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
|
"""
Process raw qstr file and output qstr data with length, hash and data bytes.
This script works with Python 2.6, 2.7, 3.3 and 3.4.
"""
from __future__ import print_function
import re
import sys
# codepoint2name is different in Python 2 to Python 3
import platform
if platform.python_version_tuple()[0] == '2':
from htmlentitydefs import codepoint2name
elif platform.python_version_tuple()[0] == '3':
from html.entities import codepoint2name
codepoint2name[ord('-')] = 'hyphen';
# add some custom names to map characters that aren't in HTML
codepoint2name[ord(' ')] = 'space'
codepoint2name[ord('\'')] = 'squot'
codepoint2name[ord(',')] = 'comma'
codepoint2name[ord('.')] = 'dot'
codepoint2name[ord(':')] = 'colon'
codepoint2name[ord('/')] = 'slash'
codepoint2name[ord('%')] = 'percent'
codepoint2name[ord('#')] = 'hash'
codepoint2name[ord('(')] = 'paren_open'
codepoint2name[ord(')')] = 'paren_close'
codepoint2name[ord('[')] = 'bracket_open'
codepoint2name[ord(']')] = 'bracket_close'
codepoint2name[ord('{')] = 'brace_open'
codepoint2name[ord('}')] = 'brace_close'
codepoint2name[ord('*')] = 'star'
codepoint2name[ord('!')] = 'bang'
codepoint2name[ord('\\')] = 'backslash'
# this must match the equivalent function in qstr.c
def compute_hash(qstr):
hash = 5381
for char in qstr:
hash = (hash * 33) ^ ord(char)
# Make sure that valid hash is never zero, zero means "hash not computed"
return (hash & 0xffff) or 1
def do_work(infiles):
# read the qstrs in from the input files
qcfgs = {}
qstrs = {}
for infile in infiles:
with open(infile, 'rt') as f:
for line in f:
line = line.strip()
# is this a config line?
match = re.match(r'^QCFG\((.+), (.+)\)', line)
if match:
value = match.group(2)
if value[0] == '(' and value[-1] == ')':
# strip parenthesis from config value
value = value[1:-1]
qcfgs[match.group(1)] = value
continue
# is this a QSTR line?
match = re.match(r'^Q\((.*)\)$', line)
if not match:
continue
# get the qstr value
qstr = match.group(1)
ident = re.sub(r'[^A-Za-z0-9_]', lambda s: "_" + codepoint2name[ord(s.group(0))] + "_", qstr)
# don't add duplicates
if ident in qstrs:
continue
# add the qstr to the list, with order number to retain original order in file
qstrs[ident] = (len(qstrs), ident, qstr)
# get config variables
cfg_bytes_len = int(qcfgs['BYTES_IN_LEN'])
cfg_max_len = 1 << (8 * cfg_bytes_len)
# print out the starte of the generated C header file
print('// This file was automatically generated by makeqstrdata.py')
print('')
# add NULL qstr with no hash or data
print('QDEF(MP_QSTR_NULL, (const byte*)"\\x00\\x00%s" "")' % ('\\x00' * cfg_bytes_len))
# go through each qstr and print it out
for order, ident, qstr in sorted(qstrs.values(), key=lambda x: x[0]):
qhash = compute_hash(qstr)
# Calculate len of str, taking escapes into account
qlen = len(qstr.replace("\\\\", "-").replace("\\", ""))
qdata = qstr.replace('"', '\\"')
if qlen >= cfg_max_len:
print('qstr is too long:', qstr)
assert False
qlen_str = ('\\x%02x' * cfg_bytes_len) % tuple(((qlen >> (8 * i)) & 0xff) for i in range(cfg_bytes_len))
print('QDEF(MP_QSTR_%s, (const byte*)"\\x%02x\\x%02x%s" "%s")' % (ident, qhash & 0xff, (qhash >> 8) & 0xff, qlen_str, qdata))
if __name__ == "__main__":
do_work(sys.argv[1:])
|
