I have added some javadoc. Could someone please commit or deny the patch. It is open for a long time, no one has complained about it and I think it's a useful enhancement. WanMil
Could someone please commit or deny the patch. It is open for a long time, no one has complained about it and I think it's a useful enhancement.
Sorry, I don't see the use of a black/whitelist of tags. The information of used tags should be contained in the style files, so why should I have to manage another additional file? Is it for speed/resource reasons? Regards, Johann
Am 04.01.2010 11:18, schrieb Johann Gail:
Could someone please commit or deny the patch. It is open for a long time, no one has complained about it and I think it's a useful enhancement.
Sorry, I don't see the use of a black/whitelist of tags. The information of used tags should be contained in the style files, so why should I have to manage another additional file? Is it for speed/resource reasons?
Regards, Johann _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Johann, yes it speeds up the map generation by magnitude of times in case the map uses only very few parameters. e.g. if you want to create a map that visualizes the maxspeed (or maxheight, maxweight....) tag only its useless to parse all other parameters. You are right that this information is also contained in the style files and in future a better patch should automatically compile this information from the style files. WanMil
Could someone please commit or deny the patch. It is open for a long time, no one has complained about it and I think it's a useful enhancement.
Sorry, I don't see the use of a black/whitelist of tags. The information of used tags should be contained in the style files, so why should I have to manage another additional file? Is it for speed/resource reasons?
Regards, Johann _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Johann,
yes it speeds up the map generation by magnitude of times in case the map uses only very few parameters. e.g. if you want to create a map that visualizes the maxspeed (or maxheight, maxweight....) tag only its useless to parse all other parameters.
You are right that this information is also contained in the style files and in future a better patch should automatically compile this information from the style files.
WanMil
I'm not really inside the style code, so I don't completely understand the background of this. But in my understanding the osm data has to be parsed from beginning to end. Before I can decide to dismiss an parameter, I have to read it in and parse it, haven't I ? If there is no action defined in the style file, no style rule is active and nothing else should be happening. So why handle it with extra code? In my personal opinion I don't like another (superfluous) file, especially the blacklist file. And the whitelist *IS* the style file. No entrys here means no action! But as I said, style file is not really my part. And i believe you, that it speeds up the process. Regards, Johann
On 06.01.2010 00:26, Johann Gail wrote:
Could someone please commit or deny the patch. It is open for a long time, no one has complained about it and I think it's a useful enhancement.
Sorry, I don't see the use of a black/whitelist of tags. The information of used tags should be contained in the style files, so why should I have to manage another additional file? Is it for speed/resource reasons?
Regards, Johann _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Johann,
yes it speeds up the map generation by magnitude of times in case the map uses only very few parameters. e.g. if you want to create a map that visualizes the maxspeed (or maxheight, maxweight....) tag only its useless to parse all other parameters.
You are right that this information is also contained in the style files and in future a better patch should automatically compile this information from the style files.
WanMil
I'm not really inside the style code, so I don't completely understand the background of this.
But in my understanding the osm data has to be parsed from beginning to end. Before I can decide to dismiss an parameter, I have to read it in and parse it, haven't I ? If there is no action defined in the style file, no style rule is active and nothing else should be happening. So why handle it with extra code?
In my personal opinion I don't like another (superfluous) file, especially the blacklist file. And the whitelist *IS* the style file. No entrys here means no action!
But as I said, style file is not really my part. And i believe you, that it speeds up the process.
Regards, Johann
The blacklist is for example needed if you want to get rid of all motorcar=no tags, because you want to use motorcar/motorcycle mode for bicycle routing. Otherwise these tags will be be enacted. This is not easily doable inside the style-file (as action rules are limited on one action rule per way id, deleting these tags via action rule will be very very complicated). The reason is, that there is a lot more about the maps, that is not influencable via the style-file. I don't need a white-list, but I am sure it will make great speed improvments if you want to layer your map (e.g. sea layer). Automatically generating a white-list would of course be great if speed improvments can be made - however the whitelist would have to be (in any case or intelligently based on things like if there is a line with road_speed road_class) appended to access tags. Actually would one have to add name=* to the whitelist in order to have names??? Or ref=*? --- or are some keys included by default? Here is my blacklist that I'm currently using (well still using delete tags option): motorcar=no motorcar=private motorcycle=no motorcycle=private hgv=no hgv=private psv=no psv=private taxi=no taxi=private emergency=no emergency=private delivery=no delivery=private goods=no goods=private
_______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Felix Hartmann schrieb:
The blacklist is for example needed if you want to get rid of all motorcar=no tags, because you want to use motorcar/motorcycle mode for bicycle routing. Otherwise these tags will be be enacted. This is not easily doable inside the style-file (as action rules are limited on one action rule per way id, deleting these tags via action rule will be very very complicated).
Yes, but this is a bug in my opinion, because in the style-branch it works better. Chris
Chris-Hein Lunkhusen schrieb am 06.01.2010 10:33:
Yes, but this is a bug in my opinion, because in the style-branch it works better.
Yes, the rule handling in the style-branch works much better than in the trunk. I really would like to see the style branch merged into the trunk, since all new, not style related features are only applied to the trunk. So the style branch gets more and more outdated. Is there anything known to be broken in the the style-branch compared to the trunk, so that it can not/should not be merged? Gruss Torsten
On 06.01.2010 15:56, Torsten Leistikow wrote:
Chris-Hein Lunkhusen schrieb am 06.01.2010 10:33:
Yes, but this is a bug in my opinion, because in the style-branch it works better.
Yes, the rule handling in the style-branch works much better than in the trunk. I really would like to see the style branch merged into the trunk, since all new, not style related features are only applied to the trunk. So the style branch gets more and more outdated.
Is there anything known to be broken in the the style-branch compared to the trunk, so that it can not/should not be merged?
Well it never worked together with the "continue" commands..... At least for me the style-branch never brought any improvements.... I think it would be better to leave the action rules untouched, but build upon the continue with_actions command. My idea would be that [ x0000 continue with_actions ] or [resolution X continue with_actions] would be implemented to work as action rule without actually being "rendered" to the map. (only problem is that it should be "forgotten" completely for later, so that one can still use continue on the same object later, and again one more time without continue).
Gruss Torsten _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
On 01/04/2010 11:18 AM, Johann Gail wrote:
Sorry, I don't see the use of a black/whitelist of tags. The information of used tags should be contained in the style files, so why should I have to manage another additional file?
At present there is no way of blacklisting in the style files. At least I haven't found one. You can omit tags, but you cannot do something like this: highway=track {blacklist} highway=* [0x00] The wildcard is very handy for catching unexpected tag values. If you want to use it and discard certain other tag values you need some way of blacklisting.
participants (6)
-
Chris-Hein Lunkhusen -
Felix Hartmann -
Johann Gail -
Ralf Kleineisel -
Torsten Leistikow -
WanMil