On 23 September 2014 19:53, Oleg Kalnichevski <[email protected]> wrote: > On Tue, 2014-09-23 at 13:00 -0400, Gary Gregory wrote: >> On Tue, Sep 23, 2014 at 12:40 PM, sebb <[email protected]> wrote: >> >> > >>> But as I said before you are welcome to extend release tools to >> > >>generate >> > >>> whatever reports you desire. >> > >> >> > >>Just because I report an issue with a process does not mean I have to >> > >>be the one to fix it. >> > >> >> > > >> > > Right. Naturally it is better to have other people do useless work than >> > do useless work yourself. >> > >> > It is not useless work. >> > The RM must - and reviewers should - run the RAT and Clirr checks. >> > This needs to be documented as part of the release voting. >> > >> >> I agree that the RM and reviewers must run or check these reports. >> It sure would waste less time if the reports were provided or instructions >> provided (I added some instructions to BUILDING.txt files yesterday FWIW). >> >> Gary >> > > Gary > > As both of us just found our reports generated by RM and those generated > locally by a reviewer may not always match.
In which case we need to find out why this is happening, and fix the problem. I also get an error running RAT on httpcore trunk unless I run mvn clean first. If I run it again, RAT complains about target/rat.txt files. No idea why this should be; I would expect target/* to be excluded wherever it occurs in the tree. > What really matters is > whether or not the source tarball compiles and passes the checks for > _you_. Reports published by RM are useless and a waste of RM time. > > If something is still unclear we should work on improving BUILDING.txt > instructions. > > Oleg > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
