A Header represents the data stored in the header section of a STEP file:

• description should be an informal description of the contents of the file.
• implementationLevel will typically be "2;1" indicating version 2 of ISO 10303, conformance class 1 (which in turn means that the file has a single data section and no anchor or reference sections, along with several other restrictions). For other possible values, see section 8.2.2 of ISO 10303-21.
• fileName may be the file name of the actual file, or it may be an abstract name for the contents of the file used when cross-referencing between files.
• timeStamp should be an ISO 8601-formatted date and time.
• author should include the name and address of the person who created the file.
• organization should be the organization that the author is associated with.
• One of preprocessorVersion or originatingSystem should identify what CAD program was used to generate the file. This does not seem to be used terribly consistently!
• authorization should include the name and address of whoever authorized sending the file.
• schemaIdentifiers identifies the EXPRESS schema used by entities in the file. This will usually be a list containing a single string, which may be either a simple string like "IFC2X3" or an 'object identifier' such as "AUTOMOTIVE_DESIGN { 1 0 10303 214 1 1 1 1 }" (more commonly known as AP214).

An Entity represents a single entity stored in the data section of a STEP file. An entity may be a point, a curve, a part, an assembly, or even an entire building. Entities may be 'simple' (having a type and a list of attributes, which can themselves be references to other entities) or 'complex' (effectively a list of simple entities combined together).

Instead of creating or inspecting Entity values directly, you will generally create them using Step.Encode.entity and extract data from them using Step.Decode.entity.

Note that if you do construct Entity values directly, you should generally use Nothing as the entity ID. Entity IDs are auto-generated anyways when encoding a STEP file, so there is no point in specifying them. However, entities read from a file will be assigned the ID they had in that file. (Note that even if those entities are written out to another file later, they will have fresh IDs generated.)

Some entities are 'complex', which means they are composed of a number of sub-entities each with its own type and attributes. A SubEntity represents one such sub-entity.

An Attribute represents a single attribute of an Entity, such as an X coordinate value, a GUID string, or a reference to another entity.

Instead of creating or inspecting Attribute values directly, you will generally create them using an encoder and extract data from them using a decoder.