Yeah I don't know the history to this type of class, but there is some stuff that looks decidedly fishy compared to standard Date methods, anyway, in this instance it checks for GMT but whoever wrote that, forgot we like DST too! ;)

I have a fix I'll pump it up later but it would probably be nice in 0.8/0.9(https://issues.apache.org/jira/browse/OODT-743) to have a todo to check the date/time routines in OODT and see if we can optimise/swap out the stuff thats in the platform for generic libraries/classes/methods.

Tom

On 26/08/14 01:30, [email protected] wrote:
It strikes me as odd that we'd see hardcoded TZ names in code like this, 
especially given the Java standard library's depth and breadth at handling 
times, dates, zones, etc.

Then again, I never ventured into the DateConvert class.

--k

PS: And that's odd too: classes are usually named after nouns!


On 2014-08-25, at 3:38 PM, Tom Barber (JIRA) <[email protected]> wrote:

    [ 
https://issues.apache.org/jira/browse/OODT-742?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14109680#comment-14109680
 ]

Tom Barber commented on OODT-742:
---------------------------------

if (tzName.equals("Greenwich Mean Time")) {
                        dateString = dateString.concat("Z");
                }

The hardcoded GMT string doesn't account for UK BST, I'll create a patch and 
send it upstream in a bit. (all you americans are very mean ;) )

Filemgr testFlat fails in GMT
-----------------------------

                Key: OODT-742
                URL: https://issues.apache.org/jira/browse/OODT-742
            Project: OODT
         Issue Type: Bug
   Affects Versions: 0.7
           Reporter: Tom Barber
           Assignee: Tom Barber

More timezone woes:
Failed tests:
  testFlat(org.apache.oodt.cas.filemgr.versioning.TestDateTimeVersioner): Generated ref 
does not equal expected ref: 
generatedRef=[file:/foo/bar/test_product/test.txt.20142508.215149], 
expected=[file:/foo/bar/test_product/test.txt.20142508.205149] 
expected:<.../test.txt.20142508.2[0]5149> but 
was:<.../test.txt.20142508.2[1]5149>


--
This message was sent by Atlassian JIRA
(v6.2#6252)


--
*Tom Barber* | Technical Director

meteorite bi
*T:* +44 20 8133 3730
*W:* www.meteorite.bi | *Skype:* meteorite.consulting
*A:* Surrey Technology Centre, Surrey Research Park, Guildford, GU2 7YG, UK

Reply via email to