Larry, Thanks for the response.
You wrote: "I think bug fixing should be done in the trunk so that you get feedback as soon as possible." I only thought of branching for two (2) reasons: (1) So I don't break the nightly build with my changes. (2) So I can produce a regular build of the branch while giving the other core programmers a chance to evaulate each change. (3) I'm killing some of my own JUMP efforts to focus on bug fixing in OpenJUMP. This means this will be a long-term campaign instead of just a couple patches. (This is my goal anyways.) It seemed that this type of long-term quality control effort might be better suited for a branch. However, if you feel strongly that the changes should occur in the core, I can do things there. Larry wrote: "What kind of unit tests - Junit? Which version? Sounds reasonable in general, although it may be a case of "closing the stable door after the horse has run away." "Tilting at windmills" and all that... :-)" I've developed a small library called SUTIJ (Simple Unit Testing In Java). I wrote the code after some frustration with trying to do some simple things with JUnit. SUTIJ doesn't use refelction, annotations, inner classes, or other Java black magic. It is just plain old Java code. It also makes it really easy to implement custom testing behavior, such as defining the order in which test methods are executed, setting up tailored test suites to excercise only specific aspects of a code base, and other things like that. I'm using it to test another project. Everything is run without a GUI and test results (pass or fail) are written to a test file. I don't want to bore you with the details, so here is a link to the SUTIJ source code: http://surveyos.svn.sourceforge.net/viewvc/surveyos/java/sutij/trunk/src/ I may be chasing horses out of the stable, but I think having a comprehensive suite of unit tests may make future refactoring easier. Larry wrote: "You may know that there are some Junit tests from the original JUMP already for:" I did not know that. Thanks for posting the list. I will reuse this code if I can. Larry wrote: "Can you be more specific about utility code consolidation?" Sure. Let me give you one example of the type of consolidation I'm talking about: - Moving the findFilesRecursively method from the PlugInManager class to the FileUtil class. I can post more examples if you want. The Sunburned Surveyor On Thu, Apr 8, 2010 at 2:38 PM, Stefan Steiniger <sst...@geo.uzh.ch> wrote: > Hei Landon, > > Yep - no problem to create a new branch. You should be able to do that. > However, I tried branching once lately, and messed-up my local system, > i.e. I wasn't able to re-integrate (merge) my changes to the branch back > without copy pasting to another folder and then do a complete new check-out. > > hope you have more luck > stefan > > Sunburned Surveyor wrote: >> Michael, >> >> You wrote: "I forgot bug (1) reported by Nacho as it has never been >> reported in the >> bug tracker. I can have a look this week end as I think I wrote the >> guilty plugin, >> and Nacho kindly supplied the data to reproduce the problem." >> >> Thanks for offering to help with Nacho's bug. I think the first step >> would be modifying the code to report the FID of the problem feature >> in the exception message. At least, that is what I was going to try. >> You might have a short cut to the problem. >> >> You wrote: "There is another bug for which I spent some time without success >> : >> 2792806 Error while deleting heterogeneous item >> If you or someone else wants to have a look, it needs a good >> understanding of the item selection process (which is quite complex in >> OpenJUMP due to partial feature slection capability)" >> >> I've done a little work with the SelectionManager in my super select >> plug-in. I'll add this to my list of things to fix if I can. I'm just >> waiting for Stefan's permission to get a branch to work in. I'll copy >> him direction on this e-mail since I know he sometimes travels and >> looses track of the messages on this list. >> >> Thanks, >> >> The Sunburned Surveyor >> >> >> 2010/4/8 Michaël Michaud <michael.mich...@free.fr>: >>> Hi Sunburned Surveyor >>> >>> Fixing these bugs would be nice and starting unit tests would be nice too. >>> >>> I forgot bug (1) reported by Nacho as it has never been reported in the >>> bug tracker. >>> I can have a look this week end as I think I wrote the guilty plugin, >>> and Nacho kindly supplied the data to reproduce the problem. >>> >>> There is another bug for which I spent some time without success : >>> 2792806 Error while deleting heterogeneous item >>> If you or someone else wants to have a look, it needs a good >>> understanding of the item selection process (which is quite complex in >>> OpenJUMP due to partial feature slection capability) >>> >>> Michaël >>> >>> >>> >>> Sunburned Surveyor a écrit : >>>> I'd like to request permission to create a branch in the JPP SVN where >>>> I can start work on two tasks: >>>> >>>> - A series of bug fixes and minor improvements. These fixes include >>>> (1) reporting the FID in the Union By Attribute plug-in to address a >>>> problem reported some time ago by Nacho; (2) making the cut polygon >>>> tool obey the "editable" status of the layer that contains the polygon >>>> being cut; and (3) consolidation of some utility code. >>>> >>>> - Creation of a set of units tests to start testing the OpenJUMP core. >>>> I'll probably start with classes in the >>>> com.vividsolutions.jump.workbench.model packages, since those are the >>>> simplest. >>>> >>>> I would like to move these improvements to the trunk of the SVN, one >>>> at a time, after getting approval from the other programmers on this >>>> mailing list. I won't be proposing any major changes to the core, and >>>> will make sure I get buy off from everyone before commiting to the >>>> trunk. >>>> >>>> With Stefan's permission I'd like to start packaging monthly builds of >>>> this branch in between his major OpenJUMP releases. >>>> >>>> Any comments or strong objections? >>>> >>>> The Sunburned Surveyor >>>> >>>> ------------------------------------------------------------------------------ >>>> Download Intel® Parallel Studio Eval >>>> Try the new software tools for yourself. Speed compiling, find bugs >>>> proactively, and fine-tune applications for parallel performance. >>>> See why Intel Parallel Studio got high marks during beta. >>>> http://p.sf.net/sfu/intel-sw-dev >>>> _______________________________________________ >>>> Jump-pilot-devel mailing list >>>> Jump-pilot-devel@lists.sourceforge.net >>>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel >>>> >>>> >>>> >>> >>> ------------------------------------------------------------------------------ >>> Download Intel® Parallel Studio Eval >>> Try the new software tools for yourself. Speed compiling, find bugs >>> proactively, and fine-tune applications for parallel performance. >>> See why Intel Parallel Studio got high marks during beta. >>> http://p.sf.net/sfu/intel-sw-dev >>> _______________________________________________ >>> Jump-pilot-devel mailing list >>> Jump-pilot-devel@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel >>> >> >> ------------------------------------------------------------------------------ >> Download Intel® Parallel Studio Eval >> Try the new software tools for yourself. Speed compiling, find bugs >> proactively, and fine-tune applications for parallel performance. >> See why Intel Parallel Studio got high marks during beta. >> http://p.sf.net/sfu/intel-sw-dev >> _______________________________________________ >> Jump-pilot-devel mailing list >> Jump-pilot-devel@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel >> >> > > ------------------------------------------------------------------------------ > Download Intel® Parallel Studio Eval > Try the new software tools for yourself. Speed compiling, find bugs > proactively, and fine-tune applications for parallel performance. > See why Intel Parallel Studio got high marks during beta. > http://p.sf.net/sfu/intel-sw-dev > _______________________________________________ > Jump-pilot-devel mailing list > Jump-pilot-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel > ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Jump-pilot-devel mailing list Jump-pilot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
