summaryrefslogtreecommitdiffstats
path: root/generator/nanopb_generator.py
blob: 9e2a30eda867fb2d59ab1a045d35497a22a31150 (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
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
'''Generate header file for nanopb from a ProtoBuf FileDescriptorSet.'''

try:
    import google.protobuf.descriptor_pb2 as descriptor
except:
    print
    print "*************************************************************"
    print "*** Could not import the Google protobuf Python libraries ***"
    print "*** Try installing package 'python-protobuf' or similar.  ***"
    print "*************************************************************"
    print
    raise

try:
    import nanopb_pb2
except:
    print
    print "***************************************************************"
    print "*** Could not import the precompiled nanopb_pb2.py.         ***"
    print "*** Run 'make' in the 'generator' folder to update the file.***"
    print "***************************************************************"
    print
    raise

import os.path

# Values are tuple (c type, pb ltype)
FieldD = descriptor.FieldDescriptorProto
datatypes = {
    FieldD.TYPE_BOOL: ('bool', 'PB_LTYPE_VARINT'),
    FieldD.TYPE_DOUBLE: ('double', 'PB_LTYPE_FIXED64'),
    FieldD.TYPE_FIXED32: ('uint32_t', 'PB_LTYPE_FIXED32'),
    FieldD.TYPE_FIXED64: ('uint64_t', 'PB_LTYPE_FIXED64'),
    FieldD.TYPE_FLOAT: ('float', 'PB_LTYPE_FIXED32'),
    FieldD.TYPE_INT32: ('int32_t', 'PB_LTYPE_VARINT'),
    FieldD.TYPE_INT64: ('int64_t', 'PB_LTYPE_VARINT'),
    FieldD.TYPE_SFIXED32: ('int32_t', 'PB_LTYPE_FIXED32'),
    FieldD.TYPE_SFIXED64: ('int64_t', 'PB_LTYPE_FIXED64'),
    FieldD.TYPE_SINT32: ('int32_t', 'PB_LTYPE_SVARINT'),
    FieldD.TYPE_SINT64: ('int64_t', 'PB_LTYPE_SVARINT'),
    FieldD.TYPE_UINT32: ('uint32_t', 'PB_LTYPE_VARINT'),
    FieldD.TYPE_UINT64: ('uint64_t', 'PB_LTYPE_VARINT')
}

class Names:
    '''Keeps a set of nested names and formats them to C identifier.
    You can subclass this with your own implementation.
    '''
    def __init__(self, parts = ()):
        if isinstance(parts, Names):
            parts = parts.parts
        self.parts = tuple(parts)
    
    def __str__(self):
        return '_'.join(self.parts)

    def __add__(self, other):
        if isinstance(other, (str, unicode)):
            return Names(self.parts + (other,))
        elif isinstance(other, tuple):
            return Names(self.parts + other)
        else:
            raise ValueError("Name parts should be of type str")
    
def names_from_type_name(type_name):
    '''Parse Names() from FieldDescriptorProto type_name'''
    if type_name[0] != '.':
        raise NotImplementedError("Lookup of non-absolute type names is not supported")
    return Names(type_name[1:].split('.'))

class Enum:
    def __init__(self, names, desc):
        '''desc is EnumDescriptorProto'''
        self.names = names + desc.name
        self.values = [(self.names + x.name, x.number) for x in desc.value]
    
    def __str__(self):
        result = 'typedef enum {\n'
        result += ',\n'.join(["    %s = %d" % x for x in self.values])
        result += '\n} %s;' % self.names
        return result

class Field:
    def __init__(self, struct_name, desc):
        '''desc is FieldDescriptorProto'''
        self.tag = desc.number
        self.struct_name = struct_name
        self.name = desc.name
        self.default = None
        self.max_size = None
        self.max_count = None
        self.array_decl = ""
        
        # Parse nanopb-specific field options
        if desc.options.HasExtension(nanopb_pb2.nanopb):
            ext = desc.options.Extensions[nanopb_pb2.nanopb]
            if ext.HasField("max_size"):
                self.max_size = ext.max_size
            if ext.HasField("max_count"):
                self.max_count = ext.max_count
        
        if desc.HasField('default_value'):
            self.default = desc.default_value
        
        # Decide HTYPE
        # HTYPE is the high-order nibble of nanopb field description,
        # defining whether value is required/optional/repeated.
        is_callback = False
        if desc.label == FieldD.LABEL_REQUIRED:
            self.htype = 'PB_HTYPE_REQUIRED'
        elif desc.label == FieldD.LABEL_OPTIONAL:
            self.htype = 'PB_HTYPE_OPTIONAL'
        elif desc.label == FieldD.LABEL_REPEATED:
            if self.max_count is None:
                is_callback = True
            else:
                self.htype = 'PB_HTYPE_ARRAY'
                self.array_decl = '[%d]' % self.max_count
        else:
            raise NotImplementedError(desc.label)
        
