On Jul 29, 2010, at 1:07 PM, ext Frank Warmerdam wrote:

> Tamas Szekeres wrote:
>> Martin,
>> 
>> I didn't follow all conclusions made earlier in this thread, but I don't 
>> think it makes much sense to enable/disable the special fields since 
>> those are derived from the attributes of the feature instead of fetching 
>> the values physically from the data source. For example OGR_STYLE 
>> corresponds to OGRFeature::GetStyleString() which can provide a 
>> meaningful value even if the styles are not handled/supported by a 
>> particular driver.
> 
> Tamas,
> 
> Geometry is generally derived from the physical data source, and knowing
> that it can be ommitted can save substantial io and processing.  I must
> confess this is less obvious with regard to styles.  I can *imagine*
> styles being expensive to collect and form in some cases but I can't
> think of any existing driver collecting the style string is particularly
> expensive.

One that occurs to me is the KML driver, where styles may exist seperately
from the features; parsing the style KML (and applying all the cascading
stylesheets, etc.) could be expensive, and excluding it could save
some processing time in theory.

In practice, I don't know whether this is the case for the current driver.

-- Chris
_______________________________________________
gdal-dev mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/gdal-dev

Reply via email to