[mkgmap-dev] Experimental patches reg. Collator.strength and other index issues

Mon Nov 22 13:52:27 GMT 2021

Hi devs,

with the attached patch for MdrCheck I see no error messages reg. wrong repeat flags for maps from Garmin.
The patch for mkgmap shoud get us closer to what MapSource (or Garmin) does.
- strength is changed from secondary to tertiary (maybe even string.equals can be used)
- mdr11 records are also sorted and grouped by city id (need a new createSortKey() method to avoid the DoubleSortKey

I've noticed that a gmapsupp index created by MapSource for a map which contains only POI but no highways has a different "extra value" in Mdr11, e.g. 0x51 instead of 0x91. 
Both have a Mdr17, so I guess we need some extra logic in Mdr11.getExtraValue(). 
MapSource writes a different 

I've started again to find out what the "unknown" byte in Mdr4 might mean. Unfortunately this section is not recalculated by MapSource, it simply seems to copy the value that is written by mkgmap.

Gerd
-------------- next part --------------
A non-text attachment was scrubbed...
Name: display-no-secondary.patch
Type: application/octet-stream
Size: 1348 bytes
Desc: display-no-secondary.patch
URL: <http://www.mkgmap.org.uk/pipermail/mkgmap-dev/attachments/20211122/18c478c3/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mkgmap-no-secondary.patch
Type: application/octet-stream
Size: 4326 bytes
Desc: mkgmap-no-secondary.patch
URL: <http://www.mkgmap.org.uk/pipermail/mkgmap-dev/attachments/20211122/18c478c3/attachment-0001.obj>