time to test

2026-08-02 19:04:52 -06:00
parent a228c0c59c
commit 3779848355
5 changed files with 310 additions and 152 deletions
@@ -2,12 +2,18 @@

 namespace spider::asm_ebnf {

+    // Token Factory
+
+    static TokenFactory tf;
+
+    // Char Functions
+
    bool isUTF8Alpha(u32 ch) {
-        return false;
+        return (u32('a') <= ch && ch <= u32('z')) || (u32('A') <= ch && ch <= u32('Z'));
    }

    bool isWhithespaceCharNotCrLf(u32 ch) {
-        return false;
+        return ch == u32(' ');
    }

    bool isUTF8CharNotCrLf(u32 ch) {
@@ -22,37 +28,117 @@ namespace spider::asm_ebnf {
        return ch != u32('"');
    }

-    LitToken numbers[] = {
-        "0",
-        "1","2","3",
-        "4","5","6",
-        "7","8","9",
-    };
-    
-    LitToken hex_digits[][2] = {
-        {"A", "a"},
-        {"B", "b"},
-        {"C", "c"},
-        {"D", "d"},
-        {"E", "e"},
-        {"F", "f"},
-    };
+    // (* Characters & Basic Predicates *)
+    static const Token* letter = tf.fn(isUTF8Alpha);
+    static const Token* digit = tf.choice("0123456789");
+    static const Token* alpha_num_char = tf.choice({ letter, digit });

-    LitToken new_line[] = { "\r\n", "\r", "\n" };
+    static const Token* hex_digit = tf.choice("0123456789ABCDEFabcdef");
+    static const Token* octal_digit = tf.choice("01234567");
+    static const Token* binary_digit = tf.choice("01");

-    LitToken symbols[] = {
-        "\\", "\'", "\"",
-        "_" , ";" , "(" ,
-        "#",
-        "$" , "." , "+" , "-", ",", ")", "@", ":",
-    };
+    static const Token* ws_char = tf.fn(isWhithespaceCharNotCrLf);
+    static const Token* ws_optional = tf.rep(ws_char);
+    static const Token* whitespace = tf.seq({ ws_char, tf.rep(ws_char) });
+    static const Token* newline = tf.choice({ tf["\r\n"], tf["\r"], tf["\n"] });
+    static const Token* utf8_char = tf.fn(isUTF8CharNotCrLf);

-    LitToken lit_letter[] = {
-        "x", "c", "b"
-    };
+    static const Token* char_escape = tf.seq({ tf["\\"], utf8_char });
+    static const Token* char_content = tf.choice({ char_escape, tf.fn(isUTF8CharLitCont) });
+    static const Token* char_lit = tf.seq({ tf["'"], char_content, tf["'"] });

