Vassil: Was this problem described in this Jira item : http://issues.apache.org/jira/browse/ESME-169
Can we close it? D. On Wed, Mar 3, 2010 at 7:27 PM, Vassil Dichev <[email protected]> wrote: > Duh, upon more careful investigation of the message it turns out that > the replacement string is not quoted. This means the message text must > contain "$1", which would normally be interpreted as the first > matching group. > > I've just committed the fix. > > > On Wed, Mar 3, 2010 at 10:25 AM, Richard Hirsch <[email protected]> > wrote: > > On Wed, Mar 3, 2010 at 9:18 AM, Vassil Dichev <[email protected]> > wrote: > >>> There is already a jira item for this problem: > >>> http://issues.apache.org/jira/browse/ESME-152 > >>> > >>> Maybe, it is the same problem..... > >> > >> Hm, there's not much detail in the issue description. > >> > >> I checked empty messages- even empty tags generate an open and closing > >> tag in Scala: > >> > >> http://www.simplyscala.com/interp?code=<body/> > >> > >> This means the message is not even empty, so probably some error > >> caused this inconsistent state. > >> > >> I'm not sure if the auto-OpenID login is related, but these should not > >> be possible anymore. The "Unknown" users in the DB are left over from > >> the time when the autologin patch was still active. Should we purge > >> the DB from these users? Most of the time they don't even have the > >> proper validated data. > > > > No problem. If the user is an old user, then it might not be cause. > > > > I'll take a look at the individual messages later and delete them 1:1 > > to find the guilty party. > > > >> > > >
