Debug Java in Bio7

Leave a reply

27.06.2018

In Bio7 R and Java code can be easily combined in the Bio7 platform. For instance to create Graphical User Interfaces in Java (with SWT or JavaFX), transfer image pixel/selection data from ImageJ and finally doing the analysis parts in R code/scripts which can be called from within Java (an easy to use API on top of Rserve in Bio7 is available, see, e.g., Groovy examples here).

However sometimes it is necessary to debug the Java language as part of such combined work. In Bio7 (based on Eclipse RCP) the very powerful Java debugging tools from Eclipse are available by default.

In addition in Bio7 it is possible to debug dynamically compiled Java classes in the same Bio7 process (using Java like a scripting language is a special feature of Bio7).

The video at the bottom of this page demonstrates the use of a remote debugging connection on the same computer (localhost connection) to debug dynamically compiled/executed Java code.

Debugging a Java process which has already started (Bio7 in this case) is only possible if Bio7 is started beforehand with the following Java arguments inside a shell to start a debug server connection:

Windows (started from the current directory = Bio7 installation directory):
Bio7.exe -vmargs -Xdebug -Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=8000

Mac (given the full path to the Bio7 executable):
/Applications/Bio7.app/Contents/MacOS/Bio7 -vmargs -Xdebug -Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=8000

Linux (started from the current directory = Bio7 installation directory):
./Bio7 -vmargs -Xdebug -Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=8000

Furthermore the compiler debug information option has to be enabled in the Bio7 Java preferences (for the dynamic compilation process) and a seperate debug connection has to be started for this Java class (with the available Eclipse debug configuration dialog).

In the video below I (hopefully) demonstrate how you can (graphically) debug a simple cellular automata simulation like the Game of Life within a running Bio7 (Java) process.

Please activate JavaScript to view this video.
Video-Link: https://www.youtube.com/watch?v=vKAGBGuQKlI

Bio7 2.8 Released

Leave a reply

30.05.2018

A new and improved release of Bio7 is available. The new Bio7 2.8 release comes with a plethora of new R features and bugfixes.

Release Notes:

General:

  • Updated Eclipse RCP to 4.7.3
  • Windows R update to version 3.5.0
  • ImageJ plugin updated to version 1.52d7
  • Added the ImageJ Edit perspective to the default opened perspectives
  • The R perspective is now selected as the opened perspective after startup of Bio7
  • Added an option to save all editors on close (default enabled on MacOSX)
  • Improved the main ‘Scripts’ menu to allow nested custom scripts in folders and subfolders which form new submenus
  • Improved the Table and R-Shell context scripts menu
  • Added some example scripts to the new scripts menu
  • Added an script to open the default script location in the OS file explorer
  • Simplified the different script locations (reduced to four locations in the Bio7 preferences)
  • Spatial import, export scripts moved to the main ‘Scripts’ and R-Shell script context menu (Deleted import, export script locations)

R:

  • Simplified the Bio7 R-Shell interface
  • Added a new view for some R templates
  • Code completion from loaded R packages can now be imported from the R-Shell or the R editor (both completions will be updated!)
  • Improved the reload code completion by only considering R functions (cleaner seperation of functions and datasets)
  • The updated R-Shell context menu can now dynamically be extended with custom scripts (updated dynamically). Folder and subfolders form new submenus.
  • Improved, bugfixed and extended the ‚Evaluate Line‘ action which was renamed to ‚Evaluate Line or Selection‘  action which now interprets selected R code, too.
  • Improved MacOSX ImageJ interaction (editor hoover focus)
  • Improved the call of the default plot device (quartz) in a native R connection on MacOSX (now default opened when, e.g., plotting, debugging, etc.)
  • Added a new menu to install Rserve for R > 3.5.0
  • New shutdown option to execute R code at shutdown (e.g., to save the current R workspace – a reload with the already existing startup option is possible, too!)
  • Improved the general error messages of the R-Shell view and the R editor evaluation. Now much simpler and printed cleaner!
  • Added an option to print the error stream of the Bio7 console to the R-Shell view (to the right output panel to get an output if the console is hidden!)
  • Fixed a bug with the shiny execution (removed annoying dialog)
  • Fixed many other bugs (unnecessary error messages, execution errors, MacOSX bugs, etc.)
  • R-Shell code completion improved (see editor code completion improvements below)
  • Removed the use of temporary files of the R-Shell and R editor code completion
  • Added an preference option to add arguments to the source command
  • Multiple selected R-Shell variables are now available in the R workspace (‘.r_shell_vars’ – a character vector) for the use in custom R scripts (e.g., to extend the R-Shell script context menu!)