        # Decide LTYPE and CTYPE
        # LTYPE is the low-order nibble of nanopb field description,
        # defining how to decode an individual value.
        # CTYPE is the name of the c type to use in the struct.
        if datatypes.has_key(desc.type):
            self.ctype, self.ltype = datatypes[desc.type]
        elif desc.type == FieldD.TYPE_ENUM:
            self.ltype = 'PB_LTYPE_VARINT'
            self.ctype = names_from_type_name(desc.type_name)
            if self.default is not None:
                self.default = self.ctype + self.default
        elif desc.type == FieldD.TYPE_STRING:
            self.ltype = 'PB_LTYPE_STRING'
            if self.max_size is None:
                is_callback = True
            else:
                self.ctype = 'char'
                self.array_decl += '[%d]' % self.max_size
        elif desc.type == FieldD.TYPE_BYTES:
            self.ltype = 'PB_LTYPE_BYTES'
            if self.max_size is None:
                is_callback = True
            else:
                self.ctype = self.struct_name + self.name + 't'
        elif desc.type == FieldD.TYPE_MESSAGE:
            self.ltype = 'PB_LTYPE_SUBMESSAGE'
            self.ctype = self.submsgname = names_from_type_name(desc.type_name)
        else:
            raise NotImplementedError(desc.type)
        
        if is_callback:
            self.htype = 'PB_HTYPE_CALLBACK'
            self.ctype = 'pb_callback_t'
            self.array_decl = ''
    
    def __cmp__(self, other):
        return cmp(self.tag, other.tag)
    
    def __str__(self):
        if self.htype == 'PB_HTYPE_OPTIONAL':
            result = '    bool has_' + self.name + ';\n'
        elif self.htype == 'PB_HTYPE_ARRAY':
            result = '    size_t ' + self.name + '_count;\n'
        else:
            result = ''
        result += '    %s %s%s;' % (self.ctype, self.name, self.array_decl)
        return result
    
    def types(self):
        '''Return definitions for any special types this field might need.'''
        if self.ltype == 'PB_LTYPE_BYTES' and self.max_size is not None:
            result = 'typedef struct {\n'
            result += '    size_t size;\n'
            result += '    uint8_t bytes[%d];\n' % self.max_size
            result += '} %s;\n' % self.ctype
        else:
            result = None
        return result
    
    def default_decl(self, declaration_only = False):
        '''Return definition for this field's default value.'''
        if self.default is None:
            return None
        
        if self.ltype == 'PB_LTYPE_STRING':
            ctype = 'char'
            if self.max_size is None:
                return None # Not implemented
            else:
                array_decl = '[%d]' % (self.max_size + 1)
            default = str(self.default).encode('string_escape')
            default = default.replace('"', '\\"')
            default = '"' + default + '"'
        elif self.ltype == 'PB_LTYPE_BYTES':
            data = self.default.decode('string_escape')
            data = ['0x%02x' % ord(c) for c in data]
            
            if self.max_size is None:
                return None # Not implemented
            else:
                ctype = self.ctype
            
            default = '{%d, {%s}}' % (len(data), ','.join(data))
            array_decl = ''
        else:
            ctype, default = self.ctype, self.default
            array_decl = ''
        
        if declaration_only:
            return 'extern const %s %s_default%s;' % (ctype, self.struct_name + self.name, array_decl)
        else:
            return 'const %s %s_default%s = %s;' % (ctype, self.struct_name + self.name, array_decl, default)
    
    def pb_field_t(self, prev_field_name):
        '''Return the pb_field_t initializer to use in the constant array.
        prev_field_name is the name of the previous field or None.
        '''
        result = '    {%d, ' % self.tag
        result += self.htype
        if self.ltype is not None:
            result += ' | ' + self.ltype
        result += ',\n'
        
        if prev_field_name is None:
            result += '    offsetof(%s, %s),' % (self.struct_name, self.name)
        else:
            result += '    pb_delta_end(%s, %s, %s),' % (self.struct_name, self.name, prev_field_name)
        
        if self.htype == 'PB_HTYPE_OPTIONAL':
            result += '\n    pb_delta(%s, has_%s, %s),' % (self.struct_name, self.name, self.name)
        elif self.htype == 'PB_HTYPE_ARRAY':
            result += '\n    pb_delta(%s, %s_count, %s),' % (self.struct_name, self.name, self.name)
        else:
            result += ' 0,'
        
        
        if self.htype == 'PB_HTYPE_ARRAY':
            result += '\n    pb_membersize(%s, %s[0]),' % (self.struct_name, self.name)
            result += ('\n    pb_membersize(%s, %s) / pb_membersize(%s, %s[0]),'
                       % (self.struct_name, self.name, self.struct_name, self.name))
        else:
            result += '\n    pb_membersize(%s, %s),' % (self.struct_name, self.name)
            result += ' 0,'
        
