Grammar railroad diagram

mingodad commented 3 months ago

Would be nice to have an EBNF accpeted by (IPV4) https://rr.red-dove.com/ui or (IPV6) https://www.bottlecaps.de/rr/ui to generate a nice navigable railroad diagram for PointerScript, I started one going through the code in https://github.com/M4GNV5/PointerScript/blob/master/parser/ast.c (see bellow) but it has many flaws due to my interpretation so far.

//
// EBNF to be viewd at
//    (IPV6) https://www.bottlecaps.de/rr/ui
//    (IPV4) https://rr.red-dove.com/ui
//
// Copy and paste this at one of the urls shown above in the 'Edit Grammar' tab
// then click the 'View Diagram' tab.
//

ptrs_parse ::=
    shebang_opt parseStmtList

parseStmtList ::=
    parseStatement+

parseStatement ::=
    parseBody
    | "const" readIdentifier parseExpression ';'
    | "var" readIdentifier type_opt var_assing_opt ';'
    | parseImport
    | "return" parseExpression ';'
    | "break" ';'
    | "continue" ';'
    | "delete" parseExpression ';'
    | "throw" parseExpression ';'
    | "try" parseBody parseCatch_opt parseFinally_opt
    | "function" readIdentifier parseFunctionInto
    | parseStruct
    | "if" '(' parseExpression ')' parseBody
    | "if" '(' parseExpression ')' parseBody "else" parseBody
    | parseSwitchCase
    | "loop" parseBody
    | "while" '(' parseExpression ')' parseBody
    | "do" parseScopelessBody "while" '(' parseExpression ')' ';'
    | "foreach" '(' readIdentifier (',' readIdentifier)* "in" parseExpression ')' parseScopelessBody
    | "for" '(' parseStatement parseExpression_opt ';' parseExpression ')' parseScopelessBody
    | parseExpression ';'

parseCatch_opt ::=
    /*empty*/
    | "catch" parseArgumentDefinitionList parseBody

parseFinally_opt ::=
    "finally" parseBody
    | "finally" '(' readIdentifier ')' parseBody

parseImport ::=
    "import"  importElement_oom ';'

importElement_oom ::=
    importElement
    | importElement_oom ',' importElement

importElement ::=
    readIdentifier '*'? (':' readNativeType)? ("as" readIdentifier)? ("from" readString)?

parseBody ::=
    parseScopelessBody

parseScopelessBody ::=
    '{' parseStmtList '}'

parseFunctionInto ::=
    parseArgumentDefinitionList parseFunctionBody

parseArgumentDefinitionList ::=
    '(' argumentDefinition_zom ')' parseOptionalTyping

argumentDefinition_zom ::=
    /*empty*/
    | "..."
    | argumentDefinition_oom
    | argumentDefinition_oom ',' "..."

argumentDefinition_oom ::=
    argumentDefinition
    | argumentDefinition_zom ',' argumentDefinition

argumentDefinition ::=
    '_'
    | readIdentifier parseOptionalTyping assignExpression_opt

assignExpression_opt ::=
    '=' Expression

parseOptionalTyping ::=
    /*empty*/
    | parseTyping

parseTyping ::=
    ':' readTypeName
    | ':' readNativeType
    | ':' readNativeType parseArrayTyping
    | ':' readIdentifier

parseFunctionBody ::=
    parseScopelessBody

parseStruct ::=
    "struct" readIdentifier '{' parseStructElements '}' ';'

parseStructElements ::=
    "static"? parseStructElement
    | parseStructElements "static"? parseStructElement

parseStructElement ::=
    "operator" '&' "this" '[' readIdentifier createParameterList parseOptionalTyping ']' ('=' readIdentifier createParameterList)?
    | "operator" '&' "this"  parseArgumentDefinitionList parseOptionalTyping ')'
    | "sizeof" "this"
    | "foreach" '(' readIdentifier ',' readIdentifier ')' "in" "this" createParameterList
    | "cast" '<' ("int"|"float"|"string") '>' "this"
    | readIdentifier "in" "this" createParameterList parseFunctionBody
    | readIdentifier
    | "constructor" parseFunction
    | "destructor" parseFunction
    | ("private"|"public"|"internal") ("get"|"set")? readIdentifier

parseSwitchCase ::=
    "switch" '(' parseExpression ')' '{' switchCaseElements '}'

switchCaseElements ::=
    "default" ':' parseStatement
    | switchCaseElement
    | switchCaseElements switchCaseElement

switchCaseElement ::=
    "case" switchCaseExpression (',' switchCaseExpression)* ':' parseStatement

switchCaseExpression ::=
    parseExpression
    | parseExpression ".." parseExpression

var_assing_opt ::=
    '=' parseExpression

parseExpression_opt ::=
    /*empty*/
    | parseExpression

parseExpression ::=
    parseUnaryExpr
    | parseBinaryExpr

parseUnaryExpr ::=
    readPrefixOperator parseUnaryExpr parseNew
    | readPrefixOperator parseUnaryExpr parseType
    | readPrefixOperator parseUnaryExpr parseSizeof
    | readPrefixOperator parseUnaryExpr "var" '(' ')'
    | readPrefixOperator parseUnaryExpr "as" '<' readTypeName '>'
    | readPrefixOperator parseUnaryExpr "cast" '<' ("int"|"float"|"string") '>'
    | readPrefixOperator parseUnaryExpr "function" parseFunctionInto
    | readPrefixOperator parseUnaryExpr parseMap

parseBinaryExpr ::=
    // >> and << need to be before > and < or we will always lookahead greater / less
    | parseBinaryExpr ">>" parseExpression //12, false, &ptrs_ast_vtable_op_sshr}, //signed shift right
    | parseBinaryExpr ">>" parseExpression //12, false, &ptrs_ast_vtable_op_ushr}, //unsigned shift right
    | parseBinaryExpr "<<" parseExpression //12, false, &ptrs_ast_vtable_op_shl}, //shift left

    | parseBinaryExpr "===" parseExpression //9, false, &ptrs_ast_vtable_op_typeequal},
    | parseBinaryExpr "!==" parseExpression //9, false, &ptrs_ast_vtable_op_typeinequal},

    | parseBinaryExpr "==" parseExpression //9, false, &ptrs_ast_vtable_op_equal},
    | parseBinaryExpr "!=" parseExpression //9, false, &ptrs_ast_vtable_op_inequal},
    | parseBinaryExpr "<=" parseExpression //10, false, &ptrs_ast_vtable_op_lessequal},
    | parseBinaryExpr ">=" parseExpression //10, false, &ptrs_ast_vtable_op_greaterequal},
    | parseBinaryExpr "<" parseExpression //10, false, &ptrs_ast_vtable_op_less},
    | parseBinaryExpr ">" parseExpression //10, false, &ptrs_ast_vtable_op_greater},

    | parseBinaryExpr "=" parseExpression //1, true, &ptrs_ast_vtable_op_assign},
    | parseBinaryExpr "+=" parseExpression //1, true, &ptrs_ast_vtable_op_add},
    | parseBinaryExpr "-=" parseExpression //1, true, &ptrs_ast_vtable_op_sub},
    | parseBinaryExpr "*=" parseExpression //1, true, &ptrs_ast_vtable_op_mul},
    | parseBinaryExpr "/=" parseExpression //1, true, &ptrs_ast_vtable_op_div},
    | parseBinaryExpr "%=" parseExpression // 1, true, &ptrs_ast_vtable_op_mod},
    | parseBinaryExpr ">>=" parseExpression //1, true, &ptrs_ast_vtable_op_sshr},
    | parseBinaryExpr "<<=" parseExpression //1, true, &ptrs_ast_vtable_op_shl},
    | parseBinaryExpr "&=" parseExpression //1, true, &ptrs_ast_vtable_op_and},
    | parseBinaryExpr "^=" parseExpression //1, true, &ptrs_ast_vtable_op_xor},
    | parseBinaryExpr "|=" parseExpression //1, true, &ptrs_ast_vtable_op_or},

    //| parseBinaryExpr "?" parseExpression // 2, true, &ptrs_ast_vtable_op_ternary},
    //| parseBinaryExpr ":" parseExpression //-1, true, &ptrs_ast_vtable_op_ternary},
    | parseBinaryExpr "?" parseExpression ":" parseExpression

    | parseBinaryExpr "instanceof" parseExpression //11, false, &ptrs_ast_vtable_op_instanceof},
    | parseBinaryExpr "in" parseExpression //11, false, &ptrs_ast_vtable_op_in},

    | parseBinaryExpr "||" parseExpression //3, false, &ptrs_ast_vtable_op_logicor},
    | parseBinaryExpr "^^" parseExpression //4, false, &ptrs_ast_vtable_op_logicxor},
    | parseBinaryExpr "&&" parseExpression //5, false, &ptrs_ast_vtable_op_logicand},

    | parseBinaryExpr "|" parseExpression //6 parseExpression //false, &ptrs_ast_vtable_op_or},
    | parseBinaryExpr "^" parseExpression //7, false, &ptrs_ast_vtable_op_xor},
    | parseBinaryExpr "&" parseExpression //8, false, &ptrs_ast_vtable_op_and},

    | parseBinaryExpr "+" parseExpression //13, false, &ptrs_ast_vtable_op_add},
    | parseBinaryExpr "-" parseExpression //13, false, &ptrs_ast_vtable_op_sub},

    | parseBinaryExpr "*" parseExpression //14, false, &ptrs_ast_vtable_op_mul},
    | parseBinaryExpr "/" parseExpression //14, false, &ptrs_ast_vtable_op_div},
    | parseBinaryExpr "%" parseExpression //14, false, &ptrs_ast_vtable_op_mod}

parseNew_opt ::=
    /*empty*/
    | parseNew

parseNew ::=
    "new"

parseType_opt ::=
    /*empty*/
    | parseType

parseType ::=
    "type" '<' readTypeName '>'

parseSizeof ::=
    "sizeof" '(' readNativeType ')'

parseMap ::=
    "map_stack"
    | "map"

type_opt ::=
    ':' readNativeType parseArrayTyping array_init_opt

array_init_opt ::=
    '=' '[' parseExpressionList ']'

shebang_opt ::=
    "#!".*

comments ::=
    "//".*
    | "/*".*?"*/"

constants ::=
    "true"
    | "false"
    | "null"
    | "undefined"
    | "NaN"
    | "Infinity"
    | "PI"
    | "E"

typeNames ::=
    "undefined"
    | "int"
    | "float"
    | "pointer"
    | "struct"
    | "function"

readTypeName ::=
    typeNames

readNativeType ::=

readIdentifier ::=
    [a-zA-Z_][a-zA-Z0-9_]*

M4GNV5 commented 3 months ago

I had some time to kill in the TGV from Paris to Munich, so I added your grammar file to the repo and made some improvements. It is still far from finished though: https://github.com/M4GNV5/PointerScript/blob/master/grammar.ebnf

Putting this into the railroad generator does produce some cool looking diagrams for Expression though.

Maybe at some point the ebnf grammar & railroad diagrams can be used for the language doc

mingodad commented 3 months ago

Thank you for reply ! Noticed that the EBNF accepted by railroad diagram tool doesn't accept escaped chars:

"\"" .* "\""

Can be :

'"' .* '"'
["] .* ["]

M4GNV5 / PointerScript

Grammar railroad diagram #8