R Editor:

  • Added a detection/quickfix function to install missing packages when a package load definition is in the editor file but the package is not installed

  • Added a quickfix to add a library definition when an unknown function is called (not defined and loaded but package with function is installed)
  • Added support for the the magrittr piping operator in code completion to autocomplete a given dataframe in the workspace (column completion)

  • Added code completion support for dataframes columns and rows, named list elements and named arrays (of arbitrary dimension)

  • Improved code completion to display current workspace variables
  • Classes of R objects are now displayed in the code completion context dialog
  • Editor variables and functions are now emphazised in code completion, too
  • Added a directory dialog to easily create a directory string for the editor file
  • Created new submenus for the organization of the context menu
  • Improved the hoover dialog (improved offset)
  • Improved the completion context of R objects (current workspace objects)
  • The styler package can now be used for an improved code formatting (requires UTF-8 Bio7 workspace on Windows – see preferences!)
  • Improved code completion for S3 and S4 objects (correctly display current cursor completion)

ImageJ:

  • Updated ImageJ plugin to version 1.52d7
  • Updated ImageJ2 libraries
  • Added new ImageJ macro functions to the code completion of the macro editor
  • Added the ImageJ Edit perspective to the default openend perspectives
  • Improved the editor hoover activation on MacOSX
  • Improved the ImageJ MacOSX touchpad resizing of the ImageJ panel

Java:

  • Updated JRE to 1.8.71
  • Improved the creation of Java projects and project imports (no annoying dialog anymore)
  • Improved the ‘Add Selected Libraries’ preference to add multiple Java libraries for dynamic compilation (e.g., to compile more embedded Eclipse library functions)
  • Now ImageJ2 libs have to be added to the dynamic compilation libraries for compilation (removed as default libraries)
  • Added a key shortcut (STRG + ‚+‘) to increase the SceneBuilder panel (2x)

Download and Installation:

Windows:

Just download the *.zip distribution file from https://bio7.org and unzip it in your preferred location. Bio7 comes bundled with a Java Runtime Environment, R and Rserve distribution and works out of the box.

Linux:

Download and extract the installation file from https://bio7.org.
For Linux you have to install R and Rserve (see Rserve installation below!).

MacOSX:

Download and extract the installation file from https://bio7.org.

If you start Bio7 a warning or error can occur because of the changes how Apple treats signatures! To allow Bio7 to start see this instructions for Yosemite and Sierra:

OS X Yosemite: Open an app from an unidentified developer

macOS Sierra: Open an app from an unidentified developer

If you have still problems with Sierra see this solution!

In addition for MacOSX you have to install R and Rserve (see below!).

Linux and MacOSX Rserve (compiled for cooperative mode) installation:

To install Rserve open the R shell and then execute the menu action “Options -> Install Rserve (coop. mode) for R …” for different R versions. This will download an install Rserve in your default R library location, see video below (please make sure that your default Linux R library install location has writing permissions!). In cooperative mode only one connection at a time is allowed (which we want for this Desktop appl.) and all subsequent connections share the same namespace (default on Windows)!

Please activate JavaScript to view this video.
Video-Link: https://youtu.be/tF7HbRBRIFQ

Installation of Useful R Packages

R packages which are useful in combination with Bio7 can easily be installed with the R-Shell context menu “Packages” action:

R-Shell context menu->Packages->Install_Default_R_Packages

Bio7 Documentation

For more information about Bio7 please consult the soon updated Bio7 User Guide.

