Mercurial > lcfOS

comparison python/ppci/frontends/lexer.py @ 96:a350055d6119

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
movage
author windel
date Mon, 24 Dec 2012 13:30:12 +0100
parents python/libs/compiler/frontends/lexer.py@60cc36ef5a50
children
comparison
equal deleted inserted replaced
95:4a37d6992bd3 96:a350055d6119
1 import collections
2 import re
3 from .errors import CompilerException
4
5 """
6 Lexical analyzer part. Splits the input character stream into tokens.
7 """
8
9 # Token is used in the lexical analyzer:
10 Token = collections.namedtuple('Token', 'typ val row col')
11
12 keywords = ['and', 'array', 'begin', 'by', 'case', 'const', 'div', 'do', \
13 'else', 'elsif', 'end', 'false', 'for', 'if', 'import', 'in', 'is', \
14 'mod', 'module', 'nil', 'not', 'of', 'or', 'pointer', 'procedure', \
15 'record', 'repeat', 'return', 'then', 'to', 'true', 'type', 'until', 'var', \
16 'while', 'asm' ]
17
18 def tokenize(s):
19 """
20 Tokenizer, generates an iterator that
21 returns tokens!
22
23 This GREAT example was taken from python re doc page!
24 """
25 tok_spec = [
26 ('REAL', r'\d+\.\d+'),
27 ('HEXNUMBER', r'0x[\da-fA-F]+'),
28 ('NUMBER', r'\d+'),
29 ('ID', r'[A-Za-z][A-Za-z\d_]*'),
30 ('NEWLINE', r'\n'),
31 ('SKIP', r'[ \t]'),
32 ('COMMENTS', r'{.*}'),
33 ('LEESTEKEN', r':=|[\.,=:;\-+*\[\]/\(\)]|>=|<=|<>|>|<'),
34 ('STRING', r"'.*?'")
35 ]
36 tok_re = '|'.join('(?P<%s>%s)' % pair for pair in tok_spec)
37 gettok = re.compile(tok_re).match
38 line = 1
39 pos = line_start = 0
40 mo = gettok(s)
41 while mo is not None:
42 typ = mo.lastgroup
43 val = mo.group(typ)
44 if typ == 'NEWLINE':
45 line_start = pos
46 line += 1
47 elif typ == 'COMMENTS':
48 pass
49 elif typ != 'SKIP':
50 if typ == 'ID':
51 if val in keywords:
52 typ = val
53 elif typ == 'LEESTEKEN':
54 typ = val
55 elif typ == 'NUMBER':
56 val = int(val)
57 elif typ == 'HEXNUMBER':
58 val = int(val[2:], 16)
59 typ = 'NUMBER'
60 elif typ == 'REAL':
61 val = float(val)
62 elif typ == 'STRING':
63 val = val[1:-1]
64 yield Token(typ, val, line, mo.start()-line_start)
65 pos = mo.end()
66 mo = gettok(s, pos)
67 if pos != len(s):
68 col = pos - line_start
69 raise CompilerException('Unexpected character {0}'.format(s[pos]), line, col)
70 yield Token('END', '', line, 0)
71