feature-access

The gdalheaders API uses the following (pseudo-code) schemes for feature access.

The different feature loops, see https://github.com/dis-organization/dirigible/issues/5

ALL

Get all the feature thingys.

  double   nFeature = gdalheaders::force_layer_feature_count(poLayer);
  List/Vector out (nFeature);

  double ii = 0;
  while( (poFeature = poLayer->GetNextFeature()) != NULL ) {
   out[ii] = poFeature-><something>; 
   OGRFeature::DestroyFeature(poFeature);
   ii++;
  }

IJ

Get a subset of feature thingys i:j - their positional index.

For c(0, 0) get the first feature, for c(0, n - 1) get all of them. For c(0, -1) that is a special case, there are no features.

  ListVector out(ij[1] - ij[0] + 1);
  double cnt = 0;
  double ii = 0;
  while( (poFeature = poLayer->GetNextFeature()) != NULL ) {
    if (ii == ij[0] || (ii > ij[0] && ii <= ij[1])) {
      out[cnt] = poFeature-><something>; 
      cnt++;
    }
    ii++;
    OGRFeature::DestroyFeature(poFeature);
  }

IA

Get a subset of feature things, arbitrary i (in order) - their positional index

  ListVector out(ia.length());
  double ii = 0;
  double cnt = 0;
  while( (poFeature = poLayer->GetNextFeature()) != NULL ) {
    if (ii == ia[cnt]) {
      out[cnt] = poFeature-><something>; 
      cnt++;
    }
    ii++;
    OGRFeature::DestroyFeature(poFeature);
  }

FA

Get a subset of feature thingys arbitrary FID (order irrelevant) - their FID unique names

  List out(fa.length());

  for (double ii = 0; ii < fa.length(); ii++) {
    GIntBig feature_id = (GIntBig)fa[ii];
      poFeature = poLayer->GetFeature(feature_id);