summaryrefslogtreecommitdiff
path: root/Tools/scripts/generate_token.py
blob: f2745e8353fc3777da8240da850ea8605fc4daad (plain)
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
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
#! /usr/bin/env python3
# This script generates token related files from Grammar/Tokens:
#
#   Doc/library/token-list.inc
#   Include/token.h
#   Parser/token.c
#   Lib/token.py


NT_OFFSET = 256

def load_tokens(path):
    tok_names = []
    string_to_tok = {}
    ERRORTOKEN = None
    with open(path) as fp:
        for line in fp:
            line = line.strip()
            # strip comments
            i = line.find('#')
            if i >= 0:
                line = line[:i].strip()
            if not line:
                continue
            fields = line.split()
            name = fields[0]
            value = len(tok_names)
            if name == 'ERRORTOKEN':
                ERRORTOKEN = value
            string = fields[1] if len(fields) > 1 else None
            if string:
                string = eval(string)
                string_to_tok[string] = value
            tok_names.append(name)
    return tok_names, ERRORTOKEN, string_to_tok


def update_file(file, content):
    try:
        with open(file, 'r') as fobj:
            if fobj.read() == content:
                return False
    except (OSError, ValueError):
        pass
    with open(file, 'w') as fobj:
        fobj.write(content)
    return True


token_h_template = """\
/* Auto-generated by Tools/scripts/generate_token.py */

/* Token types */
#ifndef Py_LIMITED_API
#ifndef Py_TOKEN_H
#define Py_TOKEN_H
#ifdef __cplusplus
extern "C" {
#endif

#undef TILDE   /* Prevent clash of our definition with system macro. Ex AIX, ioctl.h */

%s\
#define N_TOKENS        %d
#define NT_OFFSET       %d

/* Special definitions for cooperation with parser */

#define ISTERMINAL(x)           ((x) < NT_OFFSET)
#define ISNONTERMINAL(x)        ((x) >= NT_OFFSET)
#define ISEOF(x)                ((x) == ENDMARKER)


PyAPI_DATA(const char * const) _PyParser_TokenNames[]; /* Token names */
PyAPI_FUNC(int) PyToken_OneChar(int);
PyAPI_FUNC(int) PyToken_TwoChars(int, int);
PyAPI_FUNC(int) PyToken_ThreeChars(int, int, int);

#ifdef __cplusplus
}
#endif
#endif /* !Py_TOKEN_H */
#endif /* Py_LIMITED_API */
"""

def make_h(infile, outfile='Include/token.h'):
    tok_names, ERRORTOKEN, string_to_tok = load_tokens(infile)

    defines = []
    for value, name in enumerate(tok_names[:ERRORTOKEN + 1]):
        defines.append("#define %-15s %d\n" % (name, value))

    if update_file(outfile, token_h_template % (
            ''.join(defines),
            len(tok_names),
            NT_OFFSET
        )):
        print("%s regenerated from %s" % (outfile, infile))


token_c_template = """\
/* Auto-generated by Tools/scripts/generate_token.py */

#include "Python.h"
#include "token.h"

/* Token names */

const char * const _PyParser_TokenNames[] = {
%s\
};

/* Return the token corresponding to a single character */

int
PyToken_OneChar(int c1)
{
%s\
    return OP;
}

int
PyToken_TwoChars(int c1, int c2)
{
%s\
    return OP;
}

int
PyToken_ThreeChars(int c1, int c2, int c3)
{
%s\
    return OP;
}
"""

def generate_chars_to_token(mapping, n=1):
    result = []
    write = result.append
    indent = '    ' * n
    write(indent)
    write('switch (c%d) {\n' % (n,))
    for c in sorted(mapping):
        write(indent)
        value = mapping[c]
        if isinstance(value, dict):
            write("case '%s':\n" % (c,))
            write(generate_chars_to_token(value, n + 1))
            write(indent)
            write('    break;\n')
        else:
            write("case '%s': return %s;\n" % (c, value))
    write(indent)
    write('}\n')
    return ''.join(result)

def make_c(infile, outfile='Parser/token.c'):
    tok_names, ERRORTOKEN, string_to_tok = load_tokens(infile)
    string_to_tok['<>'] = string_to_tok['!=']
    chars_to_token = {}
    for string, value in string_to_tok.items():
        assert 1 <= len(string) <= 3
        name = tok_names[value]
        m = chars_to_token.setdefault(len(string), {})
        for c in string[:-1]:
            m = m.setdefault(c, {})
        m[string[-1]] = name

    names = []
    for value, name in enumerate(tok_names):
        if value >= ERRORTOKEN:
            name = '<%s>' % name
        names.append('    "%s",\n' % name)
    names.append('    "<N_TOKENS>",\n')

    if update_file(outfile, token_c_template % (
            ''.join(names),
            generate_chars_to_token(chars_to_token[1]),
            generate_chars_to_token(chars_to_token[2]),
            generate_chars_to_token(chars_to_token[3])
        )):
        print("%s regenerated from %s" % (outfile, infile))


token_inc_template = """\
.. Auto-generated by Tools/scripts/generate_token.py
%s
.. data:: N_TOKENS

.. data:: NT_OFFSET
"""

def make_rst(infile, outfile='Doc/library/token-list.inc'):
    tok_names, ERRORTOKEN, string_to_tok = load_tokens(infile)
    tok_to_string = {value: s for s, value in string_to_tok.items()}

    names = []
    for value, name in enumerate(tok_names[:ERRORTOKEN + 1]):
        names.append('.. data:: %s' % (name,))
        if value in tok_to_string:
            names.append('')
            names.append('   Token value for ``"%s"``.' % tok_to_string[value])
        names.append('')

    if update_file(outfile, token_inc_template % '\n'.join(names)):
        print("%s regenerated from %s" % (outfile, infile))


token_py_template = '''\
"""Token constants."""
# Auto-generated by Tools/scripts/generate_token.py

__all__ = ['tok_name', 'ISTERMINAL', 'ISNONTERMINAL', 'ISEOF']

%s
N_TOKENS = %d
# Special definitions for cooperation with parser
NT_OFFSET = %d

tok_name = {value: name
            for name, value in globals().items()
            if isinstance(value, int) and not name.startswith('_')}
__all__.extend(tok_name.values())

EXACT_TOKEN_TYPES = {
%s
}

def ISTERMINAL(x):
    return x < NT_OFFSET

def ISNONTERMINAL(x):
    return x >= NT_OFFSET

def ISEOF(x):
    return x == ENDMARKER
'''

def make_py(infile, outfile='Lib/token.py'):
    tok_names, ERRORTOKEN, string_to_tok = load_tokens(infile)

    constants = []
    for value, name in enumerate(tok_names):
        constants.append('%s = %d' % (name, value))
    constants.insert(ERRORTOKEN,
        "# These aren't used by the C tokenizer but are needed for tokenize.py")

    token_types = []
    for s, value in sorted(string_to_tok.items()):
        token_types.append('    %r: %s,' % (s, tok_names[value]))

    if update_file(outfile, token_py_template % (
            '\n'.join(constants),
            len(tok_names),
            NT_OFFSET,
            '\n'.join(token_types),
        )):
        print("%s regenerated from %s" % (outfile, infile))


def main(op, infile='Grammar/Tokens', *args):
    make = globals()['make_' + op]
    make(infile, *args)


if __name__ == '__main__':
    import sys
    main(*sys.argv[1:])