Hi, Working on the 58653, I see an additional reason to remove the $Revision, it's in the difference that it introduces in files when you compare GIT repo source with SVN Repo source. You get a difference on all the files that contain $Revision because GIT will contain "@version $Revision$" while SVN code will contain "@version $Revision: 1698378 $".
So unless there is a good reason to keep them I would rather remove them. Regards On Thu, Nov 12, 2015 at 12:56 AM, sebb <[email protected]> wrote: > On 11 November 2015 at 15:55, Felix Schumacher > <[email protected]> wrote: > > Am 11.11.2015 um 15:43 schrieb Philippe Mouawad: > >> > >> On Wed, Nov 11, 2015 at 3:35 PM, Felix Schumacher < > >> [email protected]> wrote: > >> > >>> Am 11.11.2015 um 14:56 schrieb Philippe Mouawad: > >>> > >>>> Hi, > >>>> Any feedback on this ? > >>>> Is there a technical reason except for SaveService why those $Revision > >>>> are > >>>> here ? > >>>> > >>>> I think for maintainbility they should be removed where they are > >>>> useless. > > They don't need maintenance. > > >>>> > >>> As long as we are on svn, they do no harm, do they? So I am +-0 on > >>> removing them. > >>> > >> Is it because you think regressions might be introduced ? In fact that's > >> why I want to remove them. > > > > I can't see a regression in both ways, that's why I am +-0 to this > change. > > > >> > >> > >> > >> > >>> For SaveService, it annoys me, that the tests fail, when run from a > >>> git-repo. I think, we could change the usage of $Revision to a sha1 sum > >>> (or > >>> sha256, for what it's worth), and compute the sum on reading the > >>> properties > >>> file. > >>> > >>> I agree. On our company repo, we replaced Revision id by a number and > >> > >> tests do not fails, so I think a SHA256 should work. > > > > See https://bz.apache.org/bugzilla/show_bug.cgi?id=58601 > > > > All tests complete correctly in my git repo. > > > > Regards, > > Felix > > > > > >> > >>> If no one objects, I will give that a try. > >>> > >>> Regards, > >>> Felix > >>> > >>> > >>>> Regards > >>>> Philippe > >>>> > >>>> On Sat, Oct 17, 2015 at 6:08 PM, Philippe Mouawad < > >>>> [email protected]> wrote: > >>>> > >>>> If they are useless why keep them ? > >>>>> > >>>>> We have coding style strategies that say that : > >>>>> - empty comments should be removed > >>>>> - dead code should be removed > >>>>> ... > >>>>> > >>>>> So if there are useless, they should be removed ? Even if they do not > >>>>> harm > >>>>> (as Empty comment do not hurt neither). > >>>>> > >>>>> Unless there is a good reason, for example related to the svn to git > >>>>> migration . > >>>>> > >>>>> Regards > >>>>> Philippe > >>>>> > >>>>> On Sat, Oct 17, 2015 at 5:47 PM, sebb <[email protected]> wrote: > >>>>> > >>>>> Why remove the marker? > >>>>>> > >>>>>> Is it doing any harm? > >>>>>> > >>>>>> On 17 October 2015 at 14:36, Philippe Mouawad > >>>>>> <[email protected]> wrote: > >>>>>> > >>>>>>> Hi, > >>>>>>> Any feedback on this ? > >>>>>>> Thanks > >>>>>>> > >>>>>>> On Sat, Oct 3, 2015 at 9:52 PM, Philippe Mouawad < > >>>>>>> > >>>>>> [email protected] > >>>>>> > >>>>>>> wrote: > >>>>>>>> > >>>>>>>> Hello, > >>>>>>>> Some classes contain this javadoc. > >>>>>>>> But a lot do not. > >>>>>>>> > >>>>>>>> Except for SaveService there does not seem to be any need of this. > >>>>>>>> > >>>>>>>> I propose to remove it from code. > >>>>>>>> > >>>>>>>> Regards > >>>>>>>> Philippe > >>>>>>>> > >>>>>>>> > >>>>>>>> > >>>>>>> -- > >>>>>>> Cordialement. > >>>>>>> Philippe Mouawad. > >>>>>>> > >>>>> -- > >>>>> Cordialement. > >>>>> Philippe Mouawad. > >>>>> > >>>>> > >>>>> > >>>>> > >> > > > -- Cordialement. Philippe Mouawad.
