You're right, it doesn't seem like parsing the feed should break the timeline, but it's good to fix this so that we eliminate any doubt that it causes problems. Besides, it's not a good idea to assume that only correctly formatted dates will be passed our way, so I wanted to put an end to ESME-275.
I've just committed a fix and closed the issue. Vassil On Sat, Nov 6, 2010 at 10:06 AM, Ethan Jewett <[email protected]> wrote: > Dick, > > If you want to disable my actions to test, that's fine. I'm fairly > confident my actions are responsible for the stack trace you sent. > (Thanks for looking in to it Vassil.) > > I just don't see how an issue with actions could be the cause of the > timeline issue. As far as I know, loading a timeline does not trigger > this action code that is failing. I could definitely be wrong, but I > think there should be a second stack trace or a Lift error message > that shows up in the logs when someone tries to load Anne or David's > user page. > > Ethan > > On Saturday, November 6, 2010, Vassil Dichev <[email protected]> wrote: >> The reason for the stack trace is that a date in an RSS feed cannot be >> parsed properly, or there is no date at all (which as I see is allowed >> by the RSS spec). It would help if we can find which RSS feed is the >> culprit, but I will try to come up with a patch anyway. >> >> Vassil >> >> >> On Sat, Nov 6, 2010 at 9:05 AM, Richard Hirsch <[email protected]> wrote: >>> I'm working on the following assumptions: >>> >>> 1) that someone anne follows is the culprit >>> 2) That the reason is some sort of an action that is called every 5 >>> minutes or so >>> 3) Checking the database shows that the most of these actions are from >>> Ethan >>> >>> What I'm going to try is to create a test user and then follow the >>> users that anne follows. >>> >>> However, you are correct in saying that I'm following you as well and >>> my timeline is working. >>> >>> If you want, we can wait until I do this test before taking such >>> drastic "Big Brother" actions. >>> >>> D. >>> >>> >>> >>> On Fri, Nov 5, 2010 at 8:25 PM, Ethan Jewett <[email protected]> wrote: >>>> I knew I shouldn't have implied it might be me :-) >>>> >>>> I definitely have RSS actions, but I'm wondering what makes you think it's >>>> my actions that are breaking only Anne & Dave's timelines? You and several >>>> others follow me, and your timelines are not broken. Before we saw this >>>> sort >>>> of thing happened because of special characters in some messages, right? >>>> >>>> I'm really wondering if the stack trace you found was the only error. I >>>> think the stack trace you found was the issue in Jira item >>>> https://issues.apache.org/jira/browse/ESME-275 This appears to still be >>>> unresolved, but I don't think we ever linked it to an issue with displaying >>>> a timeline (which should have nothing to do with a feed or action). >>>> >>>> Ethan >>>> >>>> On Fri, Nov 5, 2010 at 7:42 PM, Richard Hirsch >>>> <[email protected]>wrote: >>>> >>>>> Going through the database - looks like Ethan might be the culprit - >>>>> though I don't know why. >>>>> >>>>> @Ethan do want to disable your rss actions manually or should I do it >>>>> via big brother ;-> >>>>> >>>>> D. >>>>> >>>>> On Fri, Nov 5, 2010 at 2:07 PM, Anne Kathrine Petterøe >>>>> <[email protected]> wrote: >>>>> > LOL :-) >>>>> > >>>>> > >>>>> > On 5 November 2010 13:29, Ethan Jewett <[email protected]> wrote: >>>>> > >>>>> >> I hope it's not me. [Shakes in boots.] :-) >>>>> >> >>>>> >> On Fri, Nov 5, 2010 at 12:16 PM, Richard Hirsch <[email protected] >>>>> >> >wrote: >>>>> >> >>>>> >> > I can look at the DB when I get home and see who the culprit is. >>>>> >> > >>>>> >> > D. >>>>> >> > >>>>> >> > On Fri, Nov 5, 2010 at 11:44 AM, Anne Kathrine Petterøe >>>>> >> > <[email protected]> wrote: >>>>> >> > > Do we know who has set up an action that would create the problem? >>>>> >> > > >>>>> >> > > /Anne >>>>> >> > > >>>>> >> > > On 5 November 2010 10:07, Ethan Jewett <[email protected]> wrote: >>>>> >> > > >>>>> >> > >> Yup, that would explain it. When you sent the stack trace before, >>>>> was >>>>> >> > that >>>>> >> > >> the only one that you saw in the Stax environment or where there >>>>> >> others? >>>>> >> > >> >>>>> >> > >> Ethan >>>>> >> > >> >>>>> >> > >> On Fri, Nov 5, 2010 at 9:34 AM, Richard Hirsch < >>>>> [email protected] >>>>> >> > >> >wrote: >> -- >> Twitter: http://twitter.com/vdichev >> Blog: http://speaking-my-language.blogspot.com >> > -- Twitter: http://twitter.com/vdichev Blog: http://speaking-my-language.blogspot.com
