parser-lr 0.4.0 → 0.5.0

package/grammars/fixtures/build-optional/build-optional.grammar

@@ -0,0 +1,32 @@
+name "build-optional-repro" ;
+
+tokens
+    kw_cls = /CLS/i ;
+    kw_with = /WITH/i ;
+    ident = /[a-z]+/ ;
+
+skip
+    whitespace = /[ \t\r\n]+/ ;
+
+start program_line ;
+
+grammar
+    program_line =
+        #statement [stmt]:cls_stmt
+      ;
+
+    cls_stmt =
+        kw_cls [ kw_with ident ]
+      ;
+
+ast
+    cls_stmt =
+        #main kw_cls [withKw]:kw_with [target]:ident
+      ;
+
+transform
+    cls_stmt ->
+        #main cls_stmt.#main(kw_cls, withKw, target) ;
+
+    program_line ->
+        #statement pass(stmt) ;

package/grammars/fixtures/build-optional/input.txt

@@ -0,0 +1 @@
+CLS

package/grammars/fixtures/comparison-op/comparison-op.grammar

@@ -0,0 +1,28 @@
+name "comparison-op-repro" ;
+
+tokens
+    kw_case = /CASE/i ;
+    kw_is = /IS/i ;
+    less = /</ ;
+    greater = />/ ;
+    number = /[0-9]+/ ;
+
+skip
+    whitespace = /[ \t\r\n]+/ ;
+
+start case_selector ;
+
+grammar
+    case_selector =
+        #relational kw_case kw_is comparison_op postfix_expr
+      | #value postfix_expr
+      ;
+
+    comparison_op =
+        #less less
+      | #greater greater
+      ;
+
+    postfix_expr =
+        number
+      ;

package/grammars/fixtures/comparison-op/input.txt

@@ -0,0 +1 @@
+CASE IS < 0

package/grammars/fixtures/flatten-repeat/flatten-repeat.grammar

@@ -0,0 +1,28 @@
+name "flatten-repeat-repro" ;
+
+tokens
+    ident = /[a-z]+/ ;
+    comma = /,/ ;
+
+skip
+    whitespace = /[ \t\r\n]+/ ;
+
+start list ;
+
+grammar
+    list =
+        #list [first]:item { comma [rest]:item }
+      ;
+
+    item =
+        ident
+      ;
+
+ast
+    list =
+        #list [first]:item { [rest]:item }
+      ;
+
+transform
+    list ->
+        #list flatten(list.#list, first, list$repeat_0) ;

package/grammars/fixtures/flatten-repeat/input.txt

@@ -0,0 +1 @@
+a,b,c

package/grammars/fixtures/help-topic/help-topic.grammar

@@ -0,0 +1,28 @@
+name "help-topic-repro" ;
+
+tokens
+    kw_help = /HELP/i ;
+    kw_print = /PRINT/i ;
+
+skip
+    whitespace = /[ \t\r\n]+/ ;
+
+start program_line ;
+
+grammar
+    program_line =
+        #statement [stmt]:help_stmt
+      ;
+
+    help_stmt =
+        kw_help help_topic
+      ;
+
+    help_topic =
+        #keyword kw_print
+      | #punct kw_print
+      ;
+
+transform
+    program_line ->
+        #statement pass(stmt) ;

package/grammars/fixtures/help-topic/input.txt

@@ -0,0 +1 @@
+HELP PRINT

package/grammars/fixtures/let-target/input.txt

@@ -0,0 +1 @@
+LET x% = 1

package/grammars/fixtures/let-target/let-target.grammar

@@ -0,0 +1,33 @@
+name "let-target-repro" ;
+
+tokens
+    kw_let = /LET/i ;
+    identifier = /[a-z][a-z0-9%]*/i ;
+    equal = /=/ ;
+    number = /[0-9]+/ ;
+
+skip
+    whitespace = /[ \t\r\n]+/ ;
+
+start program_line ;
+
+grammar
+    program_line =
+        #statement [stmt]:let_stmt
+      ;
+
+    let_stmt =
+        kw_let let_target equal expr
+      ;
+
+    let_target =
+        identifier
+      ;
+
+    expr =
+        number
+      ;
+
+transform
+    program_line ->
+        #statement pass(stmt) ;

package/grammars/fixtures/pass-preserves-production/bare-terminal/bare-terminal.grammar

