[mkgmap-dev] Regression causing Heap space exhaustion?

Sat Aug 22 12:54:12 BST 2009


>   The reason for the high memory usage is that most nodes in the Danish 
>   extract look like this:
>
>   <node id='342123311' lat='55.3401904' lon='9.4886152'>
>   <tag k='addr:postcode' v='6070'/>
>   <tag k='kms:zip_no' v='6070'/>
>   <tag k='kms:street_name' v='Seggelund Hovedvej'/>
>   <tag k='kms:last_updated' v='2002'/>
>   <tag k='kms:county_name' v='Sønderjyllands Amtskommune'/>
>   <tag k='addr:street' v='Seggelund Hovedvej'/>
>   <tag k='kms:municipality_name' v='Christiansfeld'/>
>
> This gets very messy, but I suppose tags could be trimmed on reading if
> there is some way to know/declare that they won't be used.  The problem
> is then the relation matching can't use those, and this could get
> mysterious.  I wonder if the style file could have a section to declare
> node tags to trim, and then abort if they are later used in styles?
>   
In this case I would suggest to inverse the logic. While reading, trim 
all tags which are not contained in the style file. So they are not 
supposed to be used later on.