|
159
|
1 import collections, re
|
|
|
2
|
|
|
3 # Different instruction sets:
|
|
|
4 class InstructionSet:
|
|
|
5 pass
|
|
|
6
|
|
|
7 class X86(InstructionSet):
|
|
|
8 pass
|
|
|
9
|
|
|
10 # Generic assembler:
|
|
|
11
|
|
|
12 class SourceLocation:
|
|
|
13 def __init__(self, x):
|
|
|
14 self.pos = x
|
|
|
15
|
|
|
16 class SourceRange:
|
|
|
17 def __init__(self, p1, p2):
|
|
|
18 self.p1 = p1
|
|
|
19 self.p2 = p2
|
|
|
20
|
|
|
21 # Token is used in the lexical analyzer:
|
|
|
22 Token = collections.namedtuple('Token', 'typ val row col')
|
|
|
23
|
|
|
24 keywords = ['global', 'db']
|
|
|
25
|
|
|
26 def tokenize(s):
|
|
|
27 """
|
|
|
28 Tokenizer, generates an iterator that
|
|
|
29 returns tokens!
|
|
|
30
|
|
|
31 This GREAT example was taken from python re doc page!
|
|
|
32 """
|
|
|
33 tok_spec = [
|
|
|
34 ('REAL', r'\d+\.\d+'),
|
|
|
35 ('HEXNUMBER', r'0x[\da-fA-F]+'),
|
|
|
36 ('NUMBER', r'\d+'),
|
|
|
37 ('ID', r'[A-Za-z][A-Za-z\d_]*'),
|
|
|
38 ('NEWLINE', r'\n'),
|
|
|
39 ('SKIP', r'[ \t]'),
|
|
|
40 ('LEESTEKEN', r':=|[\.,=:;\-+*\[\]/\(\)]|>=|<=|<>|>|<'),
|
|
|
41 ('STRING', r"'.*?'")
|
|
|
42 ]
|
|
|
43 tok_re = '|'.join('(?P<%s>%s)' % pair for pair in tok_spec)
|
|
|
44 print(tok_re)
|
|
|
45 gettok = re.compile(tok_re).match
|
|
|
46 line = 1
|
|
|
47 pos = line_start = 0
|
|
|
48 mo = gettok(s)
|
|
|
49 while mo is not None:
|
|
|
50 typ = mo.lastgroup
|
|
|
51 val = mo.group(typ)
|
|
|
52 if typ == 'NEWLINE':
|
|
|
53 line_start = pos
|
|
|
54 line += 1
|
|
|
55 elif typ == 'COMMENTS':
|
|
|
56 pass
|
|
|
57 elif typ != 'SKIP':
|
|
|
58 if typ == 'ID':
|
|
|
59 if val in keywords:
|
|
|
60 typ = val
|
|
|
61 elif typ == 'LEESTEKEN':
|
|
|
62 typ = val
|
|
|
63 elif typ == 'NUMBER':
|
|
|
64 val = int(val)
|
|
|
65 elif typ == 'HEXNUMBER':
|
|
|
66 val = int(val[2:], 16)
|
|
|
67 typ = 'NUMBER'
|
|
|
68 elif typ == 'REAL':
|
|
|
69 val = float(val)
|
|
|
70 elif typ == 'STRING':
|
|
|
71 val = val[1:-1]
|
|
|
72 yield Token(typ, val, line, mo.start()-line_start)
|
|
|
73 pos = mo.end()
|
|
|
74 mo = gettok(s, pos)
|
|
|
75 if pos != len(s):
|
|
|
76 col = pos - line_start
|
|
|
77 raise CompilerException('Unexpected character {0}'.format(s[pos]), line, col)
|
|
|
78 yield Token('END', '', line, 0)
|
|
|
79
|
|
|
80 class Lexer:
|
|
|
81 def __init__(self, src):
|
|
|
82 self.tokens = tokenize(src)
|
|
|
83 self.curTok = self.tokens.__next__()
|
|
|
84 def eat(self):
|
|
|
85 t = self.curTok
|
|
|
86 self.curTok = self.tokens.__next__()
|
|
|
87 return t
|
|
|
88 @property
|
|
|
89 def Peak(self):
|
|
|
90 return self.curTok
|
|
|
91
|
|
|
92 class Parser:
|
|
|
93 def __init__(self, lxr):
|
|
|
94 self.lxr = lxr
|
|
|
95 def parse(self):
|
|
|
96 t = self.lxr.eat()
|
|
|
97
|
|
|
98 while True:
|
|
|
99 ins = self.parseLine()
|
|
|
100 print(ins)
|
|
|
101 t = self.lxr.eat()
|
|
|
102 def parseLine(self):
|
|
|
103 self.parseLabel()
|
|
|
104 if self.lxr.Peak == ';':
|
|
|
105 self.eatComments()
|
|
|
106 def parseLabel(self):
|
|
|
107 i = self.lxr.eat()
|
|
|
108
|
|
|
109 class Assembler:
|
|
|
110 def assemble(self, asmsrc):
|
|
|
111 print('assembling', asmsrc)
|
|
|
112 lxr = Lexer(asmsrc)
|
|
|
113 prsr = Parser(lxr)
|
|
|
114 instructions = prsr.parse()
|
|
|
115 return instructions
|
|
|
116
|
