Hi Philip

>From the traceback it looks like an area is being split into an excessive 
>number
of subdivisions. Masking the first exception won't help produce a reasonable
map.

Worth trying is to put back the -ea option and try option
  --order-by-decreasing-area
which changes the behaviour the splitting algorithm.

I've never looked at how contours work: eg. where the data comes from and how it
is merged into the map generation process. Can you briefly explain your process
and send your splitter & mkgmap options and, if not using the default style,
your style.

I'll see what I can do.

Ticker

On Fri, 2025-09-19 at 13:53 +0200, Philip Homburg via mkgmap-dev wrote:
> > In the short term, try removing the -ea option from your java
> > command. This should stop this crash but the part of resultant map
> > might be rubbish or the bit might be not noticeable. I'll have a
> > look sometime to see if I can work out why it is generating a value
> > that won't fit into the allocated #bits in the .img file.
> 
> I tried that on one the pbf files that came out of the splitter and that
> seemed fine. Then I tried it on the entire collection. I first got
> a failure that 8GB memory was not enough. So made it 16GB. But now
> it seems to be stuck in a loop.
> 
> If you need any help reproducing the problem then please let me know.
> _______________________________________________
> mkgmap-dev mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
> %(web_page_url)slistinfo/%(_internal_name)s

_______________________________________________
mkgmap-dev mailing list -- [email protected]
To unsubscribe send an email to [email protected]
%(web_page_url)slistinfo/%(_internal_name)s

Reply via email to