[ https://issues.apache.org/jira/browse/OPENJPA-2178?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13262647#comment-13262647 ]
Mark Struberg commented on OPENJPA-2178: ---------------------------------------- postgresql-6.3 was in 1998 I honestly do not expect that we need to implement that. And even if so: they can still use the trick outlined in OPENJPA-433, but now just do it the other way around. If someone really has such an old postgresql or likes to round to 1/100 seconds for another reason, he could still add the following setting: <property name="openjpa.jdbc.DBDictionary" value="DatePrecision=CENTI"/> > PostgresDictionary > ------------------- > > Key: OPENJPA-2178 > URL: https://issues.apache.org/jira/browse/OPENJPA-2178 > Project: OpenJPA > Issue Type: Bug > Affects Versions: 2.2.0 > Reporter: Mark Struberg > Assignee: Mark Struberg > Fix For: 2.3.0 > > > We hit a problem that OpenJPA always rounds to the nearest 10ms for > PostgreSQL. We found the following old issue in which a workaround got > outlined in OPENJPA-433 > But still the question remains: PostgreSQL is perfectly fine to store > milliseconds, so why does the PostgresDictionary line 146 sets: > > datePrecision = CENTI; > ? > The generated TIMESTAMP type in PostgreSQL should even be able to store > microseconds! [1] > And that seems to be the case since quite some time now (1999) [2]. > I'm really tempted to set this to MICRO; Anyone against it? > [1] http://www.postgresql.org/docs/9.0/interactive/datatype-datetime.html > [2] http://www.postgresql.org/docs/7.0/static/datatype1134.htm -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira