Parsing of IVEF timestamp fields

28 views
Skip to first unread message

Lukassen, Chris

unread,
Aug 17, 2010, 4:08:56 AM8/17/10
to ivef...@googlegroups.com

Hi everyone,

we are discussing a compatibilty feature in the IVEF SDK. Basically we
want to break compatibilty with the current version with respect to the
missing "Z" on the timestamp.

Earlier implementations sometimes "forgot" to add the Z behind a
timestamp to indicate UTC, however it would be a UTC timestamp. As a
result we added the indication in the parser. According to ISO this
should behave differently.

Question: are there still applications out there that send UTC but not
the Z?

If we implement this, it will break compatibilty and also some tests.
(nothing we cant fix)
Altenatively, we can split the generator (stil in 0.1 version) we can
create a branch with the patch behaviour and use it for IVEF 0.1
(remains compatible) and create a branch without the patch (0.2) this
could be used for IVEF 0.2. This doubles the work on the generator and i
am not looking forward to that.

Question: do you want split features in the different versions of the
library.

Please reply quick, without a request to retain the functionality i will
create a issue by the end of the week to remove it.

best regards

Chris


-----Original Message-----
From: Heres, Arjen [mailto:he...@hitt.home]
Sent: dinsdag 17 augustus 2010 9:06
To: Lukassen, Chris
Cc: Nijenhuis, Erwin
Subject: Re: Parsing of IVEF timestamp fields

De parsing is inderdaad niet zo vreemd als je weet hoe het ontstaan is.
De hack kan er wat mij betreft uit, moeten we alleen wel even checken
dat alle applicaties tegenwoordig netjes de Z sturen in alle tijden.

Groeten,

Arjen.
tel. 2586

----- Original Message -----
From: "Chris Lukassen" <luka...@hitt.nl>
To: "Erwin Nijenhuis" <nij...@hitt.nl>
Cc: "Arjen Heres" <he...@hitt.home>
Sent: Monday, 16 August, 2010 19:09:50 GMT +01:00 Amsterdam / Berlin /
Bern / Rome / Stockholm / Vienna
Subject: Re: Parsing of IVEF timestamp fields

Mijn meaning is dat Isis 1.0 het verkeerd deed dus daarom daeze
constructie

Ivef 0.1 is downwards compatibel maar zouden we nu kunnen afsplitsen

Chris

Sent from my iPhone

Chris

unread,
Aug 26, 2010, 3:37:15 AM8/26/10
to ivef
Due to the utter lack of repsonses, we will go ahead with the
suggested solution
Reply all
Reply to author
Forward
0 new messages