Hi Christian & everyone,
> the last weeks we started several branches in the git repository and I was > wondering wether we could merge some of these new things into the master. > I noticed that these three branches (labeling-revised-revised, subimg_test and sparse_labeling_experiment) had been merged into master, so I deleted the remote branches to keep things tidy. To all ImgLib2 developers: I wrote a little script to analyze remote branches: list the last author, last commit date, and number of still-unmerged commits. Script is here: https://github.com/scijava/scijava-common/blob/master/bin/remote-branch-info.sh The result on imglib.git is below. If you see your name below, can you double-check the topic branch to confirm what should happen to it? Options include "delete it" (if it is already fully merged, or no longer relevant), or "leave it alone" (if still a work in progress). Is there any reason why the branches with "0 unmerged" cannot go away? Thanks, Curtis -- KnimeInterface - Barry DeZonia - 4 months ago - 0 unmerged barcelona-tobias - Tobias Pietzsch - 9 days ago - 68 unmerged bughunting - Tobias Pietzsch - 7 weeks ago - 1 unmerged catmaid - Stephan Saalfeld - 9 days ago - 65 unmerged compressedcontainer - Stephan Preibisch - 1 year, 3 months ago - 4 unmerged connectedcomponents - Larry Lindsey - 1 year, 10 months ago - 7 unmerged contrib - Jean-Yves Tinevez - 7 weeks ago - 0 unmerged converted_access - Tobias Pietzsch - 5 months ago - 0 unmerged converters - Stephan Saalfeld - 2 weeks ago - 40 unmerged edgeldetection - Tobias Pietzsch - 28 hours ago - 3 unmerged for-steffi - Stephan Preibisch - 3 months ago - 60 unmerged ij2-examples - Stephan Preibisch - 3 months ago - 35 unmerged imglib-hough - Larry Lindsey - 2 years, 2 months ago - 0 unmerged imglib2-dirty - Stephan Saalfeld - 1 year, 1 month ago - 0 unmerged imglib2-dirty-kdtree - Tobias Pietzsch - 1 year, 1 month ago - 14 unmerged imglib2-scripting - Stephan Preibisch - 5 months ago - 0 unmerged imglib2-views - Tobias Pietzsch - 1 year, 2 months ago - 0 unmerged imgs - Stephan Saalfeld - 30 hours ago - 3 unmerged kdtree - Tobias Pietzsch - 1 year ago - 0 unmerged knime-algorithms - Christian Dietz - 4 days ago - 3 unmerged knip_deployment - Christian Dietz - 5 days ago - 6 unmerged labeling-revised - Tobias Pietzsch - 7 weeks ago - 4 unmerged localneighborhoodcursor - Benjamin Schmid - 4 days ago - 3 unmerged mandelbox - Stephan Saalfeld - 9 days ago - 60 unmerged modified-abstract-hierarchy - Tobias Pietzsch - 7 weeks ago - 0 unmerged mser - Tobias Pietzsch - 5 months ago - 0 unmerged newinterpolation - Stephan Preibisch - 8 weeks ago - 1 unmerged ops - Stephan Preibisch - 9 months ago - 0 unmerged poster-bioimaging2 - Tobias Pietzsch - 8 months ago - 3 unmerged realtransform - Stephan Saalfeld - 9 weeks ago - 0 unmerged roi-dscho - Johannes Schindelin - 7 weeks ago - 5 unmerged roialgorithm-opt - Larry Lindsey - 1 year, 3 months ago - 16 unmerged serialization - Christian Dietz - 7 weeks ago - 1 unmerged shapeimages - Albert Cardona - 1 year, 6 months ago - 0 unmerged shapelists - Albert Cardona - 12 months ago - 5 unmerged tutorial - Tobias Pietzsch - 7 weeks ago - 12 unmerged utils - Larry Lindsey - 2 years ago - 11 unmerged On Fri, Apr 20, 2012 at 11:11 AM, Christian Dietz < [email protected]> wrote: > Hi all, > > the last weeks we started several branches in the git repository and I was > wondering wether we could merge some of these new things into the master. > > labeling-revised-revised: > Martin and me extended Lee's version of the Labeling to allow arbitrary > Img<? extends IntegerType<?>> as the index container. > Furthermore we changed the way Labelings are created, for e.g. a > NativeImgLabeling just needs the Img as an index container which makes the > usage of Labelings much easier. > Lee already agreed on these changes. > > subimg_test: > Thanks to Tobias we have an so called SubImg in KNIP now, directly using > the Views from imglib2. A SubImg is simply a subinterval of a Img acting as > an Img itself. > > sparse_labeling_experiment ("Octree" (N-Tree) representation of an image > in n-dimensions): > Thanks again to Tobias, we got a sparse Img container running in KNIP > which is really useful if you have e.g. big connected areas with the same > value or just a few smaller dots in a labeling or image. > Memory is reduced dramatically, but of course writing is slower than into > a e.g. ArrayImg. > > So if you have any comments on these branches, feel free. > For example the subimg_test could also just be part of KNIP if nobody > needs this. For us it's very helpful to iterate over certain subintervals > in images (e.g. PlaneSelection XYZ, iterate over CT), if there are > algorithms which explicitly require an Img as an input. > > Best regards and have a nice weekend, > > Martin & Christian > > > ______________________________**_________________ > ImageJ-devel mailing list > [email protected] > http://imagej.net/mailman/**listinfo/imagej-devel<http://imagej.net/mailman/listinfo/imagej-devel> >
_______________________________________________ ImageJ-devel mailing list [email protected] http://imagej.net/mailman/listinfo/imagej-devel
