- mkgmap-dev - The mkgmap lists

[Patch v2] evaluate type=associatedStreet
by Gerd Petermann 02 Dec '14

02 Dec '14

Hi all, attached is a slightly modified version of the patch with improved logging. I suggest to use uk.me.parabola.mkgmap.osmstyle.housenumber.level=FINE in the logging props if you want to find out more about housenumber processing. If I hear no complains I'll commit this patch on thursday. Gerd

1 0

Can't split into big tiles
by Andrzej Popowski 01 Dec '14

01 Dec '14

Hi, I would like to split Germany data into 5 tiles. I have added contour lines and sea area to OSM data but this probably is not important. My file has about 200M of nodes. If I split with max-nodes 50M, then I get about 50 tiles, which is wrong. If I split with num-tiles=5 then bounding polygon is ignored and I get tiles with boxes like: 63240002: 2176512,-935424 to 2796288,355584 # : 46.702881,-20.072021 to 60.001831,7.630005 This is useless for me, since I need true bounding boxes for tiles. I have used splitter r412. Any solution for this problem? Temporary I'm going to clip data with osmosis. -- Best regards, Andrzej

4 14

relation associatedStreet
by Andrzej Popowski 01 Dec '14

01 Dec '14

Hi, I can't find references in mkgmap code to relation associatedStreet. Is it supported? http://wiki.openstreetmap.org/wiki/Relation:associatedStreet This relation bounds houses to streets. I think it should be used by mkgmap to add housenumbers to streets. Relation can't be described in style, since it is defined basing on members role instead of tags. Example of use could be following: <node id="101" Lat="12.31" Lon="4.56"> <tag k="addr:housenumber" v="5"/> </node> <way id="12"> <nd ref="200"> <nd ref="201"> <tag k="highway" v="tertiary"/> <tag k="name" v="Long Street"/> </node> <relation id="3"> <member type="node" ref="101" role="house" /> <member type="way" ref="12" role="street" /> <tag k="type" v="associatedStreet" /> </relation> Can mkgmap find that house number 5 belongs to Long Street? -- Best regards, Andrzej

4 7

[Patch v1] splitter: keep associatedStreet rel complete
by GerdP 01 Dec '14

01 Dec '14

Hi all, for those who want to test the patch evaluate-associatedStreet-rel-v1.patch here is a small patch for splitter to make sure that the relations are complete at tile borders. keep-associatedStreet-rel-complete-v1.patch <http://gis.19327.n5.nabble.com/file/n5825929/keep-associatedStreet-rel-comp…> Gerd -- View this message in context: http://gis.19327.n5.nabble.com/Patch-v1-splitter-keep-associatedStreet-rel-… Sent from the Mkgmap Development mailing list archive at Nabble.com.

1 0

Mailing list
by Steve Ratcliffe 30 Nov '14

30 Nov '14

Hi There was a problem last week where certain posts were not received by list members that had Hotmail or Yahoo email addresses and this eventually resulted in all such email addresses being disabled due to bounced posts. I have now re-enabled everyone that was affected so they should receive mail from the list again. It is no longer possible to sign up or post to the list with a Yahoo email address. If you are already subscribed with a Yahoo email address, you should still receive posts, but if you want to post, then you should do so from a different email address. [This also applies to AOL, although nobody on the list actually uses AOL.] Sorry about this restriction, it may be removed at a later time. ..Steve

1 0

Mailing list maintenance
by Steve Ratcliffe 29 Nov '14

29 Nov '14

Hi everyone I've put the list into emergency moderation mode, so that all messages will be held until problems with bouncing messages due to a change made by Yahoo! are solved. https://www.virusbtn.com/blog/2014/04_15.xml ..Steve

1 0

Substring returns full string
by gbbickerton 29 Nov '14

29 Nov '14

I have (ncn=yes | rcn=yes) { name '${rcnname | substring:1:14}'} [0x10018 resolution 22 continue with_actions] in my lines file, rcnname is 123456789123456789 for test purposes, I get the full string showing on the map in basecamp not the 1 - 14 substring am I doing something wrong ? -- View this message in context: http://gis.19327.n5.nabble.com/Substring-returns-full-string-tp5825603.html Sent from the Mkgmap Development mailing list archive at Nabble.com.

5 10

[Patch v2] style filter country-ISO
by Gerd Petermann 28 Nov '14

28 Nov '14

Attached is v2 of the patch, now based on r3357, as v1 causes a conflict. I've also added a unit test and a bit of documentation. If I hear no complains I'll commit this on Monday. Gerd

1 0

Commit: r3357: Allow a space between the pipe symbol and the filter name.
by svn commit 28 Nov '14

28 Nov '14

Version mkgmap-r3357 was committed by steve on Fri, 28 Nov 2014 Allow a space between the pipe symbol and the filter name. In the unquoted syntax allow spaces '${rcnname | substring:1:5}' Also added an error when the filter name is not recognised, instead of silently ignoring it.

1 0

Manage drive-on-left/drive-on-right in resources\LocatorConfig.xml
by Gerd Petermann 28 Nov '14

28 Nov '14

Hi all, this point is on the to do list, but I am not sure what the target is. I guess it should replace the two options drive-on-right and drive-on-left. ("dor","dol") A possible implementation looks like this: 1) add property driveOnLeft="true" to the appropriate countries in LocatorConfig.xml 2) add methods to lookup the property for a given roundabout 3) count the number of "dol" and "dor" roundabouts 4) If neither the --drive-on-left option or --drive-on-right option is set, use the counters: if the "dol" number is larger than the "dor" number, set the "dol" flag in the NOD header to 1. So far so easy. I guess in most usage cases one of the counters is 0 and the other is not, so that would work fine. Open questions: 1) What should happen for those tiles containing both "dol" and "dor" roundabouts? 2) What should happen if the --drive-on-left option is given and the tile contains only "dor" roundabouts (or vice-versa) ? Should we ignore the option given by the user and print a corresponding warning? 3) We have the --check-roundabouts option which also might set the flag in the NOD header if the user doesn't specify --drive-on-left or --drive-on-right. The current implementation simply uses the direction of first roundabout (clockwise or not) to decide which driving side is to be used and it reverses the direction of all following roundabouts which don't match the direction of the first one. I assume this should be changed so that we use the above counters as well? @Steve: The data in the LocatorConfig.xml seems to be only used in combination with the --bounds option, without precompiled bounds the default style will not set mkgmap:country which is used to get the data from the LocatorConfig.xml. I did expect that an option like --country-name or --country-abbr could be used to set a default, and the code really passes the values to a method in Locator called setDefaultCountry(), but that method is only adding /overwriting a HashMap of country-name+iso-code which is filled from LocatorConfig.xml . Is that intended?

6 16