here: www.geocaching-dresden.de/mkgmap/style.zip i found a workaround just now: the style contains a file called "lines workaround", which does the naming of the roads in a separate step - no freezing. micha Am 20.10.2011 17:58, schrieb WanMil:

...

i'm trying to create a lines-only routable map. from a certain amount of lines in the style mkgmap starts turning out 2 empty img files (empty means 0 bytes, not empty maps), then memory usage freezes pretty much, cpu usage oscillates wildly and nothing more happens.

this problem is tied to --route, leaving the option out turns things back to normal.

a certain amount of poi's in the style solve the problem too - with some lines for cities in the points file everything works as expected. it almost feels as if mkgmap "needs something to do" between all the roads.

any ideas? Can you please send the problematic style and your mkgmap parameters so that we can try to reproduce the problem?

Thanks WanMil _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

On 20/10/11 21:36, WanMil wrote:

I can confirm that behaviour. Although in real mkgmap does not freeze (endless loop) but is working very very hard. Possibly the mkgmap code is not very optimal for the special test case.

@Steve, the code requires a very long time in the NETFile.simplifySortedRoads method because I think there are many roads with the same name. I think I remember a similar discussion some time ago? The method seems to perform a brute force method to detect connected roads with the same name. Can you have a look on it?

Yes that routine is a disaster when there are many roads with the same name. I'm not sure what it is supposed to do and how important it is. I'll have a go at fixing it. ..Steve

Am 29.10.2011 15:57, schrieb Steve Ratcliffe:

Hi

I would like some opinions please from those that make maps.

If you have a style that say sets the name of every path to 'Path'

eg: highway=path | highway=footway | highway=track {name 'Path' }

or anything similar that results in thousands of roads with the same name, then mkgmap appears to freeze.

The code that is slow is creating a list of roads that will be used when you search for an address in MapSource or on the device.

I can fix this in various ways, but I want to ask how useful it is to have thousands of roads with the same name appear as a result of a search?

For a start in mapsource, you only see a few of the results anyway.

Should I drop the name completely from the index when there are more than say 500 roads with the same name in the same city?

There is a branch called simplify-sorted-roads where I am trying out different approaches to the problem, if anyone else wants to try it out or look at it. I think it would be better to control it via style-file. Eg. you should add a tagg like mkgmap:nomerge=yes for such ways.

Henning

it would make a lot of sense if such names do not find their way into the index. to control that behaviour flagging those ways via the style file feels like the most flexible method. Am 29.10.2011 15:57, schrieb Steve Ratcliffe:

Hi

I would like some opinions please from those that make maps.

If you have a style that say sets the name of every path to 'Path'

eg: highway=path | highway=footway | highway=track {name 'Path' }

or anything similar that results in thousands of roads with the same name, then mkgmap appears to freeze.

The code that is slow is creating a list of roads that will be used when you search for an address in MapSource or on the device.

I can fix this in various ways, but I want to ask how useful it is to have thousands of roads with the same name appear as a result of a search?

For a start in mapsource, you only see a few of the results anyway.

Should I drop the name completely from the index when there are more than say 500 roads with the same name in the same city?

There is a branch called simplify-sorted-roads where I am trying out different approaches to the problem, if anyone else wants to try it out or look at it.

...

I can confirm that behaviour. Although in real mkgmap does not freeze (endless loop) but is working very very hard. Possibly the mkgmap code is not very optimal for the special test case.

@Steve, the code requires a very long time in the NETFile.simplifySortedRoads method because I think there are many roads with the same name. I think I remember a similar discussion some time ago? The method seems to perform a brute force method to detect connected roads with the same name. Can you have a look on it? ..Steve

---

mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

Yes that is probably best. I set names pretty extensively, and even I would only need about 80 additional lines of code in my lines style with mkgmap:noindex=true (as I have to duplicate all lines where names are set to have one with name!=* set mkgmap:nopreviousname=yes and then check that condition on any future setting of names, the other option of doing a (highway=* & name!=* & ref!=* & route_name!=*) set mkgmap:nopreviousname=yes wouldn't mean less lines of new code). On 31.10.2011 11:40, WanMil wrote:

I think it's a good idea to use a flag like 'mkgmap:noindex=true' which can be set in the style file. That provides the best flexibility.

WanMil

...

Hi

I would like some opinions please from those that make maps.

If you have a style that say sets the name of every path to 'Path'

eg: highway=path | highway=footway | highway=track {name 'Path' }

or anything similar that results in thousands of roads with the same name, then mkgmap appears to freeze.

The code that is slow is creating a list of roads that will be used when you search for an address in MapSource or on the device.

I can fix this in various ways, but I want to ask how useful it is to have thousands of roads with the same name appear as a result of a search?

