DateTime object with timezone GMT+XX.XX

[Amit Khan]

DateTime() returns a wrong DateTime object when local time zone offset from GMT is not an integer. Like for a local time zone GMT+05.30, DateTime() object is returning an DateTime object where time zone offset is set to GMT+5 - this is resulting 30 minutes of difference when we are translating the current time to some other time zope.

It seems the problem is in constructor (init() method) of DateTime class.

The problem remains in 2.5+ versions as well like 2.6, 2.7 etc.
Best Regards
Amit

[Andreas Jung (ajung)]

Changes: submitter email, importance (critical => medium)

[Amit Khan (amit-khan)]

When can we expect a patch for the same? Is there any work around we may use till the patch is available for it?

Best Regards
Amit Khan

[Andreas Jung (ajung)]

I just changed the importance because this issue is not critical and it does not mean that someone is assigned to handle this issue at this time...patches are welcome.

[Lupa Zurven (yaybee)]

Perhaps I am provincial, but in my part of the world there are no fractional time zones. I don't see a bug here. DateTime should only accept an int for an offset value because time zones come in full hours of offset, not fractional hours. I think it is nice (a feature, not a bug) that it accepts a float like 5.30 at all, and converts it to an int for the time zone conversion. I would consider helping on this one if I could see what the problem might be, but this behavior sounds exactly right to me. Amit, do you have more information about what you think the proper behavior should be?

[Andreas Jung (ajung)]

Please submit a patch if it is important for you but there is nothing about this problem that makes it "critical".

[TinoW (tino-wildenhain)]

Changes: edited transcript, new comment

India that is. See:
http://lists.gnu.org/archive/html/bug-cvs/2004-04/msg00320.html
there are other people biten by this :-)
Maybe the OP could provide a patch?

[Andreas Jung (ajung)]

What is the "OP"?

[TinoW (tino-wildenhain)]

amit? :-)

[Lennart Regebro (regebro-gmail)]

I think that this has nothing to do with fractionality, but timezone formatting. GMT+0530 works fine, GMT+05.00 does not.

This may indeed be a bug, but I agree it's minor. Both TZ GMT+0530 and Asia/Calcutta works fine for me.

[Florent Guillaume (efge)]

FYI: OP means Original Poster.

[Florent Guillaume (efge)]

Note that I can find no standard that allow GMT+05.30. Only GMT+0530 or GMT+05:30 are correct. Still, DateTime doesn't understant the latter. RFC2822 only allows +HHMM timezones.

See http://www.cl.cam.ac.uk/~mgk25/iso-time.html for an ISO 8601 reference.

[Jens Vagelpohl (dataflake-deactivatedaccount-deactivatedaccount)]

The single issue here is supporting the colon-separated notation (e.g. "GMT+05.30" or "+05:30"), the dot-separated notation "GMT+05.30" is not covered by any standard.

Unfortunately the code in DateTime.DateTime.DateTime._parse is a complete hairball.

[Jens Vagelpohl (dataflake-deactivatedaccount-deactivatedaccount)]

Corrected comment:

The single issue here is supporting the colon-separated notation (e.g. "GMT+05:30" or "+05:30"), the dot-separated notation "GMT+05.30" is not covered by any standard.

[Colin Watson (cjwatson)]

The datetime project on Launchpad has been archived at the request of the Zope developers (see https://answers.launchpad.net/launchpad/+question/683589 and https://answers.launchpad.net/launchpad/+question/685285). If this bug is still relevant, please refile it at https://github.com/zopefoundation/datetime.