[mkgmap-dev] Error in TreReader? Are our limits too small?

Fri Jan 28 16:02:58 GMT 2022

Hi Ticker,

yes, the patch is only for the reader.
Reg.  MAX_RGN_OFFSET_SIZE : You are right, it is about the pointers in RGN like indPointOffset in RgnFileReader.
Anyway, with WANTED_MAX_AREA_SIZE = 0x3fff we want to produce even smaller subdivs .

Gerd

________________________________________
Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von Ticker Berkin <rwb-mkgmap at jagit.co.uk>
Gesendet: Freitag, 28. Januar 2022 13:04
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] Error in TreReader? Are our limits too small?

Hi Gerd

Subdivision.write needs adjusting to merge startRgnPointer and
getType()

MAX_RGN_OFFSET_SIZE limits to 16 bits rather than 24 so probably not be
directly related.

The 24 to 28 bit increase presumably covers all the subdivisions in the
tile and so should allow larger maps. I didn't notice anything that
checked the 24 bit limit apart from the put3u asserts

Ticker

On Fri, 2022-01-28 at 10:14 +0000, Gerd Petermann wrote:
> Hi programmers!
>
> Our code in TreFileReader differs from that in GPXSee treFile.cpp.
> Our code only reads 24 bits into field lastRgnOffset, but GPXSee
> reads up to 28 bits as offset.
> We assume 1 byte for the flags and 3 bytes for the offset. GPXSee
> assumes 4 bits for the flags and 28 bit for the offset.
>
> Since our flags field has values > 0x10 I think the code in GPXSee
> looks better.
> Not sure what that means. Maybe we can write larger tiles?
> Is MapSplitter.MAX_RGN_OFFSET_SIZE related to this?
>
> Gerd
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________
mkgmap-dev mailing list
mkgmap-dev at lists.mkgmap.org.uk
https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev