El 06/05/12 20:23, WanMil escribió:
El 04/05/12 19:33, Clinton Gladstone escribió:
On May 3, 2012, at 23:55, WanMil wrote:
I have already found a few tiles with flooding (e.g. NE of Russia). I have to investigate that. But I will be happy if you find some other things :-)
I've also been testing a little bit. So far the results have been pretty good.
One point I'm not sure on is how the --precomp-sea option should work together with the other coastline related options.
- I assume that the --coastlinefile is now redundant, correct?
- What about the various --generate-sea parameters?
I currently use the following when compiling:
--generate-sea=multipolygon,extend-sea-sectors,close-gaps=1000,floodblocker
Are most or all of these parameters now redundant?
If you use --precomp-sea you don't need --generate-sea at all, just omit it.
That's true. --precomp-sea uses multipolygon processing. The --generate-sea parameters are not used at all if --precomp-sea is set. Also the --coastlinefile is ignored too. I've been using the patched mkgmap for a while and for several countries and it worked fine. Perhaps it's time to commit the patch if nobody reports any problem with it.
Hi Carlos, I don't commit release 2 of this patch. I want to use PBF/OSM format for the precompiled files. I realized that there is no need for a proprietary format for the precompiled data. This has some advantages but I have to finish the implementation. Have fun! WanMil