Tiled output. And a bugfix.

[wziard]

Hi, I didn't await your reaction to my issue #62 and implemented it anyway. It is quite a big commit so you'll probably have some reservations about applying it. However, the speedup in my tile generation is a factor 100, so I still created the pull request. I'm currently very busy at work so it's a bit of a 'take it or leave it' situation as I don't have the time to do extensive reworking of the patch, sorry.

The other commit is a workaround for when the player label (--drawplayers) or the origin circle (--draworigin) crosses the image boundary, which can happen if you specify --geometry. It really should be two separate pull requests, but I have no idea how to get github to do that from a single branch.

[wziard]

Hm, apparently the libgd Travis uses is older? I'll look into it.

[sfan5]

IMO generating tiling pyramids is better left to an external utility in your programming language of choice.
(which is Python, if you were to ask me)

[wziard]

That's what I did first. But it is much easier from within minetestmapper because there I already have the needed info available:

• how many tiles there are in each direction.
• which tiles are actually available.
• To generate leaflet compatible tiles I need to save them with the correct zoomlevel in the filename. I could of course rename the tiles afterwards with some other program, but this was easier.

I agree that it's not very hard to autodetect everything from the filenames and rename stuff, but still it was a lot easier from within minetestmapper.

My pull request was created only for the --tiled option though. I didn't know github auto-updated pull requests to always point tot the tip of a branch. I can put the --tiled option on a separate branch if you'd like.

[wziard]

But even that code needs some polishing first :-)