December 2013 - mkgmap-dev - The mkgmap lists

Commit: r2857: Now really fix the wrong ref assignment in the relations file.
by svn commit 01 Dec '13

01 Dec '13

Version mkgmap-r2857 was committed by wanmil on Sun, 01 Dec 2013 Now really fix the wrong ref assignment in the relations file. The previous revert reverted one wrong line but also a good fix. Now it should be ok :-)

1 0

relations file difference
by Gerd Petermann 01 Dec '13

01 Dec '13

Hi WanMil, please check: I've just committed r2851, a merge from trunk in the mergeroads branch, but I am not sure about the change in the relations style file. Gerd

2 1

Commit: r2849: remove obsolete call to checkFixmeCoords
by svn commit 01 Dec '13

01 Dec '13

Version mkgmap-r2849 was committed by gerd on Sun, 01 Dec 2013 remove obsolete call to checkFixmeCoords

1 0

Commit: r2848: minor correction: evaluate the fixme tag of points before
by svn commit 01 Dec '13

01 Dec '13

Version mkgmap-r2848 was committed by gerd on Sun, 01 Dec 2013 minor correction: evaluate the fixme tag of points before using routines which replace coords and drop the info.

1 0

The road network and duplicate ways
by Gerd Petermann 01 Dec '13

01 Dec '13

Hi Steve, some styles assign create multiple routable lines for the same OSM way, e.g. to set different road-class or speed values for bike and car. This seems to work, but reg. performanc I see two points: 1) We do all kinds of calculations multiple times, e.g. all the bearing calculations and (Douglas-Peucker) filters. 2) We save the points for the same line multiple times in the img file. A possible alternative would be to have a class that stores the list of points and a list of assigned types, but that would involve a lot of changes. I wonder if the img format allows to store the points only once, means, can we store one NET entry and multiple NOD entries referring to the same NET entry? Another point that I don't fully understand: In RoadNetwork.addRoad() we always add the reverse arc for each arc. I wonder why oneway roads are not treated different here, esp. if they have foot=no? Gerd P.S. A while ago I tried to find out the meaning of the curve data encoded with two bytes. I failed because I was not able to change display tool to read my Garmin map NOD file. I saw too many "sync lost" messages. I guess the format is too different, e.g. NOD header length is 0x3f. I have created a local copy of NodConvert that is more up to date. Should I commit that?

1 0