Sometimes you don't want the user to run more than one instance of a program. But if the user associates a filetype with your programm it shall open the filetype in this first instance. The code below uses interprocess communication to find out if there is already another instance. It further allows the first instance to register an eventhandler which can be called from a second instance in order pass its command line parameters to the first instance.

This is how to use the code:

Yesterday I updated my system to Gnome 3.8. Unfortunately I was not able to boot it anymore. I would only see the boot manager but then It would just hang with a black screen. Switching to a console (Ctrl + Alt + F2) didn't work.

First of all you need to be able to work with your system. The easiest way to achieve that is starting it with runlevel 1.

Using SSH to copy files

Copy something from this system to some other system:

scp /path/to/local/file username@hostname:/path/to/remote/file

Copy something from some system to some other system:

scp username1@hostname1:/path/to/file username2@hostname2:/path/to/other/file

Copy something from another system to this system:

I worked on a wpf window containing some controls that were required and that validated the format of the entered data. Once the user pressed a button I wanted to check if there was any validation error and notify the user with an error message. This is how to do it:

Helper class:

I had an empty standard installation of MySQL but somehow I could not create a database using the InnoDB engine. The engine was just not activated (you can check that with the

Implementation

Uses the HeapPriorityQueue.
Code on Github

Implementation

Uses a heap for fast retrieval of the smallest element as well as a dictionary for a fast access to an element which is already in the data strucutre. This priority queue is not too bad for implementing a Dijkstra (a fibonacci heap would be perfect...).
Code on Github

Implementation

It is a simple implementation of a priority queue which is based on a SortedDictionary. This makes the implementation quite small.
Code on Github

Exporting the table using SQLWorkbench

Imagine you have a table images with a character column filename and a clob column image_data

First you need to install the SQL-Workbench. Install does just mean to extract the archive in that case.

The following command exports the table into a textfile. It does also generate the loader file which can be used with sqlldr. The CLOB data is exported as files with the name image_data_0, image_data_1, ... and will be in the same directory. Those files are referenced in the textfile.

Sometimes after you come home from holidays you realize that you forgot to adjust the time of your digital camera. So all pictures might have a wrong date and/or time in the exif header. Fortunately it is pretty easy to fix that with linux:

jhead -ta10:11 *

=> Adds 10 hours and 11 minutes to the exif datetime of all files (that have exif headers) in the current folder

jhead -ta-10:11 *

=> Subtracts 10 hours and 11 minutes from the exif datetime of all files (that have exif headers) in the current folder