IMPORT POLYGONS Statement

Nomenclature

By examining the sample 'waterbody.dbf' shapefile (found in the 'GeoBase Examples\Alchemy\Allegheny' folder) using a spreadsheet package we see that there are 7 columns: 'ID', 'AREA', 'PERIMETER', and so on.

Knowing that bodies of water are polygons, we will use the IMPORT POLYGONS command to import this data. The 'ID' column (above) provides a unique ID for each feature and we will use the 'NAME' column as the label. The language is English.

The IMPORT POLYGONS statement to import the geometries of the polygons would be structured as follows:

IMPORT POLYGONS [
  ID = %ID,
  LABEL = %NAME,
  LANG = "ENG"
] FROM "waterbody";

To import only the bodies of water with an area greater than (for example) 1000 we would add a WHERE clause to the IMPORT POLYGONS statement, as follows:

IMPORT POLYGONS [
  ID = %ID,
  LABEL = %NAME,
  LANG = "ENG"
] FROM "waterbody" WHERE %AREA > 1000;

The following GBFS columns are available for population through the Alchemy IMPORT POLYGONS command.

This table shows the different polygon types. The polygon TYPE field should contain one of the type values listed below (in the left column).

To import Polygons from CSV format text files, use the following format.

IMPORT POLYGONS [id = %id, name = %name, geom = wkt((%polygon))] FROM "csv://src_file";

polygon_id,polygon_label,polygon_type,polygon_data
1,Telogis Park,15,"POLYGON((-118.268898 34.045484,-118.267866 34.045484,-118.267866 34.044802,-118.268902 34.044798,-118.268898 34.045484))"

Note
Note the formatting of our polygon geometry. For Well Known Text, the location of each vertice should be formatted as 'Longitude Latitude', reversing the conventional GeoBase standard of 'Latitude Longitude'. Note also the use of a space as a separator rather than a comma (commas are is used instead to separate field values).

The IMPORT POLYGONS statement required to import data from the example CSV file shown above is:

IMPORT POLYGONS [
  id = %polygon_id,
  label = %polygon_label,
  type = %polygon_type,
  geom = wkt(%polygon_data)
] FROM "csv://mycsvfile.csv";

The 'polygon_id' column provides a unique ID for each polygon shape, the 'polygon_label' column a name for each polygon and the 'polygon_type' the type of polygon (refer to the 'Polygon Types' table above). The 'polygon_data' column provides polygon data stored as WKT string Polygons.

Polygon data stored in the CSV file should be contained within quotes, unless data is separated by pipes; for example:

polygon_id|polygon_label|polygon_type|polygon_data
1|Telogis Park|15|POLYGON((-118.268898 34.045484,-118.267866 34.045484,-118.267866 34.044802,-118.268902 34.044798,-118.268898 34.045484))
.....