        if self.ltype == 'PB_LTYPE_SUBMESSAGE':
            result += '\n    &%s_fields}' % self.submsgname
        elif self.default is None or self.htype == 'PB_HTYPE_CALLBACK':
            result += ' 0}'
        else:
            result += '\n    &%s_default}' % (self.struct_name + self.name)
        
        return result
    
    def largest_field_value(self):
        '''Determine if this field needs 16bit or 32bit pb_field_t structure to compile properly.
        Returns numeric value or a C-expression for assert.'''
        if self.ltype == 'PB_LTYPE_SUBMESSAGE':
            if self.htype == 'PB_HTYPE_ARRAY':
                return 'pb_membersize(%s, %s[0])' % (self.struct_name, self.name)
            else:
                return 'pb_membersize(%s, %s)' % (self.struct_name, self.name)

        return max(self.tag, self.max_size, self.max_count)        


class Message:
    def __init__(self, names, desc):
        self.name = names
        self.fields = [Field(self.name, f) for f in desc.field]
        self.ordered_fields = self.fields[:]
        self.ordered_fields.sort()

    def get_dependencies(self):
        '''Get list of type names that this structure refers to.'''
        return [str(field.ctype) for field in self.fields]
    
    def __str__(self):
        result = 'typedef struct {\n'
        result += '\n'.join([str(f) for f in self.ordered_fields])
        result += '\n} %s;' % self.name
        return result
    
    def types(self):
        result = ""
        for field in self.fields:
            types = field.types()
            if types is not None:
                result += types + '\n'
        return result
    
    def default_decl(self, declaration_only = False):
        result = ""
        for field in self.fields:
            default = field.default_decl(declaration_only)
            if default is not None:
                result += default + '\n'
        return result

    def fields_declaration(self):
        result = 'extern const pb_field_t %s_fields[%d];' % (self.name, len(self.fields) + 1)
        return result

    def fields_definition(self):
        result = 'const pb_field_t %s_fields[%d] = {\n' % (self.name, len(self.fields) + 1)
        
        prev = None
        for field in self.ordered_fields:
            result += field.pb_field_t(prev)
            result += ',\n\n'
            prev = field.name
        
        result += '    PB_LAST_FIELD\n};'
        return result

def iterate_messages(desc, names = Names()):
    '''Recursively find all messages. For each, yield name, DescriptorProto.'''
    if hasattr(desc, 'message_type'):
        submsgs = desc.message_type
    else:
        submsgs = desc.nested_type
    
    for submsg in submsgs:
        sub_names = names + submsg.name
        yield sub_names, submsg
        
        for x in iterate_messages(submsg, sub_names):
            yield x

def parse_file(fdesc):
    '''Takes a FileDescriptorProto and returns tuple (enum, messages).'''
    
    enums = []
    messages = []
    
    if fdesc.package:
        base_name = Names(fdesc.package.split('.'))
    else:
        base_name = Names()
    
    for enum in fdesc.enum_type:
        enums.append(Enum(base_name, enum))
    
    for names, message in iterate_messages(fdesc, base_name):
        messages.append(Message(names, message))
        for enum in message.enum_type:
            enums.append(Enum(names, enum))
    
    return enums, messages

def toposort2(data):
    '''Topological sort.
    From http://code.activestate.com/recipes/577413-topological-sort/
    This function is under the MIT license.
    '''
    for k, v in data.items():
        v.discard(k) # Ignore self dependencies
    extra_items_in_deps = reduce(set.union, data.values(), set()) - set(data.keys())
    data.update(dict([(item, set()) for item in extra_items_in_deps]))
    while True:
        ordered = set(item for item,dep in data.items() if not dep)
        if not ordered:
            break
        for item in sorted(ordered):
            yield item
        data = dict([(item, (dep - ordered)) for item,dep in data.items()
                if item not in ordered])
    assert not data, "A cyclic dependency exists amongst %r" % data

def sort_dependencies(messages):
    '''Sort a list of Messages based on dependencies.'''
    dependencies = {}
    message_by_name = {}
    for message in messages:
        dependencies[str(message.name)] = set(message.get_dependencies())
        message_by_name[str(message.name)] = message
    
    for msgname in toposort2(dependencies):
        if msgname in message_by_name:
            yield message_by_name[msgname]

def generate_header(dependencies, headername, enums, messages):
    '''Generate content for a header file.
    Generates strings, which should be concatenated and stored to file.
    '''
    
    yield '/* Automatically generated nanopb header */\n'
    
