idea for the manual

[Dirk Hohndel]

On Wed, Dec 10, 2014 at 03:12:18PM +0000, Pedro Neves wrote:
> Subsurface's XML format supports four coordinate formats according to the
> manual:
> 
> "The coordinates can be entered by hand if they are known, using one of four
> formats with latitude followed by longitude:
> ISO 6709 Annex D format e.g. 30°13'28.9"N 30°49'1.5"E
> Degrees and decimal minutes, e.g. N30° 13.49760' , E30° 49.30788'
> Degrees minutes seconds, e.g. N30° 13' 29.8" , E30° 49' 1.5"
> Decimal degrees, e.g. 30.22496 , 30.821798
> Southern hemisphere latitudes are given with a S, e.g. S30°, or with a
> negative value, e.g. -30.22496. Similarly, western longitudes are given with
> a W, e.g. W07°, or with a negative value, e.g. -7.34323."
> 
> However I can only import GPS coordinates from a .CSV file if they are
> expressed as decimal degrees (and without a comma separating latitude and
> longitude): "30.22496 30.821798". All the other formats result in empty
> coordinates on Subsurface.
> 
> I'm not sure if this has to do with the import filter code or with my
> locale, but it would be helpfull if someone could run a few tests on other
> machines just to make sure. I'm inclined to write that the GPS coordinates
> on the .CSV file have to follow the format "dd.dddd dd.dddd", with a "-" for
> S latitudes and W longitudes

This is almost certainly Miika's input filter... we have some rather
fragile C++ functions to parse all the other formats, but I don't think he
can easily call those from XSLT...

/D