.POV File Format – Reserved words

Token categories

Reserved words are matched against the text word by word, so a word that contains another (boxed, box) is a different word; the scanner takes the whole identifier first and looks it up afterwards.

The header reservedwords.h numbers the tokens in blocks. The order matters: the parser tests a token against the end of a block to ask which kind of expression the token can start. A token that is in no block is a plain keyword. There are four categories.

A float expression is therefore recognised by its first token: a number, a float identifier, or one of the float words. The block order of the source also puts sqr outside the float block: sqr is a keyword of the julia_fractal object only and is not a function. The expression rules are on Expressions, vectors and identifiers, and the functions one by one on Functions and constants.

Examples

The two spellings of a word give the same token, so they can be mixed. A word that is compiled out (see the end of the page) is free to use as a name.

#declare A = color rgb <0.2, 0.4, 0.6>;
#declare B = colour red 0.2 green 0.4 blue 0.6;
#declare Same = ((A.red = B.red) & (A.green = B.green) & (A.blue = B.blue));
#debug concat("same colour: ", str(Same, 0, 0), "\n")
#declare breakpoint = 7;
#declare obj = 8;
#debug concat("free words: ", str(breakpoint + obj, 0, 0), "\n")
sphere { 0, 1 pigment { gradient x color_map { [0 color A] [1 colour B] } } }

A reserved word cannot be declared. The messages of the two common mistakes:

#declare box = 1;
  Parse Error: Expected 'undeclared identifier', box found instead

#declare a = sqr(2);
  Parse Error: Expected 'RValue to declare', sqr found instead

Directive and statement words

The words that follow #, and the words that belong to declarations and file access.

Objects

Keywords that begin a shape or a group of shapes. See Finite objects, Implicit objects, Meshes and CSG.

Keywords inside objects

Arguments and sub-keywords that belong to one kind of object.

Object modifiers

Items that may follow an object. See Object modifiers.

Transformations and warps

Camera words

See Cameras.

Light words

See Lights.

Texture, pigment, normal and finish words

See Textures and pigments and Normals and finishes.

Interior, media and atmosphere words

See Media and interior.

Global settings words

See Global settings.

Pattern words

See Patterns and maps.

File-type and gamma words

Functions

Float, vector and string functions. Their arguments and results are on Functions and constants.

Constants and built-in variables

Colour words

Spline, array and dictionary words

Alternative spellings

These words are listed in a block of their own in the table. Each gives the same token as the word in the second column.

Words that are compiled out

Eight words of the table are inside #if blocks of reservedwords.cpp. In this build they are ordinary identifiers: a scene that declares #declare breakpoint = 1; or #declare obj = 2; is accepted and runs.

Operator and punctuation tokens

The table of reserved words also holds 34 operator and punctuation tokens. Three of them have two characters: >=, <= and !=. How the scanner reads them is on Lexical structure, and how the parser uses them in Expressions, vectors and identifiers.

Pseudo-tokens

The last part of the table holds tokens that are not words of the language. They give the parser names for what it has just read, and the second column is what the error messages call them. The identifier tokens are the type tags of the symbol table; the table of identifier types, with what declares each, is on Expressions, vectors and identifiers.