Ant Build Successful - IzPack Installer Under Construction

I was able to fix the internationalization problem I was having with the OpenJUMP JAR files created by my new Ant build file, after I received some help from Paolo Rizzi on the JUMP Pilot Project developer mailing list. I’ve now got a (mostly) functioning Ant build file for OpenJUMP. I tested out the JAR files produced with the new Ant build file with my launch script on Linux and everything appears to be working just fine. I hope to distribute new Ant build files for use by other OpenJUMP developers on the SurveyOS site by the end of the week.

I also have succeeded in setting up the IzPack installer program on my Linux box, and was able to create a functioning installer for OpenJUMP. I still have a few more days of work left on the installer. I need to write the release notes, get the generation of shortcuts for both Windows and Linux working properly, and also work on a custom panel that will generate a configured launch script for OpenJUMP on Linux.

The bug tracker on SourceForge has a number of different bugs that have been reported. I selected 3 bugs from the list that I would like to tackle and squish before this next release.

I’m still toying with the idea of adding OpenJUMP’s classpath to the manifest file, and haven’t decided if I want to tackle a Debian package for this release of OpenJUMP.

At any rate, I am making progress on the 1.02 release of OpenJUMP. I’ll update the readers of my blog when I have more news.

The Sunburned Surveyor

Read More

Update On OpenJUMP Release 1.02

I was able to make some progress last week on the 1.02 release of OpenJUMP. I've modified the Ant build file used for the nightly build of OpenJUMP. It's quite a bit simpler now, and doesn't mess with the stuff for the CVS repository at the JPP. I imagine this build file will be a lot more useful for individual developers that want to modify, build, and test OpenJUMP.

I hope to release 3 versions of this Ant build file in the next week or two.

The first version will simply build OpenJUMP without any Javadoc target or distribution targets. This build file will only produce the 2 main OpenJUMP JAR files. This will be the build file will be the quickest and the one most commonly used by developers that want to repeatedly tweak and test OpenJUMP.

The second version of the build file will include the Javadoc target, and will bundle all of the necessary files to create an OpenJUMP “distro”. This means all of the available documentation and the necessary library files.

The third version of the build file will also prepare an Izpack installer for OpenJUMP.

I've finished the first 2 versions on the Ant build file described above, but I'm having some trouble with some of the classes that support internationalization. (That's what I think based on the error message I get when I try to run OpenJUMP from the files produced in the Ant build.) I'm going to try and get some help for this from the other developers on the JPP mailing list. Next week, after this is fixed, I'll be working on the Izpack installer for the 1.02 release. It will be based on the work Erwan and the SIGLE team has already done on the installer for their brand of JUMP.

When I have successfully built the installer, I'll get that incorporated into the third and final version of the Ant build file for OpenJUMP.

Then it will be time to squish a couple of OpenJUMP bugs. We may see a release by the end of February, but don't quote me on that date. I've been wrong on the date too many times already. :]

The Sunburned Surveyor

Read More

Update On The "Mapping Math" Chapter Of The Free GIS Book

I wasn't able to make any additions to my chapter in the Free GIS Book last weekend, but I did take some time to do so this morning. I've started on the part of the section on measurements that deals with distance measurements specifically.

You can read this new material on the OSGeo wiki here:
http://wiki.osgeo.org/index.php/Math_For_Measurements:_Measuring_Distances

As I wrote this part of the chapter today I found myself struggling to define some abstract geometric concepts like “space” and a “point”. It is more difficult than I had thought it would be. There is so much knowledge that I just take for granted until I need to explain something in detail! Many of the things I hope to teach are interrelated, and it can be difficult to determine what should be taught first. I think I will be introducing some topics and returning to discuss them in more detail later in the chapter, when it is more appropriate.

At any rate, today I did take a stab at the definition of a “point” and even tried to describe “space”. (How do you explain that a point is a single location in space without telling the reader what you mean by space?) Hopefully I can explain what the geometric definition of a “path” is next weekend. Then I'll tackle the simplest definition of a distance, which is the length of a segment of a straight path, or a straight line segment. Then I may talk a little about distances that are slightly more complex, like the distance along a circular arc or other complicated curve. However, the distances that can be represented by straight line segments are the most important for this section of the chapter, because those are the distances we most typically measure when we want to locate features on a map.

I welcome comments and suggestions on this new material. Perhaps some of my more experienced readers will have ways I can improve the definitions I've written so far.
I'm also finding that it will be a challenge to stay focused in this chapter.

In an effort not to get really sidetracked and to avoid cluttering the main body of the chapter I will be adding extra information sections when I see fit. These sections will be brief discussion on a topic that is slightly related to the topic under discussion. The reader can take a look these extra information sections for a deeper understanding of the material that isn't absolutely necessary for the reader in more of a hurry. :] I added a link to the first of these extra information sections today, which isn't yet written. It will be entitled “Modeling with Maps”.

In another couple of weeks I actually hope to write some simple word problems involving the measurement of distances, so my readers can actually “do” some math.

The Sunburned Surveyor

Read More