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
|
cdef extern from "graminit.c":
ctypedef struct grammar:
pass
cdef grammar _PyParser_Grammar
cdef int Py_file_input
cdef extern from "node.h":
ctypedef struct node
void PyNode_Free(node* n)
int NCH(node* n)
node* CHILD(node* n, int ix)
node* RCHILD(node* n, int ix)
short TYPE(node* n)
char* STR(node* n)
cdef extern from "parsetok.h":
ctypedef struct perrdetail:
pass
cdef void PyParser_SetError(perrdetail *err) except *
cdef node * PyParser_ParseStringFlagsFilenameEx(
const char * s,
const char * filename,
grammar * g,
int start,
perrdetail * err_ret,
int * flags)
import distutils.sysconfig
import os
import re
def extract_names(path):
# All parse tree types are #defined in these files as ints.
type_names = {}
with open(path) as fid:
for line in fid:
if line.startswith('#define'):
try:
_, name, value = line.strip().split()
type_names[int(value)] = name
except:
pass
return type_names
cdef dict type_names = {}
cdef print_tree(node* n, indent=""):
if not type_names:
type_names.update(extract_names(
os.path.join(distutils.sysconfig.get_python_inc(), 'token.h')))
type_names.update(extract_names(
os.path.join(os.path.dirname(__file__), 'graminit.h')))
print indent, type_names.get(TYPE(n), 'unknown'), <object>STR(n) if NCH(n) == 0 else NCH(n)
indent += " "
for i in range(NCH(n)):
print_tree(CHILD(n, i), indent)
def handle_includes(source, path):
# TODO: Use include directory.
def include_here(include_line):
included = os.path.join(os.path.dirname(path), include_line.group(1)[1:-1])
if not os.path.exists(included):
return include_line.group(0) + ' # no such path: ' + included
with open(included) as fid:
return handle_includes(fid.read(), path)
# TODO: Proper string tokenizing.
return re.sub(r'^include\s+([^\n]+[\'"])\s*(#.*)?$', include_here, source, flags=re.M)
def p_module(path):
cdef perrdetail err
cdef int flags
cdef node* n
with open(path) as fid:
source = fid.read()
if '\ninclude ' in source:
# TODO: Tokanizer needs to understand includes.
source = handle_includes(source, path)
path = "preparse(%s)" % path
n = PyParser_ParseStringFlagsFilenameEx(
source,
path,
&_PyParser_Grammar,
Py_file_input,
&err,
&flags)
if n:
# print_tree(n)
PyNode_Free(n)
else:
PyParser_SetError(&err)
|