For a start in mapsource, you only see a few of the results anyway.

Should I drop the name completely from the index when there are more than say 500 roads with the same name in the same city?

There is a branch called simplify-sorted-roads where I am trying out different approaches to the problem, if anyone else wants to try it out or look at it.

...

I can confirm that behaviour. Although in real mkgmap does not freeze (endless loop) but is working very very hard. Possibly the mkgmap code is not very optimal for the special test case.

@Steve, the code requires a very long time in the NETFile.simplifySortedRoads method because I think there are many roads with the same name. I think I remember a similar discussion some time ago? The method seems to perform a brute force method to detect connected roads with the same name. Can you have a look on it? ..Steve

---

mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

---

mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

Actually I just thought about one possible problem, as I don't know how the mkgmap:noindex=true will be implemented. What happens when a user simply rebuilds the address index of existing .img files without using a style-file when mkgmap:noindex=true had been set for problematic cases? a) will it then slow down if many roads have the same name b) doesn't matter, the names are now in such a position that mkgmap will not use them for address index? If a) is the case, then I think there would also need to be a switch to enable say 300 roads same name, don't index. If b) is the case then the mkgmap:noindex=true setup is sufficient. On 10.11.2011 13:25, Steve Ratcliffe wrote:

On 31/10/11 10:40, WanMil wrote:

...

I think it's a good idea to use a flag like 'mkgmap:noindex=true' which can be set in the style file. That provides the best flexibility. The option appears a popular choice! It might be a good addition, although I think it would have to work with all element types and not just roads.

So I will take it that there is no-one in favour of just removing often repeated names automatically, and really as long as they don't cause any harm I don't see any problem with them.

So I will merge later the current simplify-sorted-roads branch which now only uses the modified algorithm when there is a large number of same named roads. It never takes more than a fraction of a second.

..Steve

_______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

On 11.11.2011 11:34, Steve Ratcliffe wrote:

Hi

...

What happens when a user simply rebuilds the address index of existing .img files without using a style-file when mkgmap:noindex=true had been set for problematic cases?

a) will it then slow down if many roads have the same name b) doesn't matter, the names are now in such a position that mkgmap will not use them for address index?

It doesn't make much difference for the global index. There is only ever one copy of a name for a map/city/region/country combination in the global index. It is only the individual map tiles that contain an entry for each road with the same name. The global index only says that a particular map has at least one road with a given name.

If Mapsource or a device crashes because of the large number of names, then we will just have to limit the number, it would not be optional.

..Steve

...

I think you did not understand what I meant. 1. I compile maps using a style-file that does highway=track & name!=* {set name 'track'; mkgmap:noindex=true} 2. A user/me wants to add contourlines to the map and uses mkgmap to recompile the maps including the contourlines into a new mapset. tdb/mdr.img needs to be written from scratch, however the user uses the default style-file that does not have mkgmap:noindex=true information for highway=track. So I don't want that the index in case 1. is correct (omits those thousands of roads with name "track", but for step 2. the index is incorrect and actually has thousands of "track" roads in the index. If this is how it works then we would need (additionally to mkgmap:noindex=true) an option like mkgmap:noindex:same=XXX for the command line.

thanks for clearing that up. I didn't know enough about the indexes. So style-file command is enough for anything that I can imagine.

maybe it would be a good idea to have a command line switch *and* a style rule. the command line switch would have to default to "no-index=false", otherwise indexing in all older styles would break. a 2nd optinon "no-index=true" could automatically add a no-index tag to all lines which then can be set to false in the style whenever we want a line indexed. Am 29.10.2011 15:57, schrieb Steve Ratcliffe:

Hi

I would like some opinions please from those that make maps.

If you have a style that say sets the name of every path to 'Path'

eg: highway=path | highway=footway | highway=track {name 'Path' }

or anything similar that results in thousands of roads with the same name, then mkgmap appears to freeze.

The code that is slow is creating a list of roads that will be used when you search for an address in MapSource or on the device.

I can fix this in various ways, but I want to ask how useful it is to have thousands of roads with the same name appear as a result of a search?

For a start in mapsource, you only see a few of the results anyway.

Should I drop the name completely from the index when there are more than say 500 roads with the same name in the same city?

There is a branch called simplify-sorted-roads where I am trying out different approaches to the problem, if anyone else wants to try it out or look at it.

...

I can confirm that behaviour. Although in real mkgmap does not freeze (endless loop) but is working very very hard. Possibly the mkgmap code is not very optimal for the special test case.

@Steve, the code requires a very long time in the NETFile.simplifySortedRoads method because I think there are many roads with the same name. I think I remember a similar discussion some time ago? The method seems to perform a brute force method to detect connected roads with the same name. Can you have a look on it? ..Steve

---

mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev