On Tue, Feb 23, 2010 at 3:49 PM, Robert Osfield <[email protected]>wrote:
> Hi Philip, > > On Tue, Feb 23, 2010 at 12:19 PM, Philip Lowman <[email protected]> wrote: > > (submitter) >> $ svn diff -u > changes.diff >> $ # email changes.diff >> >> (reviewer) >> $ # download changes.diff >> $ cd clean_trunk >> $ cat changes.diff |patch -p0 >> $ meld . >> > > And... how many steps are there for the reviewer? > > Right now I just do: > > download changes. > mergeall ~/OpensceneGraph > You're assuming that the submission is actually worthy of being applied ;) With a context diff, it's often easy to see by inspection that the submitter is off in the weeds, or to quickly determine that the patch is worth further study. This is from the point of view of someone who might give feedback on a patch rather than check it into the mainline sources. You might consder the work saved if people were doing more of that before you took on the individual submissions yourself. > And then step through each file. mergeall is a script I have for just > stepping through subdirectories running xxdiff on the files it finds. I > don't have to maintain multiple checkouts just for the purpose of reviews. > I do what I do because it's the most efficient way for me to work. > > >> As for the occasional patch not applying without errors, I think most >> open-source projects deal with this just by asking the submitter to >> generate a new patch against the latest version of the source. Odds >> are if their patch is against out of date code you're probably not >> going to want their original source file anyways since you might >> accidentally revert an unrelated change you have made to the file >> since. >> > > This round trip takes time, and sometimes the reply never comes. I > continuously try to reduce the amount of time dedicated to communication, > the amount of downtime between. It's my experience that diffs don't help. > Whole files, even ones out of date can be far more useful than a broken and > useless patch file. > If the reply never comes, then you've saved yourself the trouble of applying the patch... > > Given how pressed I am for time, I really do care about waste minutes here > or there. Whole files wins hands down for me. > > Now git might handle things better, but unless we can do a graphics diff > against the trunk then we are still stuck with having to apply to branch > locally and then doing a graphics diff and merge. > > git format-patch, the command for submitting patches from local work, does include info that makes it more reliable to merge the patches against the current head of the maintainer regardless of the original versions used by the submitter. Tim
_______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