    symbol = headername.replace('.', '_').upper()
    yield '#ifndef _PB_%s_\n' % symbol
    yield '#define _PB_%s_\n' % symbol
    yield '#include <pb.h>\n\n'
    
    for dependency in dependencies:
        noext = os.path.splitext(dependency)[0]
        yield '#include "%s.pb.h"\n' % noext
    yield '\n'
    
    yield '/* Enum definitions */\n'
    for enum in enums:
        yield str(enum) + '\n\n'
    
    yield '/* Struct definitions */\n'
    for msg in sort_dependencies(messages):
        yield msg.types()
        yield str(msg) + '\n\n'
        
    yield '/* Default values for struct fields */\n'
    for msg in messages:
        yield msg.default_decl(True)
    yield '\n'
    
    yield '/* Struct field encoding specification for nanopb */\n'
    for msg in messages:
        yield msg.fields_declaration() + '\n'
    
    count_required_fields = lambda m: len([f for f in msg.fields if f.htype == 'PB_HTYPE_REQUIRED'])
    largest_msg = max(messages, key = count_required_fields)
    largest_count = count_required_fields(largest_msg)
    if largest_count > 64:
        yield '\n/* Check that missing required fields will be properly detected */\n'
        yield '#if PB_MAX_REQUIRED_FIELDS < %d\n' % largest_count
        yield '#warning Properly detecting missing required fields in %s requires \\\n' % largest_msg.name
        yield '         setting PB_MAX_REQUIRED_FIELDS to %d or more.\n' % largest_count
        yield '#endif\n'
    
    worst = 0
    worst_field = ''
    checks = []
    for msg in messages:
        for field in msg.fields:
            status = field.largest_field_value()
            if isinstance(status, (str, unicode)):
                checks.append(status)
            elif status > worst:
                worst = status
                worst_field = str(field.struct_name) + '.' + str(field.name)

    if worst > 255 or checks:
        yield '\n/* Check that field information fits in pb_field_t */\n'
        yield '/* (Largest message has %d fields' % worst
        if checks: yield ' and submessages have to be checked at compile-time.'
        yield ') */\n'
        
        if worst < 65536:
            yield '#if !defined(PB_FIELD_16BIT) && !defined(PB_FIELD_32BIT)\n'
            if worst > 255:
                yield '#error Field descriptor for %s is too large. Define PB_FIELD_16BIT to fix this.\n' % worst_field
            else:
                assertion = ' && '.join(str(c) + ' < 256' for c in checks)
                yield 'STATIC_ASSERT((%s), YOU_MUST_DEFINE_PB_FIELD_16BIT)\n' % assertion
            yield '#endif\n\n'
        
        if worst > 65535 or checks:
            yield '#if !defined(PB_FIELD_32BIT)\n'
            if worst > 65535:
                yield '#error Field descriptor for %s is too large. Define PB_FIELD_32BIT to fix this.\n' % worst_field
            else:
                assertion = ' && '.join(str(c) + ' < 65536' for c in checks)
                yield 'STATIC_ASSERT((%s), YOU_MUST_DEFINE_PB_FIELD_32BIT)\n' % assertion
            yield '#endif\n'
    
    # End of header
    yield '\n#endif\n'

def generate_source(headername, enums, messages):
    '''Generate content for a source file.'''
    
    yield '/* Automatically generated nanopb constant definitions */\n'
    yield '#include "%s"\n\n' % headername
    
    for msg in messages:
        yield msg.default_decl(False)
    
    yield '\n\n'
    
    for msg in messages:
        yield msg.fields_definition() + '\n\n'

if __name__ == '__main__':
    import sys
    import os.path
    
    if len(sys.argv) != 2:
        print "Usage: " + sys.argv[0] + " file.pb"
        print "where file.pb has been compiled from .proto by:"
        print "protoc -ofile.pb file.proto"
        print "Output fill be written to file.pb.h and file.pb.c"
        sys.exit(1)
    
    data = open(sys.argv[1], 'rb').read()
    fdesc = descriptor.FileDescriptorSet.FromString(data)
    enums, messages = parse_file(fdesc.file[0])
    
    noext = os.path.splitext(sys.argv[1])[0]
    headername = noext + '.pb.h'
    sourcename = noext + '.pb.c'
    headerbasename = os.path.basename(headername)
    
    print "Writing to " + headername + " and " + sourcename
    
    # List of .proto files that should not be included in the C header file
    # even if they are mentioned in the source .proto.
    excludes = ['nanopb.proto', 'google/protobuf/descriptor.proto']
    dependencies = [d for d in fdesc.file[0].dependency if d not in excludes]
    
    header = open(headername, 'w')
    for part in generate_header(dependencies, headerbasename, enums, messages):
        header.write(part)

    source = open(sourcename, 'w')
    for part in generate_source(headerbasename, enums, messages):
        source.write(part)