@@ -0,0 +1,24 @@
+name "pass-collapse-repro" ;
+
+tokens
+    kw_cls = /CLS/i ;
+    kw_with = /WITH/i ;
+    ident = /[a-z]+/ ;
+
+skip
+    whitespace = /[ \t\r\n]+/ ;
+
+start program_line ;
+
+grammar
+    program_line =
+        #statement [stmt]:cls_stmt
+      ;
+
+    cls_stmt =
+        kw_cls [ kw_with ident ]
+      ;
+
+transform
+    program_line ->
+        #statement pass(stmt) ;

package/grammars/fixtures/pass-preserves-production/bare-terminal/input.txt

@@ -0,0 +1 @@
+CLS

package/grammars/fixtures/pass-preserves-production/labeled-variant/input-end-if.txt

@@ -0,0 +1 @@
+END IF

package/grammars/fixtures/pass-preserves-production/labeled-variant/input-end.txt

@@ -0,0 +1 @@
+END

package/grammars/fixtures/pass-preserves-production/labeled-variant/labeled-variant.grammar

@@ -0,0 +1,24 @@
+name "labeled-variant-repro" ;
+
+tokens
+    kw_end = /END/i ;
+    kw_if = /IF/i ;
+
+skip
+    whitespace = /[ \t\r\n]+/ ;
+
+start program_line ;
+
+grammar
+    program_line =
+        #statement [stmt]:end_stmt
+      ;
+
+    end_stmt =
+        #program kw_end
+      | #if kw_end kw_if
+      ;
+
+transform
+    program_line ->
+        #statement pass(stmt) ;

package/grammars/fixtures/pass-preserves-production/optional-absent/input.txt

@@ -0,0 +1 @@
+CLS

package/grammars/fixtures/pass-preserves-production/optional-absent/optional-absent.grammar

@@ -0,0 +1,24 @@
+name "optional-absent-repro" ;
+
+tokens
+    kw_cls = /CLS/i ;
+    kw_with = /WITH/i ;
+    ident = /[a-z]+/ ;
+
+skip
+    whitespace = /[ \t\r\n]+/ ;
+
+start program_line ;
+
+grammar
+    program_line =
+        #statement [stmt]:cls_stmt
+      ;
+
+    cls_stmt =
+        kw_cls [ kw_with ident ]
+      ;
+
+transform
+    program_line ->
+        #statement pass(stmt) ;

package/grammars/fixtures/pass-preserves-production/optional-present/input.txt

@@ -0,0 +1 @@
+CLS WITH foo

package/grammars/fixtures/pass-preserves-production/optional-present/optional-present.grammar

@@ -0,0 +1,24 @@
+name "optional-present-repro" ;
+
+tokens
+    kw_cls = /CLS/i ;
+    kw_with = /WITH/i ;
+    ident = /[a-z]+/ ;
+
+skip
+    whitespace = /[ \t\r\n]+/ ;
+
+start program_line ;
+
+grammar
+    program_line =
+        #statement [stmt]:cls_stmt
+      ;
+
+    cls_stmt =
+        kw_cls [ kw_with ident ]
+      ;
+
+transform
+    program_line ->
+        #statement pass(stmt) ;

package/grammars/grammar.grammar

@@ -0,0 +1,428 @@
+// Meta-grammar for parser-lr grammar files (.grammar).
+// Dogfood target: this file should parse itself once lexer, table build, and parse are wired.
+//
+// Calc example (parse + ast + transform):
+//
+//   grammar
+//       expr = term { #add plus term | #sub minus term } ;
+//       term = factor { #mul star factor | #div slash factor } ;
+//       factor = #number number | #group lpar expr rpar ;
+//
+//   ast
+//       expr = #binary [left]:expr [operator]:operator [right]:expr
+//            | #literal number
+//            | #grouped expr ;
+//       operator = #plus plus | #minus minus | #star star | #slash slash ;
+//
+//   transform
+//       expr ->
+//           #add fold-left(expr.#binary, left, operator, right)
+//         | #sub fold-left(expr.#binary, left, operator, right)
+//         | #term pass(term) ;
+//       term ->
+//           #mul fold-left(expr.#binary, left, operator, right)
+//         | #div fold-left(expr.#binary, left, operator, right)
+//         | #factor pass(factor) ;
+//       factor ->
+//           #number expr.#literal(number)
+//         | #group pass(expr) ;
+
+name "grammar" ;
+
+tokens
+    name_kw = /name/ ;
+    tokens_kw = /tokens/ ;
+    skip_kw = /skip/ ;
+    states_kw = /states/ ;
+    start_kw = /start/ ;
+    grammar_kw = /grammar/ ;
+    ast_kw = /ast/ ;
+    transform_kw = /transform/ ;
+    pass_kw = /pass/ ;
+    drop_kw = /drop/ ;
+    fold_left_kw = /fold-left/ ;
+    fold_right_kw = /fold-right/ ;
+    flatten_kw = /flatten/ ;
+
+    identifier = /[A-Za-z_][A-Za-z0-9_$]*/ ;
+
+    equal = /=/ ;
+    semicolon = /;/ ;
+    comma = /,/ ;
+    bar = /\|/ ;
+    hash = /#/ ;
+    arrow = /->/ ;
+    colon = /:/ ;
+    dot = /\./ ;
+
+    lpar = /\(/ ;
+    rpar = /\)/ ;
+    lbracket = /\[/ ;
+    rbracket = /\]/ ;
+    lbrace = /\{/ ;
+    rbrace = /\}/ ;
+
+    // Lexer: / pattern / then zero or more flag letters from g i m s u y
+    regex_literal = /\/(\\.|[^\\\/])+\/[gimsuy]*/ ;
+    string_literal = /"(?:\\.|[^"\\])*"/ ;
+
+skip
+    whitespace = /[ \t\r\n]+/ ;
+    comment = /\/\/[^\n\r]*/ ;
+
+start grammar_file ;
+
+grammar
+    grammar_file =
+        name_decl
+        { section }
+        start_decl
+        grammar_section
+        [ ast_section ]
+        [ transform_section ]
+    ;
+
+    name_decl =
+        name_kw grammar_name semicolon
+    ;
+
+    grammar_name =
+        identifier
+      | string_literal
+    ;
+
+    section =
+        token_section
+      | skip_section
+      | states_section
+    ;
+
+    token_section =
+        tokens_kw token_def { token_def }
+    ;
+
+    skip_section =
+        skip_kw token_def { token_def }
+    ;
+
+    states_section =
+        states_kw state_name { comma state_name } semicolon
+    ;
+
+    token_def =
+        identifier equal regex_literal semicolon
+    ;
+
+    state_name =
+        identifier
+    ;
+
+    start_decl =
+        start_kw identifier semicolon
+    ;
+
+    grammar_section =
+        grammar_kw production { production }
+    ;
+
+    ast_section =
+        ast_kw ast_type { ast_type }
+    ;
+
+    transform_section =
+        transform_kw transform_rule { transform_rule }
+    ;
+
+    production =
+        identifier equal expression semicolon
+    ;
+
+    ast_type =
+        identifier equal expression semicolon
+    ;
+
+    transform_rule =
+        identifier arrow labeled_transform { bar labeled_transform } semicolon
+    ;
+
+    labeled_transform =
+        hash identifier transform_expr
+    ;
+
+    transform_expr =
+        drop_kw
+      | pass_expr
+      | fold_left_expr
+      | fold_right_expr
+      | flatten_expr
+      | build_expr
+    ;
+
+    pass_expr =
+        pass_kw lpar reference rpar
+    ;
+
+    fold_left_expr =
+        fold_left_kw lpar build_target comma reference_list rpar
+    ;
+
+    fold_right_expr =
+        fold_right_kw lpar build_target comma reference_list rpar
+    ;
+
+    flatten_expr =
+        flatten_kw lpar build_target comma reference comma reference rpar
+    ;
+
+    build_expr =
+        build_target [ lpar reference_list rpar ]
+    ;
+
+    build_target =
+        identifier dot hash identifier
+    ;
+
+    reference_list =
+        reference { comma reference }
+    ;
+
+    reference =
+        identifier
+    ;
+
+    expression =
+        choice
+    ;
+
+    choice =
+        alternative { bar alternative }
+    ;
+
+    alternative =
+        labeled_alternative
+    ;
+
+    labeled_alternative =
+        [ hash identifier ] sequence
+    ;
+
+    sequence =
+        factor { factor }
+    ;
+
+    factor =
+        bracketed
+      | repeat
+      | group
+      | primary
+    ;
+
+    bracketed =
+        lbracket expression rbracket [ colon identifier ]
+    ;
+
+    repeat =
+        lbrace expression rbrace
+    ;
+
+    group =
+        lpar expression rpar
+    ;
+
+    primary =
+        string_literal
+      | identifier
+    ;
+
+ast
+    grammar_file =
+        name_decl
+        { section }
+        start_decl
+        grammar_section
+        [ ast_section ]
+        [ transform_section ]
+    ;
+
+    name_decl =
+        name_kw grammar_name semicolon
+    ;
+
+    grammar_name =
+        identifier
+      | string_literal
+    ;
+
+    section =
+        token_section
+      | skip_section
+      | states_section
+    ;
+
+    token_section =
+        tokens_kw token_def { token_def }
+    ;
+
+    skip_section =
+        skip_kw token_def { token_def }
+    ;
+
+    states_section =
+        states_kw state_name { comma state_name } semicolon
+    ;
+
+    token_def =
+        identifier equal regex_literal semicolon
+    ;
+
+    state_name =
+        identifier
+    ;
+
+    start_decl =
+        start_kw identifier semicolon
+    ;
+
+    grammar_section =
+        grammar_kw production { production }
+    ;
+
+    ast_section =
+        ast_kw ast_type { ast_type }
+    ;
+
+    transform_section =
+        transform_kw transform_rule { transform_rule }
+    ;
+
+    production =
+        identifier equal expression semicolon
+    ;
+
+    ast_type =
+        identifier equal expression semicolon
+    ;
+
+    transform_rule =
+        identifier arrow labeled_transform { bar labeled_transform } semicolon
+    ;
+
+    labeled_transform =
+        hash identifier transform_expr
+    ;
+
+    transform_expr =
+        drop_kw
+      | pass_expr
+      | fold_left_expr
+      | fold_right_expr
+      | flatten_expr
+      | build_expr
+    ;
+
+    pass_expr =
+        pass_kw lpar reference rpar
+    ;
+
+    fold_left_expr =
+        fold_left_kw lpar build_target comma reference_list rpar
+    ;
+
+    fold_right_expr =
+        fold_right_kw lpar build_target comma reference_list rpar
+    ;
+
+    flatten_expr =
+        flatten_kw lpar build_target comma reference comma reference rpar
+    ;
+
+    build_expr =
+        build_target [ lpar reference_list rpar ]
+    ;
+
+    build_target =
+        identifier dot hash identifier
+    ;
+
+    reference_list =
+        reference { comma reference }
+    ;
+
+    reference =
+        identifier
+    ;
+
+    expression =
+        choice
+    ;
+
+    choice =
+        alternative { bar alternative }
+    ;
+
+    alternative =
+        labeled_alternative
+    ;
+
+    labeled_alternative =
+        [ hash identifier ] sequence
+    ;
+
+    sequence =
+        factor { factor }
+    ;
+
+    factor =
+        bracketed
+      | repeat
+      | group
+      | primary
+    ;
+
+    bracketed =
+        lbracket expression rbracket [ colon identifier ]
+    ;
+
+    repeat =
+        lbrace expression rbrace
+    ;
+
+    group =
+        lpar expression rpar
+    ;
+
+    primary =
+        string_literal
+      | identifier
+    ;
+
+transform
+    grammar_file$repeat_0 ->
+        #main flatten(grammar_file.#sections, section, grammar_file$repeat_0) ;
+
+    token_section$repeat_1 ->
+        #main flatten(token_section.#definitions, token_def, token_section$repeat_1) ;
+
+    skip_section$repeat_2 ->
+        #main flatten(skip_section.#definitions, token_def, skip_section$repeat_2) ;
+
+    grammar_section$repeat_4 ->
+        #main flatten(grammar_section.#productions, production, grammar_section$repeat_4) ;
+
+    ast_section$repeat_5 ->
+        #main flatten(ast_section.#types, ast_type, ast_section$repeat_5) ;
+
+    transform_section$repeat_6 ->
+        #main flatten(transform_section.#rules, transform_rule, transform_section$repeat_6) ;
+
+    transform_rule$repeat_7 ->
+        #main flatten(transform_rule.#alternatives, labeled_transform, transform_rule$repeat_7) ;
+
+    reference_list$repeat_8 ->
+        #main flatten(reference_list.#references, reference, reference_list$repeat_8) ;
+
+    choice$repeat_9 ->
+        #main flatten(choice.#alternatives, alternative, choice$repeat_9) ;
+
+    sequence$repeat_10 ->
+        #main flatten(sequence.#factors, factor, sequence$repeat_10) ;