-
Notifications
You must be signed in to change notification settings - Fork 9
Expand file tree
/
Copy pathcparser.py
More file actions
231 lines (196 loc) · 6.23 KB
/
cparser.py
File metadata and controls
231 lines (196 loc) · 6.23 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
# emacs: -*- mode: python; py-indent-offset: 4; indent-tabs-mode: nil -*-
# vi: set ft=python sts=4 ts=4 sw=4 et:
from builtins import next
from past.builtins import basestring
from builtins import object
from ctypes import *
import re
TOKENS=re.compile(r'[\w_]+|[()*,]')
WORD=re.compile(r'^[_\w]+$')
_keywords=set()
_types={}
def define(name, t):
if isinstance(t,basestring):
# referencing an existing type by name
t=_parse_type(t)
_types[name]=t
# assume defined types are keywords
for token in re.findall(r'\w+', name):
_keywords.add(token)
return t # return the defined-type
# define standard ctypes
define('void', None)
define('void*', c_void_p)
define('char', c_char)
define('wchar_t', c_wchar)
define('unsigned char', c_ubyte)
define('short', c_short)
define('unsigned short', c_ushort)
define('int', c_int)
define('unsigned int', c_uint)
define('long', c_long)
define('unsigned long', c_ulong)
define('long long', c_longlong)
define('unsigned long long', c_ulonglong)
define('float', c_float)
define('double', c_double)
define('char*', c_char_p)
define('wchar_t*', c_wchar_p)
def _parse_type(type_str):
# see if the type is there
if type_str in _types:
return _types[type_str]
if type_str.endswith('*'):
type_str=type_str[:-1]
return POINTER(_parse_type(type_str))
else:
raise ValueError("unknown type: " + type_str)
class tokenizer(object):
def __init__(self,s):
self.s=s
# parse raw tokens, then group adjacent keywords
# together to treat them like one token (e.g. 'unsigned int')
tokens=TOKENS.findall(s)
self.tokens=[]
k=[]
for t in tokens:
if t in _keywords:
k.append(t)
else:
if len(k) > 0:
self.tokens.append(' '.join(k))
k=[]
self.tokens.append(t)
if len(k) > 0:
self.tokens.append(' '.join(k))
self.i=-1
def __next__(self):
if self.empty():
raise ValueError('no more tokens found parsing - %s' % self.s)
self.i+=1
return self.current()
def current(self):
return self.tokens[self.i]
def push_back(self):
if self.i < 0:
raise ValueError('pushed back too far parsing - %s' % self.s)
self.i-=1
def empty(self):
# are we at the last element
return self.i >= (len(self.tokens)-1)
class c_type(object):
def __init__(self,type_name,name=''):
self.type_name=type_name
self.name=name
@property
def ctype(self):
return _parse_type(self.type_name)
@property
def cstruct(self):
'''convert the type for use in a struct'''
return (self.name, self.ctype)
def cast(self,value):
return cast(value,self.ctype)
def __repr__(self):
if self.name:
return u'c_type(%s %s)' % (self.type_str, self.name)
return u'c_type(%s)' % self.type_str
class c_function(object):
def __init__(self,return_type, name, param_list):
self.return_type=return_type
self.name=name
self.param_list=param_list
@property
def ctype(self):
params=[p.ctype for p in self.param_list]
return CFUNCTYPE(self.return_type.ctype, *params)
@property
def cstruct(self):
'''convert the type for use in a struct'''
return (self.name, self.ctype)
def from_lib(self,lib):
fn=getattr(lib,self.name)
fn.restype=self.return_type.ctype
fn.argtypes=[p.ctype for p in self.param_list]
return fn
def __repr__(self):
return u'c_function(%s %s %s)' % (self.return_type, self.name, self.param_list)
def parse_type(t):
# grab variable name (including * for pointers)
base_type=next(t)
while not t.empty():
if next(t) == '*':
base_type += '*'
else:
t.push_back()
break
return c_type(base_type)
def parse_fn_name(t):
if next(t) == '(':
# fn pointer name
assert next(t) == '*'
name=next(t)
assert next(t) == ')'
else:
name=t.current()
return name
def parse_param(t):
param_type=parse_type(t)
name=next(t)
if WORD.match(name):
param_type.name=name
else:
# un-named param
t.push_back()
return param_type
def parse_param_list(t):
assert next(t) == '('
params=[]
while next(t) != ')':
t.push_back()
params.append(parse_param(t))
if next(t) != ',':
break
assert t.current() == ')'
return params
def parse(s):
'''
parse a c definition/declaration returning a variable def, function def etc
as appropriate
'''
t=tokenizer(s)
def_type = parse_type(t)
if not t.empty():
if next(t) == '(':
# looks like we're parsing a function pointer definition
# e.g. void (*my_fn)(void)
t.push_back()
fn_name = parse_fn_name(t)
param_list = parse_param_list(t)
if not t.empty():
raise ValueError("unexpected tokens at end of function def in - %s" % s)
return c_function(def_type, fn_name, param_list)
else:
t.push_back()
name = next(t)
if not t.empty():
if next(t) == '(':
# parsing function def again (regular def not pointer)
# e.g. void my_fn(void)
t.push_back()
param_list = parse_param_list(t)
if not t.empty():
raise ValueError("unexpected tokens at end of function def in - %s" % s)
return c_function(def_type, name, param_list)
else:
pass
# might be parsing array definition
if not t.empty():
raise ValueError("unexpected tokens at end of variable def in - %s" % s)
# named variable declaration
def_type.name=name
return def_type
else:
# un-named variable type declatation
return def_type
__all__ = ['parse', 'define']