Den ons 10 jan. 2024 kl 10:45 skrev Stefan Sperling <s...@apache.org>:
> On Wed, Jan 10, 2024 at 09:44:51AM +0100, Johan Corveleyn wrote: > > Interesting discussion. I agree it should at least be documented, and > > perhaps be made a bit more clear from the output of 'revert' (but not > > sure how far we can go without breaking compat). Changing the current > > behavior is probably a more risky move, given the maturity of SVN and > > backwards compatibility etc. > > Adding new notification types should not cause compatibility concerns. > If adding new notifications breaks other software which reads command > line client output than this other software has a bug: It should have > been ignoring unknown lines of output in the first place. > > API users would likewise simply need to catch the new notification type > in a switch-like statement and should also be ignoring unknown values. > > I would only see a compatibility problem if an existing notification > about an important event no longer appears. > > In the past we have made significant changes to output from commands, > such as when tree conflict detection was added in 1.6. That by itself > has not resulted in any problems, as far as I know. > Thanks for the feedback (and sorry for mixing up the threads previously saying there was no feedback here). I've made a separate commit r1915215 adding code to catch the inconsistency with the auth bits and report a separate notification. Kind regards, Daniel