logo separator

[mkgmap-dev] [Patch] Faster BoundaryGrid

From WanMil wmgcnfg at web.de on Wed May 2 21:14:32 BST 2012

Hi Gerd,

the patch looks good but I am a bit in doubt if it really improves the 
overall performance for most cases (I haven't measured myself).

Did you measure the performance diffs when all bounds files are 
available? This is more likely than the case that the tile contains 
mostly sea.
I believe the patched version is slower if the bounds files are 
available because the 
BoundaryUtil.getBoundaryDirContent(boundaryDirName) method might take 
some time to list the several thousand bounds files.

WanMil

> Hi WanMil,
>
> attached is a small patch which drastically improves BoundaryGrid for
> tiles that contain mostly sea. I found such a tile in south-america, and the
> patched version is>  10 times faster.
> It contais a large part of the pacific, and BoudaryGrid searches the
> boundary zip file for
>> 5000 different *.bnd files, most of them do not exist in your
>> bounds_20120401.zip , so the
> code in BoundaryUtil searches through all elements of the zip file to find
> out that it doesn't exist.
>
> The patch creates a HashSet first with the  available files and calls the
> load routine only for
> those that exist.
>
> Gerd
> http://gis.19327.n5.nabble.com/file/n5680180/BoundaryGrid.java.patch
> BoundaryGrid.java.patch
>
>
> --
> View this message in context: http://gis.19327.n5.nabble.com/Patch-Faster-BoundaryGrid-tp5680180.html
> Sent from the Mkgmap Development mailing list archive at Nabble.com.
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev




More information about the mkgmap-dev mailing list