-    LitToken type_letter[] = {
-        "B", "S", "I", "L", "F", "D"
-    };
+    static const Token* string_char = tf.choice({ char_escape, tf.fn(isUTF8StringLitCont) });
+    static const Token* string_lit = tf.seq({ tf["\""], tf.rep(string_char), tf["\""] });
+
+    // (* Literals *)
+    static const Token* identifier = tf.seq({
+        tf.choice({ letter, tf["_"] }),
+        tf.rep(tf.choice({ alpha_num_char, tf["_"] }))
+    });
+
+    static const Token* comment = tf.seq({ tf[";"], tf.rep(utf8_char) });
+
+    static const Token* sign = tf.choice("+-");
+    static const Token* exponent_marker = tf.choice("eE");
+    static const Token* exponent = tf.seq({ exponent_marker, tf.opt(sign), digit, tf.rep(digit) });
+
+    static const Token* decimal_lit = tf.seq({
+        tf.opt(sign),
+        digit,
+        tf.rep(digit),
+        tf.opt(tf.choice("BSIL"))
+    });
+
+    static const Token* float_lit = tf.seq({
+        tf.opt(sign),
+        tf.choice({
+            tf.seq({ digit, tf.rep(digit), tf["."], digit, tf.rep(digit), tf.opt(exponent) }),
+            tf.seq({ tf["."], digit, tf.rep(digit), tf.opt(exponent) }),
+            tf.seq({ digit, tf.rep(digit), exponent })
+        }),
+        tf.opt(tf.choice("FD"))
+    });
+
+    static const Token* hex_lit = tf.seq({ tf.opt(sign), tf["0x"], hex_digit, tf.rep(hex_digit) });
+    static const Token* octal_lit = tf.seq({ tf.opt(sign), tf["0c"], octal_digit, tf.rep(octal_digit) });
+    static const Token* binary_lit = tf.seq({ tf.opt(sign), tf["0b"], binary_digit, tf.rep(binary_digit) });
+
+    static const Token* literal = tf.choice({ float_lit, decimal_lit, hex_lit, octal_lit, binary_lit, string_lit, char_lit });
+    static const Token* literal_cast = tf.seq({ tf.choice("BSILFD"), ws_optional, tf["("], ws_optional, literal, ws_optional, tf[")"] });
+    static const Token* literal_decl = tf.choice({ literal, literal_cast });
+
+    // (* Operands *)
+    static const Token* register_tok = tf.seq({ tf["R"], alpha_num_char });
+
+    static const Token* addrm_ind = tf.seq({ tf["["], ws_optional, literal_decl, ws_optional, tf["]"] });
+    static const Token* addrm_ptr = tf.seq({ tf["["], ws_optional, register_tok, ws_optional, tf["]"] });
+
+    static const Token* addrm_idx = tf.seq({
+        tf["["], ws_optional, register_tok, ws_optional,
+        tf["+"], ws_optional, literal_decl, ws_optional, tf["]"]
+    });
+
+    static const Token* addrm_sca = tf.seq({
+        tf["["], ws_optional, register_tok, ws_optional,
+        tf["+"], ws_optional, register_tok, ws_optional,
+        tf["*"], ws_optional, literal_decl, ws_optional, tf["]"]
+    });
+
+    static const Token* addrm_dis = tf.seq({
+        tf["["], ws_optional, register_tok, ws_optional,
+        tf["+"], ws_optional, register_tok, ws_optional,
+        tf["*"], ws_optional, literal_decl, ws_optional,
+        tf["+"], ws_optional, literal_decl, ws_optional, tf["]"]
+    });
+
+    static const Token* addr_modes = tf.choice({ addrm_dis, addrm_sca, addrm_idx, addrm_ptr, addrm_ind });
+    static const Token* operand = tf.choice({ register_tok, identifier, literal_decl, addr_modes });
+
+    // (* Generalized Instructions *)
+
+    static const Token* opcode = tf.seq({ letter, tf.rep(alpha_num_char) });
+    static const Token* operand_list = tf.seq({ operand, tf.rep(tf.seq({ tf[","], ws_optional, operand })) });
+    static const Token* instruction = tf.seq({opcode, tf.opt(tf.seq({ whitespace, operand_list })) });
+
+    // (* Added Preprocessor, Annotation *)
+
+    static const Token* annotation_named = tf.seq({ identifier, ws_optional, tf["="], ws_optional, literal_decl });
+    static const Token* annotation_arg = tf.choice({ annotation_named, literal_decl });
+    static const Token* annotation_args = tf.seq({ annotation_arg, tf.rep(tf.seq({ ws_optional, tf[","], ws_optional, annotation_arg })) });
+    static const Token* annotation_pars = tf.seq({ tf["("], ws_optional, annotation_args, ws_optional, tf[")"] });
+    static const Token* annotation = tf.seq({ tf["@"], identifier, tf.opt(annotation_pars) });
+
+    static const Token* preprocessor_val = tf.choice({ identifier, string_lit });
+    static const Token* preprocessor = tf.seq({ tf["#"], identifier, whitespace, preprocessor_val });
+
+    // (* Line Structure & Program *)
+
+    static const Token* label = tf.seq({ identifier, tf[":"] });
+    static const Token* line_label = tf.seq({ label, tf.opt(tf.seq({ whitespace, instruction })) });
+    static const Token* line_annotation = tf.seq({ annotation, tf.opt(tf.seq({ whitespace, instruction })) });
+    static const Token* line_content = tf.choice({ preprocessor, line_annotation, line_label, instruction });
+    static const Token* line = tf.seq({ ws_optional, tf.opt(line_content), ws_optional, tf.opt(comment), newline });
+    static const Token* line_last = tf.seq({ ws_optional, tf.opt(line_content), ws_optional, tf.opt(comment) });
+    static const Token* program = tf.seq({ tf.rep(line), tf.opt(line_last) });

 }
@@ -6,4 +6,6 @@ namespace spider::asm_ebnf {

    extern LitToken letter;

+    void createTokens();
+
 }