[gdal-dev] S57 field definition incorrect

[Frank Warmerdam]

ogr user wrote:
> I see that now in code. I added my own wrapper to parse the list, but
> there are two issues remaining:
> 1) efficiency - if a field must be read multiple times, string will have
> to be parsed multiple times, whereas if it were treated as integer list
> from the start, string parsing would be done once. This may not be much,
> but it adds up when thousands of features are processed.
> 2) it is impossible to define default behavior for integer list fields,
> because they self-identify as strings - i.e. even if I were to parse it
> in my code (which I do now), I have to know from another source whether
> the field is an integer list.

OGR User,

I will say I find the second issue stronger than the first.

> May be it coud be done as an option, controlled by flag to a class
> registrar (and default behavior be made as it was previously)?

This is possible, but to me it seems like additional complication with
relatively little value.  The existing number of S57 processing options
is mostly due to the need to provide the funding client with some very
specific abilities.

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent