summaryrefslogtreecommitdiffstats
path: root/Tools/scripts/generate_global_objects.py
blob: 27e3530f38574da354b980b959b8195ebc299776 (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
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
import contextlib
import io
import os.path
import re

__file__ = os.path.abspath(__file__)
ROOT = os.path.dirname(os.path.dirname(os.path.dirname(__file__)))
INTERNAL = os.path.join(ROOT, 'Include', 'internal')


IGNORED = {
    'ACTION',  # Python/_warnings.c
    'ATTR',  # Python/_warnings.c and Objects/funcobject.c
    'DUNDER',  # Objects/typeobject.c
    'RDUNDER',  # Objects/typeobject.c
    'SPECIAL',  # Objects/weakrefobject.c
}
IDENTIFIERS = [
    # from ADD() Python/_warnings.c
    'default',
    'ignore',

    # from GET_WARNINGS_ATTR() in Python/_warnings.c
    'WarningMessage',
    '_showwarnmsg',
    '_warn_unawaited_coroutine',
    'defaultaction',
    'filters',
    'onceregistry',

    # from WRAP_METHOD() in Objects/weakrefobject.c
    '__bytes__',
    '__reversed__',

    # from COPY_ATTR() in Objects/funcobject.c
    '__module__',
    '__name__',
    '__qualname__',
    '__doc__',
    '__annotations__',

    # from SLOT* in Objects/typeobject.c
    '__abs__',
    '__add__',
    '__and__',
    '__divmod__',
    '__float__',
    '__floordiv__',
    '__getitem__',
    '__iadd__',
    '__iand__',
    '__ifloordiv__',
    '__ilshift__',
    '__imatmul__',
    '__imod__',
    '__imul__',
    '__int__',
    '__invert__',
    '__ior__',
    '__irshift__',
    '__isub__',
    '__itruediv__',
    '__ixor__',
    '__lshift__',
    '__matmul__',
    '__mod__',
    '__mul__',
    '__neg__',
    '__or__',
    '__pos__',
    '__pow__',
    '__radd__',
    '__rand__',
    '__rdivmod__',
    '__rfloordiv__',
    '__rlshift__',
    '__rmatmul__',
    '__rmod__',
    '__rmul__',
    '__ror__',
    '__rpow__',
    '__rrshift__',
    '__rshift__',
    '__rsub__',
    '__rtruediv__',
    '__rxor__',
    '__str__',
    '__sub__',
    '__truediv__',
    '__xor__',
]


#######################################
# helpers

def iter_files():
    for name in ('Modules', 'Objects', 'Parser', 'PC', 'Programs', 'Python'):
        root = os.path.join(ROOT, name)
        for dirname, _, files in os.walk(root):
            for name in files:
                if not name.endswith(('.c', '.h')):
                    continue
                yield os.path.join(dirname, name)


def iter_global_strings():
    id_regex = re.compile(r'\b_Py_ID\((\w+)\)')
    str_regex = re.compile(r'\b_Py_DECLARE_STR\((\w+), "(.*?)"\)')
    for filename in iter_files():
        try:
            infile = open(filename, encoding='utf-8')
        except FileNotFoundError:
            # The file must have been a temporary file.
            continue
        with infile:
            for lno, line in enumerate(infile, 1):
                for m in id_regex.finditer(line):
                    identifier, = m.groups()
                    yield identifier, None, filename, lno, line
                for m in str_regex.finditer(line):
                    varname, string = m.groups()
                    yield varname, string, filename, lno, line


def iter_to_marker(lines, marker):
    for line in lines:
        if line.rstrip() == marker:
            break
        yield line


class Printer:

    def __init__(self, file):
        self.level = 0
        self.file = file
        self.continuation = [False]

    @contextlib.contextmanager
    def indent(self):
        save_level = self.level
        try:
            self.level += 1
            yield
        finally:
            self.level = save_level

    def write(self, arg):
        eol = '\n'
        if self.continuation[-1]:
            eol = f' \\{eol}' if arg else f'\\{eol}'
        self.file.writelines(("    "*self.level, arg, eol))

    @contextlib.contextmanager
    def block(self, prefix, suffix="", *, continuation=None):
        if continuation is None:
            continuation = self.continuation[-1]
        self.continuation.append(continuation)

        self.write(prefix + " {")
        with self.indent():
            yield
        self.continuation.pop()
        self.write("}" + suffix)


@contextlib.contextmanager
def open_for_changes(filename, orig):
    """Like open() but only write to the file if it changed."""
    outfile = io.StringIO()
    yield outfile
    text = outfile.getvalue()
    if text != orig:
        with open(filename, 'w', encoding='utf-8') as outfile:
            outfile.write(text)
    else:
        print(f'# not changed: {filename}')


#######################################
# the global objects

START = '/* The following is auto-generated by Tools/scripts/generate_global_objects.py. */'
END = '/* End auto-generated code */'


def generate_global_strings(identifiers, strings):
    filename = os.path.join(INTERNAL, 'pycore_global_strings.h')

    # Read the non-generated part of the file.
    with open(filename) as infile:
        orig = infile.read()
    lines = iter(orig.rstrip().splitlines())
    before = '\n'.join(iter_to_marker(lines, START))
    for _ in iter_to_marker(lines, END):
        pass
    after = '\n'.join(lines)

    # Generate the file.
    with open_for_changes(filename, orig) as outfile:
        printer = Printer(outfile)
        printer.write(before)
        printer.write(START)
        with printer.block('struct _Py_global_strings', ';'):
            with printer.block('struct', ' literals;'):
                for literal, name in sorted(strings.items(), key=lambda x: x[1]):
                    printer.write(f'STRUCT_FOR_STR({name}, "{literal}")')
            outfile.write('\n')
            with printer.block('struct', ' identifiers;'):
                for name in sorted(identifiers):
                    assert name.isidentifier(), name
                    printer.write(f'STRUCT_FOR_ID({name})')
            with printer.block('struct', ' ascii[128];'):
                printer.write("PyASCIIObject _ascii;")
                printer.write("uint8_t _data[2];")
            with printer.block('struct', ' latin1[128];'):
                printer.write("PyCompactUnicodeObject _latin1;")
                printer.write("uint8_t _data[2];")
        printer.write(END)
        printer.write(after)


def generate_runtime_init(identifiers, strings):
    # First get some info from the declarations.
    nsmallposints = None
    nsmallnegints = None
    with open(os.path.join(INTERNAL, 'pycore_global_objects.h')) as infile:
        for line in infile:
            if line.startswith('#define _PY_NSMALLPOSINTS'):
                nsmallposints = int(line.split()[-1])
            elif line.startswith('#define _PY_NSMALLNEGINTS'):
                nsmallnegints = int(line.split()[-1])
                break
        else:
            raise NotImplementedError
    assert nsmallposints and nsmallnegints

    # Then target the runtime initializer.
    filename = os.path.join(INTERNAL, 'pycore_runtime_init_generated.h')

    # Read the non-generated part of the file.
    with open(filename) as infile:
        orig = infile.read()
    lines = iter(orig.rstrip().splitlines())
    before = '\n'.join(iter_to_marker(lines, START))
    for _ in iter_to_marker(lines, END):
        pass
    after = '\n'.join(lines)

    # Generate the file.
    with open_for_changes(filename, orig) as outfile:
        printer = Printer(outfile)
        printer.write(before)
        printer.write(START)
        with printer.block('#define _Py_global_objects_INIT', continuation=True):
            with printer.block('.singletons =', ','):
                # Global int objects.
                with printer.block('.small_ints =', ','):
                    for i in range(-nsmallnegints, nsmallposints):
                        printer.write(f'_PyLong_DIGIT_INIT({i}),')
                printer.write('')
                # Global bytes objects.
                printer.write('.bytes_empty = _PyBytes_SIMPLE_INIT(0, 0),')
                with printer.block('.bytes_characters =', ','):
                    for i in range(256):
                        printer.write(f'_PyBytes_CHAR_INIT({i}),')
                printer.write('')
                # Global strings.
                with printer.block('.strings =', ','):
                    with printer.block('.literals =', ','):
                        for literal, name in sorted(strings.items(), key=lambda x: x[1]):
                            printer.write(f'INIT_STR({name}, "{literal}"),')
                    with printer.block('.identifiers =', ','):
                        for name in sorted(identifiers):
                            assert name.isidentifier(), name
                            printer.write(f'INIT_ID({name}),')
                    with printer.block('.ascii =', ','):
                        for i in range(128):
                            printer.write(f'_PyASCIIObject_INIT("\\x{i:02x}"),')
                    with printer.block('.latin1 =', ','):
                        for i in range(128, 256):
                            printer.write(f'_PyUnicode_LATIN1_INIT("\\x{i:02x}"),')
                printer.write('')
                with printer.block('.tuple_empty =', ','):
                    printer.write('.ob_base = _PyVarObject_IMMORTAL_INIT(&PyTuple_Type, 0)')
        printer.write('')
        printer.write("static inline void")
        with printer.block("_PyUnicode_InitStaticStrings(void)"):
            printer.write(f'PyObject *string;')
            for i in sorted(identifiers):
                # This use of _Py_ID() is ignored by iter_global_strings()
                # since iter_files() ignores .h files.
                printer.write(f'string = &_Py_ID({i});')
                printer.write(f'PyUnicode_InternInPlace(&string);')
        printer.write(END)
        printer.write(after)


def get_identifiers_and_strings() -> 'tuple[set[str], dict[str, str]]':
    identifiers = set(IDENTIFIERS)
    strings = {}
    for name, string, *_ in iter_global_strings():
        if string is None:
            if name not in IGNORED:
                identifiers.add(name)
        else:
            if string not in strings:
                strings[string] = name
            elif name != strings[string]:
                raise ValueError(f'string mismatch for {name!r} ({string!r} != {strings[name]!r}')
    return identifiers, strings


#######################################
# the script

def main() -> None:
    identifiers, strings = get_identifiers_and_strings()

    generate_global_strings(identifiers, strings)
    generate_runtime_init(identifiers, strings)


if __name__ == '__main__':
    main()