This is what probably happens. The application designated to open files with the file types you mention, is the same across all of those. .gif, .jpg, .png: these are all handled by EOG (GNOME's Eye Of Gnome). I suspect this application does not determine the file type based on the extension, rather it seeks for the magic number.
Note that the behavior of EOG may very well be exceptional, and also beware that no actual conversion is done. It's just that Linux applications tend to rely on different detection methods than file extensions, though many modern desktops actually look for the file extension to determine what application to open, since this method is probably quicker.
This all may apply to Ubuntu, Windows is a completely different story. Explorer practices a behavior by default that cuts off the file extension from the filename. This way, a file named "document.pdf" will show as "document", and renaming it to "document.dvi" will actually be processed as "document.dvi.pdf". Viruses sometimes exploit this "feature" by making the user believe they are opening a .jpg, while in reality they are being tricked into running a malicious exe!
To be really sure about filetypes, there is a command called file that shows you detailed file information.
From my personal experience, I can tell that, at least on my system, EOG cannot be tricked into recognizing a renamed image file. You must be doing something different than I do.
Best Answer
I don't have much time on my hands right now but here's a quick overview*
Download Pyglossary from this link. Extract the archive and move the folder to a location of your choice. Make
pyglossary.pywexecutable and start it by double-clicking on the file.Alternatively you can download a prepackaged older version of pyglossary on the google code project page and install it with Ubuntu-Software Center or gdebi. Because this version is quite old (dating back to 2009) your mileage may vary.
Rename your wordlist files to whatever you would like your dictionary to be called (e.g.
DICTCC_EN_DE.txt)Point Pyglossary to your file and set the input and output formats as follows:
Click apply. WARNING: Depending on the length of the word list conversion might be very memory-intensive.
Import the dictionary to GoldenDict. This as well may use a lot of system memory (building the index, etc.)
Formatting should be acceptable (not perfect but usable). For further formatting tweaks (e.g. to remove annoying gender denominations) you would have to manually edit the input file.
*: please feel free to edit this post as you see fit to add more detail. I might do so myself later.