Hi Enrico, sounds like you just have to compare the definitions for the type to find out what's different. Gerd demon.box wrote

Hi GerdP, no it isn't caused by different points in the map but I'm sure that is it just a matter of different type files (in the lines files I think)... I don't understand what cause it. In my map the footway concrete line looks more long than in the ExtremeCarver's map. Have you got any ideas? Thanks. --enrico

-- View this message in context: http://gis.19327.n5.nabble.com/Different-rendering-tp5803738p5803748.html Sent from the Mkgmap Development mailing list archive at Nabble.com.

Hi Enrico, I think you can't use r3211 with the same style rules as 2889. See http://www.mkgmap.org.uk/news/2013/12/22/more-flexibility-in-style-files-sty... Besides that you probably see the effect of the WrongAngleFixer. I can't say more without knowing the style, the options and the OSM data you are using. Gerd demon.box wrote

I make some test:

with mkgmap release 3211: <http://gis.19327.n5.nabble.com/file/n5803764/immagine1.jpg>

with an old mkgmap release 2889 and the same style rules: <http://gis.19327.n5.nabble.com/file/n5803764/Immagine3.jpg>

I don't understand....

-- View this message in context: http://gis.19327.n5.nabble.com/Different-rendering-tp5803738p5803765.html Sent from the Mkgmap Development mailing list archive at Nabble.com.

Hi Enrico, the WrongAnfgleFixer contains the code that tries to fix distorted lines. I still don't fully understand what you try to do. If you want help, please post all information that is needed to reproduce the result: 1) a link to the style and the typ file 2) the OSM data (the id of one of the ways in the screenshot should be enough) Gerd demon.box wrote

Yes, sure, I delete ONLY the finalize section from the style and the rest is the same but how can you see the result is very different.

This is my command:

start /low /b /wait java -enableassertions -Xmx1000m -jar ..\bin\mkgmap.jar --country-name="%country%" --family-id=%FID% --mapname=66%FID%001 --overview-mapname=66%FID%000 --series-name="OSM_%date%" --family-name="OSM_%date%" --style-file=..\bin\styles\ --style=%style% --name-tag-list=name:it,name --description="%country%" -c ..\bin\template_basic.args -c template.args %TypFileToAdd%

This is in the "options" file: levels = 0:24, 1:22, 2:20, 3:18, 4:16

What is the WrongAngleFixer? What can I do? Thanks! --enrico

-- View this message in context: http://gis.19327.n5.nabble.com/Different-rendering-tp5803738p5803771.html Sent from the Mkgmap Development mailing list archive at Nabble.com.

Hi Enrico, I think there are two problems. 1st problem is that you use a non-routable type for way 268146383. (You have a lot of rules in the lines file which assign road_class=x in combination with extended types. This is allowed, but the ways are not routable.) That means it is not connected to the routable ways. 2nd problem is that mkgmap optimizes this way as if it is not connected to any other way. I think that is not okay, I'll have a look at it when the refactoring branch is done. I've attached a modified lines file that solves problem 1. Gerd lines.lines <http://gis.19327.n5.nabble.com/file/n5803787/lines.lines> demon.box wrote

Ok, thanks for your kindness, here can you find my complete style: http://dropcanvas.com/7x3um <http://dropcanvas.com/7x3um>

and these are the 3 ways id:

268146389 268146383 268146390

Thanks a lot! --enrico

-- View this message in context: http://gis.19327.n5.nabble.com/Different-rendering-tp5803738p5803787.html Sent from the Mkgmap Development mailing list archive at Nabble.com.

Wow, with your modifications (I understand than the difference is the sequence of lines) the problem is solved as can you see below: <http://gis.19327.n5.nabble.com/file/n5803790/Immagine5.jpg> Thanks very much!! --enrico -- View this message in context: http://gis.19327.n5.nabble.com/Different-rendering-tp5803738p5803790.html Sent from the Mkgmap Development mailing list archive at Nabble.com.