====== Attributes map file ======

The L3DT //attributes map file// (AMF) stores the land type of each pixel in the height map. The 'amf.gz' extension indicates the AMF file has been compressed using the gzip algorithm.

The data is preceded by a 10-byte header containing the following information:

^  Field name  ^ Contents ^ Length^ Offset ^ Purpose ^
|  ASCII marker  |  "L3DT"  |  4  |  0  | Tells L3DT and humans that this is an L3DT file. Note string is not null terminated. |
|  Binary marker  |  USHORT  |  2  |  4  | Tells L3DT what type of map file this is. Must be 520 for AMF. |
|  Map width  |  USHORT  |  2  |  6  | East-west extent of the map (number of pixels). |
|  Map height  |  USHORT  |  2  |  8  | North-south extent of the map (number of pixels). |

The map data itself has two bytes per pixel, the first being 'land type ID' and the second being 'climate ID'. The pixels are ordered in rows going west-to-east, with south-to-north row-order.

The climate ID value indicates the ID number of the relevant climate within the group climate list (which you can find in the //Climates// section of the [[l3dt:formats:specs:proj|project file]]). The land type ID indicates the ID number of the land type in that climate (see //LandTypeList// section of the [[l3dt:formats:specs:cli.xml|climate file]]). You also can get the combined 2-byte values using the [[l3dt:userguide:ops:am#exporting_the_type_list|export type list]] option.

===== Addenda for L3DT v2.8 and later =====

In L3DT v2.8 and later versions, the AMF file contains an additional list of climate names after the map data. This extra data consist of:
  * A 32-bit integer for the number of climates.
  * Then, for each climate:
    * A 32-bit integer for the climate name length.
    * The climate name, the length of which is given by the preceding integer. The string is not null-terminated.

This additional data is optional, and can be safely omitted or ignored.