complete a crude parse

.gitignore

@@ -0,0 +1,2 @@
+.vscode
+__pycache__

py/parse.py

@@ -0,0 +1,60 @@
+from representation import Pair
+from split import bulid_exp_list
+
+class Stack:
+    def __init__(self, line=None):
+        self.line = line or []
+        self.index = len(self.line)
+
+    def pop(self):
+        self.index -= 1
+        return self.line.pop()
+
+    def top(self):
+        return self.line[self.index-1]
+
+    def stash(self, val):
+        self.index += 1
+        self.line.append(val)
+
+    def not_empty(self):
+        return self.index > 0
+
+def parse(exp_stack):
+    buffer = Stack()
+    while exp_stack.not_empty():
+        val = exp_stack.pop()
+        if val == ')':
+            buffer.stash(val)
+        elif val == '(':
+            car = buffer.pop()
+            buffer.pop()
+            if buffer.not_empty():
+                if buffer.top() != ')':
+                    cdr = buffer.pop()
+                    val = Pair(car, cdr)
+                    buffer.stash(val)
+                else:
+                    buffer.stash(car)
+            else:
+                return car
+        else:
+            if buffer.top() == ')':
+                buffer.stash(Pair(val, None))
+            else:
+                cdr = buffer.pop()
+                buffer.stash(Pair(val, cdr))
+
+if __name__ == "__main__":
+    #b1 = Stack(["(", "*", 1, 2, ")"])
+    #b2 = Stack(["(", "*", "(", "+", 1, 2, ")", 3, ")"])
+    #val1 = parse(b1)
+    #val2 = parse(b2)
+    #print(val1)
+    #print(val2)
+    exp = "(lambda (x) (+ 1 x))"
+    print(exp)
+    exp = bulid_exp_list(exp)
+    print(exp)
+    b3 = Stack(exp)
+    print(parse(b3))

py/representation.py

@@ -0,0 +1,29 @@
+class Pair:
+    """Pair: has car and cdr"""
+
+    def __init__(self, car, cdr):
+        self.car = car
+        self.cdr = cdr
+
+    def __repr__(self):
+        return "<Pair: {car} {cdr}>".format(car=repr(self.car), 
+                                            cdr=repr(self.cdr))
+
+    def __str__(self):
+        s = "(" + str(self.car)
+        cdr = self.cdr
+        while isinstance(cdr, Pair):
+            s += " " + str(cdr.car)
+            cdr = cdr.cdr
+        if cdr is not None:
+            s += " . " + str(cdr)
+        return s + ")"
+
+    def __len__(self):
+        n, cdr = 1, self.cdr
+        while isinstance(cdr, Pair):
+            n += 1
+            cdr = cdr.cdr
+        if cdr is not None:
+            raise TypeError("length attempted on improper list")
+        return n

py/split.py

@@ -0,0 +1,46 @@
+import string
+
+_NUMERAL_STARTS = set(string.digits) | set('+-.')
+_SYMBOL_CHARS = (set('!$%&*/:<=>?@^_~') | set(string.ascii_lowercase) |
+                 set(string.ascii_uppercase) | _NUMERAL_STARTS)
+_STRING_DELIMS = set('"')
+_WHITESPACE = set(' \t\n\r')
+_SINGLE_CHAR_TOKENS = set("()'`")
+_TOKEN_END = _WHITESPACE | _SINGLE_CHAR_TOKENS | _STRING_DELIMS | {',', ',@'}
+DELIMITERS = _SINGLE_CHAR_TOKENS | {'.', ',', ',@'}
+
+def split_exp_str(line, index=0):
+    """返回分割得到的字符串"""
+    var = []
+    while index < len(line):
+        var.clear()
+        char = line[index]
+        if char in _WHITESPACE:
+            index += 1
+        elif char in _STRING_DELIMS:
+            index += 1 # 跳过"
+            char = line[index]
+            while char not in _STRING_DELIMS:
+                var.append(char)
+                index += 1
+                char = line[index]
+            index += 1 # 跳过"
+            yield "".join(var)
+        elif char in _SINGLE_CHAR_TOKENS:
+            index += 1
+            yield char
+        elif char in _SYMBOL_CHARS:
+            while char not in _WHITESPACE | _SINGLE_CHAR_TOKENS:
+                var.append(char)
+                index += 1
+                char = line[index]
+            yield "".join(var)
+
+def bulid_exp_list(line):
+    exp = []
+    for var in split_exp_str(line):
+        exp.append(var)
+    return exp
+
+if __name__ == "__main__":
+    print(bulid_exp_list('(define "asd" (+ 1 2) (* 3 4) "adf")'))