A plethora of Bio7 videotutorials can be found on YouTube.

 

R Improvements for Bio7 2.8

Leave a reply

18.05.2018

The next release of Bio7 adds a lot of new R features and improvements. One minor change is that the default perspective after the startup of Bio7 now is the R perspective to emphazise the importance of R within this software.

The R-Shell view has been simplified and the R templates have been moved in it’s own simple view for an improved usability (see screenshot from R perspective below).

In addition the context menu has been enhanced to allow the creation of submenus from scripts found in folders and subfolders (recursively added) which you can create for a menu structure.
Scripts can be added created in R, JavaScript, Groovy, Jython, BeanShell, ImageJ Macros.
Java (with dependant classes) can be dynamically compiled and executed like a script, too.

Several improvements have also been added to the R-Shell and the R editor for an easier generation of valid R code. The R-Shell and the R editor now display R workspace objects with it’s class and structure in the code completion dialog (marked with a new workspace icon – see below).

R-Shell:

R editor:

In the R editor a new quick fix function has been added to detect and install missing packages (from scanned default packages folder of an R installation – has to be enabled in the Bio7 R code analysis preferences).

Also the detection of missing package imports are fixable (when a function is called but the installed package declaration is missing in the code but the package is installed to deliver the function).

The code assistance in the R-Shell and in the R editor now offers completions for, e.g., dataframes (columns) in the %>% operator of piped function calls:

In addition code assistance is available for list, vectors, dataframes and arrays of named rows and columns, etc., when available in the current R environment.

Code completion for package functions can now easily added with the R-Shell or the R editor which loads the package function help for both interfaces. The editor will automatically be updated (see updated editor marking unknown functions in screencast below).

Numerous other features, improvements and bugfixes have been added, too.

Bio7 2.8 will hopefully be available soon at:

https://bio7.org

Overview videos on YouTube

 

 

Bio7 2.7 Bugfix Release for MacOSX

Leave a reply

26.02.2018

Last week I released a new bugfix version of Bio7 which fixes some annoying bugs on MacOSX and updates some plugins.

  • Updated ImageJ to version 1.51v1
  • Improved the import of Java projects (No annoying dialog anymore)
  • Improved the execution of Shiny apps (No dialog when stopping the server)
  • Fixed several bugs and possible deadlocks regarding the swt_awt bridge and the OpenGL panel
  • Fixed a swt_awt key listener bug (ImageJ panel) to restore key events on the panel
  • Fixed a Mac bug in the R editor color dialog (hoover dialog of name R colors)
  • Huge fontsize for Swing (e.g., ImageJ dialogs) can be corrected for Retina displays (Bio7 ImageJ preferences – ‘-6’ for Retina displays fits well)
  • Bio7 editors and the Bio7 Workspace will be automatically saved before application shutdown

 

Download here

Bio7 2.7 Retina-Display Fix for MacOSX

2 Replies

08.01.2018

I recently discovered on a high dpi display (Retina-Display) enabled 13’’ notebook that the latest release of Bio7 for MacOSX is not displayed in high dpi but in a lower resolution which makes text and icons blurry (scaled) and unnecessary strains the eyes when editing R scripts with the R editor.

So I uploaded a new, fixed version for MacOSX.

However, to fix an already downloaded version of Bio7 you have to open open the ‘/Bio7.app/Contents/Info.plist’ file with a text editor (right-click on Bio7.app, execute ‘Show Package Contents’ and open ‘/Contents/Info.plist’) and add the following lines before the </dict> tag:

<key>NSHighResolutionCapable</key>
<true/>

In addition to update the Info.plist file for the Operating System you have to open the Terminal (in ‘Applications/Utilities’) and execute the following command (default install directory – shell command is one line!):

/System/Library/Frameworks/CoreServices.framework/Frameworks/LaunchServices.framework/Support/lsregister -v -f /Applications/Bio7.app

Here some screenshots from before and after the fix was applied on a Retina-Display (zoom in for details):

Before:

After:

An updated version with the fix already applied can be downloaded here

See also:

Release notes for Bio7 2.7