enum RuleToSourceFlags in Tinman.Core.Parsing

No format flags for emitting source code.

Emits compact computerish source code instead of nice-looking human-readable code.

Omit AST-building rule syntax.

Omit whitespace trimming rules?

Also emit external rules in source code?

Outputs grammar rules in a way that represents the hierarchical structure of the abstract syntax tree that is created by the grammar.

Remarks:

AST rules are retained, all others are dropped (see Ast, AstIfLeft, AstIfRight). AST nodes are represented as token rules (see Token), using the AST node type as the value (see Type). The start and end of the children list of an AST node are marked with '{' and '}' token rules.

For example, the AST grammar of GrammarCode grammar looks like this:

grammar       := 'GRAMMAR' '{' (decl-skip | decl-rule)+ '}' ;

decl-rule     := 'RULE' '{' 'NAME' rule '}' | 'RULE_FLAGS' '{' 'RULE_FLAG' 'RULE' '{' 'NAME' rule '}' '}' ;
decl-skip     := 'RULE' '{' 'SKIP' rule '}' ;

rule          := rule-list ;
rule-ast      := 'EXPR_AST' '{' 'NAME' rule (('IF_LEFT' | 'IF_RIGHT') rule)? '}' ;
rule-brace    := 'EXPR_BRACE' '{' rule '}' ;
rule-choice   := rule-sequence | 'EXPR_CHOICE' '{' rule-sequence rule-sequence+ '}' ;
rule-except   := rule-not | 'EXPR_EXCEPT' '{' rule-not rule-not '}' ;
rule-list     := rule-choice | 'EXPR_LIST' '{' rule-choice rule-choice '}' ;
rule-not      := rule-primary | 'EXPR_NOT' '{' rule-primary '}' ;
rule-primary  := rule-char | rule-token | rule-name | rule-brace | rule-ast ;
rule-repeat   := rule-except | 'EXPR_REPEAT' '{' rule-except ('ZERO_OR_ONE' | 'ZERO_OR_MORE' | 'ONE_OR_MORE' | 'MIN_MAX' '{' number number? '}') '}' ;
rule-sequence := rule-trim | 'EXPR_SEQUENCE' '{' rule-trim rule-trim+ '}' ;
rule-trim     := rule-repeat | 'EXPR_TRIM_RIGHT' '{' rule-repeat '}' | 'EXPR_TRIM_LEFT' '{' (rule-repeat | 'EXPR_TRIM_RIGHT' '{' rule-repeat '}') '}' ;

rule-char     := char-range | char-set | char-set-inv | char-eof ;

char-range    := char-one | 'CHAR_RANGE' '{' char-one char-one '}' ;

char-one      := 'CHAR_ONE' '{' char-esc '}' ;
char-set      := 'CHAR_SET' '{' (char-esc | 'CHAR_ONE')+ '}' ;
char-set-inv  := 'CHAR_SET_INV' '{' (char-esc | 'CHAR_ONE')+ '}' ;
rule-token    := 'EXPR_TOKEN' '{' char-esc* '}' | 'EXPR_TOKEN_NOCASE' '{' 'EXPR_TOKEN' '{' char-esc* '}' '}' ;

char-eof      := 'CHAR_EOF' ;
char-esc      := 'CHAR_ESC' ;
number        := 'NUMBER' ;
rule-name     := 'EXPR_SUBRULE' ;

See also:

Optimize output by removing redundant grammar elements?