From b5cc10cff15797bc9f89724ab53ac7d296fbbc0c Mon Sep 17 00:00:00 2001
From: Daniel Weipert <code@drogueronin.de>
Date: Fri, 25 Aug 2023 16:24:59 +0200
Subject: trying ply

---
 parse.py | 197 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 197 insertions(+)
 create mode 100755 parse.py

(limited to 'parse.py')

diff --git a/parse.py b/parse.py
new file mode 100755
index 0000000..17e3be1
--- /dev/null
+++ b/parse.py
@@ -0,0 +1,197 @@
+#!/bin/env python3
+
+
+import sys
+import ply.lex as lex
+import ply.yacc as yacc
+
+
+input = open(sys.argv[1]).read()
+
+
+reserved = {
+    "function": "FUNCTION",
+    "if": "IF",
+    "else": "ELSE",
+    "return": "RETURN",
+    "const": "CONST",
+    "var": "VAR",
+
+
+    "integer": "TYPE_INTEGER",
+    "string": "TYPE_STRING",
+    "void": "TYPE_VOID",
+}
+
+tokens = [
+    "OPERATOR_PLUS",
+    "OPERATOR_MINUS",
+    "OPERATOR_MULTIPLY",
+    "OPERATOR_DIVIDE",
+    "ASSIGN",
+
+    "OPERATOR_PIPE",
+    "OPERATOR_PIPE_REPLACEMENT",
+
+
+    "PARENTHESIS_LEFT",
+    "PARENTHESIS_RIGHT",
+    "BRACE_LEFT",
+    "BRACE_RIGHT",
+    "BRACKET_LEFT",
+    "BRACKET_RIGHT",
+
+
+    "BIT_OR",
+
+
+    "NAMESPACE_ACCESSOR",
+
+
+    "COMMA",
+    "COLON",
+    "SEMICOLON",
+
+
+    "IDENTIFIER",
+    "NUMBER",
+    "STRING",
+    "COMMENT",
+] + list(reserved.values())
+
+
+t_OPERATOR_PLUS = r'\+'
+t_OPERATOR_MINUS = r'-'
+t_OPERATOR_MULTIPLY = r'\*'
+t_OPERATOR_DIVIDE = r'/'
+t_ASSIGN = '='
+
+t_OPERATOR_PIPE = r'\|>'
+t_OPERATOR_PIPE_REPLACEMENT = r'\$'
+
+
+t_PARENTHESIS_LEFT = r'\('
+t_PARENTHESIS_RIGHT = r'\)'
+t_BRACE_LEFT = r'\['
+t_BRACE_RIGHT = r'\]'
+t_BRACKET_LEFT = r'{'
+t_BRACKET_RIGHT = r'}'
+
+
+t_BIT_OR = r'\|'
+
+
+t_NAMESPACE_ACCESSOR = r'\.'
+
+
+t_COMMA = r','
+t_COLON = r':'
+t_SEMICOLON = r';'
+
+
+def t_IDENTIFIER(t):
+    r'[a-zA-Z][a-zA-Z0-9_]*'
+
+    t.type = reserved.get(t.value, 'IDENTIFIER')
+
+    return t
+
+def t_NUMBER(t):
+    r'[0-9]+'
+
+    t.value = int(t.value)
+
+    return t
+
+def t_STRING(t):
+    r'(".+"|\'.*\')'
+
+    return t
+
+def t_COMMENT(t):
+    r'//.*'
+
+    pass
+
+
+def t_newline(t):
+    r'\n+'
+
+    t.lexer.lineno += len(t.value)
+
+t_ignore = ' \t'
+
+def t_error(t):
+    print("undefined: '%s'" % t.value[0])
+    t.lexer.skip(1)
+
+
+lexer = lex.lex()
+lexer.input(input)
+
+lineno = 0
+for token in lexer:
+    if token.lineno != lineno:
+        lineno = token.lineno
+        print("\nLine %s:" % token.lineno)
+
+    print('%s: "%s" --' % (token.type, token.value), end=' ')
+print("\n")
+
+
+precedence = (
+    ('left', 'OPERATOR_PLUS', 'OPERATOR_MINUS'),
+    ('left', 'OPERATOR_MULTIPLY', 'OPERATOR_DIVIDE'),
+)
+
+
+def p_statements(p):
+    '''statements : statement
+                  | statement statements'''
+    pass
+
+def p_statement(p):
+    '''statement : variable_declaration_statement
+                 | variable_reassignment_statement'''
+    pass
+
+# def p_function(p):
+#     '''function : FUNCTION IDENTIFIER PARENTHESIS_LEFT function_parameters PARENTHESIS_RIGHT COLON return_type BRACKET_LEFT statements BRACKET_RIGHT'''
+#     pass
+
+def p_variable_declaration_statement(p):
+    '''variable_declaration_statement : variable_declarator IDENTIFIER COLON variable_type ASSIGN expression SEMICOLON'''
+    pass
+
+def p_variable_reassignment_statement(p):
+    '''variable_reassignment_statement : IDENTIFIER ASSIGN expression SEMICOLON'''
+    pass
+
+def p_variable_declarator(p):
+    '''variable_declarator : CONST
+                           | VAR'''
+    pass
+
+def p_variable_type(p):
+    '''variable_type : TYPE_INTEGER
+                     | TYPE_STRING'''
+    pass
+
+def p_expression(p):
+    '''expression : IDENTIFIER
+                  | NUMBER
+                  | STRING
+                  | expression OPERATOR_PLUS expression
+                  | expression OPERATOR_MINUS expression
+                  | expression OPERATOR_MULTIPLY expression
+                  | expression OPERATOR_DIVIDE expression'''
+    pass
+
+
+def p_error(p):
+    print("Syntax error in input!", p)
+
+
+parser = yacc.yacc()
+result = parser.parse(input)
+print(result)
-- 
cgit v1.2.3