Open GoogleCodeExporter opened 8 years ago
Is it possible to fix this issue in the near future?
Currently this is a real show-stopper in our multinational project.
Original comment by holger.k...@siemens.com
on 2 Jul 2009 at 11:38
Original comment by jsakuda
on 11 Dec 2009 at 6:47
Multinational is not the main problem here.
The problem is that date strings created by Java 6 (like 2010-05-11 ::
10:50:37:968
MESZ) are not parsable in Java 5 due to the timezone name.
The Z solution sounds like a good temporary fix, personally I would recommend
using a
proper XML date format.
Original comment by rolf.woj...@googlemail.com
on 11 May 2010 at 9:15
I created a patch against Trunk (3.5.1?) that switches Jupiter to the Z Format.
It makes the following changes:
1) Date-Strings are parsed without a timezone, if a Parse Exception occurs
2) Date-Strings are stored as "yyyy-MM-dd :: HH:mm:ss:SSS Z" instead of
"yyyy-MM-dd :: HH:mm:ss:SSS z"
Some notes regarding backwards compatibility:
- This version is backwards compatibel to reviews created with the official
version in JDK5 or JDK6
- After changes to a review it is stored in the new format (+0200 instead of
MESZ, Pattern yyyy-MM-dd :: HH:mm:ss:SSS z instead of yyyy-MM-dd ::
HH:mm:ss:SSS Z)
- the new format can be read by unmodified Jupiter versions but when they write
to it, they will keep using the "bad" z format.
Original comment by rolf.woj...@googlemail.com
on 18 Jun 2010 at 8:43
Attachments:
Find attached a jupiter.jar that was compiled using the patch above.
Replace the official one with it and your timezone problems should be a thing
of the past.
If this helps you, please drop a note so I know I didn't waste my time when I
posted this ;)
Original comment by rolf.woj...@googlemail.com
on 18 Jun 2010 at 8:47
Attachments:
Original comment by jsakuda
on 22 Jun 2010 at 11:59
We have decided get rid of the date string altogether. Both the creation and
modification dates will be stored in UTC instead. Any new reviews will use
UTC. Jupiter will still work with the exiting reviews.
And I looked at your patch. Thank you for taking the time to write it. But I
don't think we will be applying it to the existing code base as it is about to
change. I may take your idea of parsing without the time zone for the older
reviews that may have issues with the time zone name though.
Original comment by jsakuda
on 23 Jun 2010 at 10:30
Original comment by jsakuda
on 18 Jan 2011 at 5:50
Original issue reported on code.google.com by
jsakuda
on 24 Jun 2009 at 9:41