The IFC exchange format “STEP physical file” uses characters represented by decimal value 32 to 126 from the code table in ISO 8859-1. Any other character, like some Western characters, like the German “Umlaut”, Greek or Cyrillic letters, or Asian characters, has to be encoded before being exchanged as part of a string value. Up until IFC4.x this encoding is used in IFC. In the future, IFC will adopt the UTF8 encoding.
The rules for decoding and encoding are defined in ISO10303-21: “Industrial automation systems and integration — Product data representation and exchange — Part 21: Implementation methods: Clear text encoding of the exchange structure“. A short summary and guideline is included in the IFC Implementation Guide.
Example: The following encodings define the character “Upper A umlaut” Ä – the hexadecimal character code is xC4 (decimal 196)
| Characters | Description |
|---|---|
| ‘\S\D’ | character code of D = x44 (decimal 68) added to x80 (128) is x44 + x80 (68+128) = xC4 (196); since Ä is defined in ISO 8859-1 it is the default code page and no P encoding is required. |
| ‘\PA\\S\D’ | same as above, but the PA directive at the begin of the string explicitly defines that the value of xC4 (196) is taken from ISO 8859-1 |
| ‘\X\C4’ | character code xC4 as 8-bit character code found in ISO 10646 (first 255 characters – also referred to as “row 0”) |
| ‘\X2\00C4\X0\’ | character code xC4 as 16-bit character x00C4 in ISO 10646 (Unicode) |