I created a quick service release to incorporate the changes since the last release.
This includes some minor adjustments in order to provide a more seamless user experience (sorry for the bullshit, but I just couldn’t withstand it 🙂 ).
Well, the new release fixes a nullpointer exception, and adds an option to save the current graph.
There are further tweaking regarding usability: the default layout is changed to Tree Layout in order to allow faster responses, and it is possible to hide the Local context node. These changes are the beginning of some serious overhaul, our goal is to increase the usability of the plugin.
Today I created a new release from Debug Visualisation.
The new release is called 0.7.0, as it introduces major new features:
- Preliminary support for Filtering nodes: similar to the Logical Structures feature it is possible to define the structure, and only visualise that (e.g. in case of Lists only the content should be depicted, not the other attributes).
- Drill in: it is possible to define a new top-level element instead of the Local Context, and hide other items.
There are also some minor updates, such as the reintroduction of the Simulated cooling layout algorithm.
On the other hand most changes are not visible in the user interface, but the revamped code should allow us to create some new features in the next service releases.
We recommend the update for everybody. Some technical information:
- After some fighting with the update site project the category information is lost. I have no idea of the cause.
- There is a new feature present in the update site: the Java Filter project defines filters for some items in the collections API.
I’ve commited the Debug Visualization project a long ago, but the previous few days I’ve had some spare time to kill on the project. The work mostly focused on the input package, which I didn’t liked at all when I wrote it in the first time. I just didn’t have better idea. But i have now and rewrote the whole part.
I’m satisfied with the currently used approach (for now), as I’ve managed to separate different functionalities, and connect them in a simple way. Also, it is reasonably faster than before, but it still makes the eclipse lagging when the user steps the debug context, which I don’t know why (yet). Anyway, navigating in the graph, hiding/showing nodes and other display-related functions are now fast enough.
For a more detailed explanation of the new operation, see the wiki page InputPackage.
ps.: for a change as big as this, I’m thinking of a maintenance step in the version numbering (0.7.0 instead of 0.6.2), what do you think? I’m hesitating because of the lack of new features.
Version 0.6.0 introduced some serious regressions from the previous versions, so we decided to create a quick service release.
This release contains fixes for Stack Overflow and Hidden nodes issues, and added some minor new features as the ability to zoom the graph.
You can use P2 to automatically update your installation, or install the new version from the http://eclipse.cubussapiens.hu update site.
We are hoping, this release will be usable than the previous one.
It was a while ago I created this introductory screencast (it used a modified version of 0.5.1 (perhaps it can be considered as a never-released first alpha of 0.5.2), so it is not totally correct but it helps a lot to understand how this can be used.
If not, please don’t hesitate to inform us how to correct it.