With the dearest person I have ever met, I am finally starting to overcome my laziness, apathy and reluctance to do anything at all, and yesterday, after waking up for the second time, I started working on evaluating file name similarity for media files. Of course, the fact that I live in this Hell (a.k.a. “home”, or rather just ‘permanent residence’) plays a part in it too. So what exactly is it about? One of the first test runs will tell you more:
I simply download a video from YT, and this program tells me whether I already have it on my disk, for example even as a different remix or something like that. It can also do plenty of other things; this is just the feature I worked on today, and that is what this post is about. In the folder I was searching for duplicates I had these files:
As you can see, some that should be there are missing.
Then I tried it on the “real” files, and this was the result:
Which is great, this is exactly what it was supposed to find. Finally, I tinkered with the method that evaluates similarity and shuffled some more files around so that it would recognize them:
As you can see, the only ones still missing are Hradistan Skupina-Modlitba Za and Hradistan-Modlitba ZaVodu. I am not resting on my laurels with these, though, and at some point in the future I would like to improve this recognizer further. For now, though, it is fine as it is, because it is enough for my needs. O:-)
And finally, some code from this (specifically just the Song class, but the evaluation method is in there): http://pastebin.com/YjPV7ipv