[mkgmap-dev] mkgmap: String index out of range: 48

Thu May 24 08:39:24 BST 2012

Hi WanMil,

I can confirm that the depth can be deeper, it is not wanted but I think
it's correct.
The patch is ok.
In some cases the patched MapSplitter "splits" the  data into two areas, one
is completely empty and the other contains all the data. The difference to
the unpatched version is only in the related bbox. I'll try to detect that
situation and avoid to save unwanted sub areas when I have finished the work
on tile splitter.

Gerd

WanMil wrote
> 
>> On Wed, May 23, Thorsten Kukuk wrote:
>>
>>>
>>> Hi WanMil,
>>>
>>> On Wed, May 23, WanMil wrote:
>>>
>>>> Hi Thorsten,
>>>>
>>>> r2278 changed only some parts of the generate-sea option. I think the
>>>> change to r2276 is more suspicious.
>>>> Which mkgmap release did you use before?
>>>
>>> r2263
>>>
>>>> Anyhow please try the following simple patch.
>>>
>>> I will try.
>>
>> Somehow the patch gots malformed from the mail system.
>> I incresed the number of white spaces and now it works
>> again.
>>
>>   Thanks,
>>    Thorsten
>>
> 
> Thanks Thorsten,
> 
> I will commit that to fix that by now.
> 
> Gerd, can you please check why your patch for r2276 increases the depth 
> of subdivision splits and if that's intended? Can you calculate a 
> maximum depth or a better fix than just adding some whitespaces like I 
> did so that it will work in all cases?
> 
> Thanks!
> WanMil
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at .org
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> 

--
View this message in context: http://gis.19327.n5.nabble.com/mkgmap-String-index-out-of-range-48-tp5709968p5710015.html
Sent from the Mkgmap Development mailing list archive at Nabble.com.