 3339?

[T. V. Raman]

all uses of rfc 3339 I've seen in atom feeds use the T. 

>>>>> "Miles" == Miles Bader <miles@gnu.org> writes:
    Miles> "T. V. Raman" <raman@users.sf.net> writes:
    >> The only optional part in the format is the fractional
    >> seconds.
    >> 
    >> You can also write just a date, i.e. no time part, or just
    >> a time. TZ can also be specified as [zZ] -
    Miles> 
    Miles> It seems that nobody actually uses the "T" separator
    Miles> though, they just use space (which makes it far more
    Miles> readable).
    Miles> 
    Miles> -Miles -- o The existentialist, not having a pillow,
    Miles> goes everywhere with the book by Sullivan, _I am going
    Miles> to spit on your graves_.

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs

Re: Emacs Lisp date/time functions and RFC 3339?

[Miles Bader]

"T. V. Raman" <raman@users.sf.net> writes:
> all uses of rfc 3339 I've seen in atom feeds use the T. 

I was being a bit facetious when I said "nobody" -- but indeed, my
impression is that the T is pretty unpopular, so many people adopt the
the same format without it.

[So any parsing function should parse it with or without the T
(presumably accepting a space or underline in place of it).  Of course
it should be easy to print either variant using format-time-string.]

-Miles
-- 
My spirit felt washed.  With blood.  [Eli Shin, on "The Passion of the
Christ"]

Re: Emacs Lisp date/time functions and RFC 3339?

[T. V. Raman]

to re-iterate, the only thing format-time-string  doesn't handle,
and something I find myself hacking in privately, is the printing
of the timezone offset i.e. -08:00 vs -0800

the -0800  problem is also inherent to unix' date command.

>>>>> "Miles" == Miles Bader <miles@gnu.org> writes:
    Miles> "T. V. Raman" <raman@users.sf.net> writes:
    >> all uses of rfc 3339 I've seen in atom feeds use the T.
    Miles> 
    Miles> I was being a bit facetious when I said "nobody" --
    Miles> but indeed, my impression is that the T is pretty
    Miles> unpopular, so many people adopt the the same format
    Miles> without it.
    Miles> 
    Miles> [So any parsing function should parse it with or
    Miles> without the T (presumably accepting a space or
    Miles> underline in place of it).  Of course it should be
    Miles> easy to print either variant using
    Miles> format-time-string.]
    Miles> 
    Miles> -Miles -- My spirit felt washed.  With blood.  [Eli
    Miles> Shin, on "The Passion of the Christ"]

-- 
Best Regards,
--raman

      
Email:  raman@users.sf.net
WWW:    http://emacspeak.sf.net/raman/
AIM:    emacspeak       GTalk: tv.raman.tv@gmail.com
PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
Google: tv+raman 
IRC:    irc://irc.freenode.net/#emacs