On the one hand, this stuff is completely separate from the core River code, so it's inclusion shouldn't have any effect on any of the stuff you or anyone else has done recently.
On the other hand, I gone through enough stress in the past trying to work out why something failed when something unrelated "shouldn't have [had] any effect" on the task in hand. :-) So, I'm happy to wait on moving it into the trunk until the build has settled down. On Tue, Nov 2, 2010 at 12:01 PM, Patricia Shanahan <p...@acm.org> wrote: > Sim IJskes - QCG wrote: >> >> On 02-11-10 12:22, Tom Hobbs wrote: >>> >>> I'd like to give people a chance to pass their eye over it first. It >>> is just a reference implementation/example of how to do a certain >>> thing. I doubt that the code is bullet proof. >> >> If there is something wrong with it, people can file a jira for it. >> >> The selfhealingproxy is in the extra section, and there are not a lot of >> river parts depending on it, so my stake would be: commit it. > > I'd like a short delay before the next significant surgery on the trunk. > > Yesterday, I checked in a qa/build.xml change that should increase by > several hundred the number of tests in a full QA run. I've also checked in > bug fixes that make those tests pass, at least on my computer. Can we hold > off until there has been a clean Hudson run with the added tests? > > Patricia > >
