pub/python-betterproto
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Make plugin use betterproto generated classes internally

Browse Source 
This means the betterproto plugin no longer needs to depend durectly on
protobuf.

This requires a small runtime hack to monkey patch some google types to
get around the fact that the compiler uses proto2, but betterproto
expects proto3.

Also:
- regenerate google.protobuf package
- fix a regex bug in the logic for determining whether to use a google
  wrapper type.
- fix a bug causing comments to get mixed up when multiple proto files
  generate code into a single python module
This commit is contained in:
Nat Noordanus 2020-10-18 22:47:58 +02:00 committed by Basileus
parent 7a358a63cf
commit fe1e712fdb
11 changed files with 2381 additions and 1122 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@ -443,9 +443,9 @@ Assuming your `google.protobuf` source files (included with all releases of `pro
```sh ```sh
protoc \ protoc \
--plugin=protoc-gen-custom=betterproto/plugin.py \ --plugin=protoc-gen-custom=src/betterproto/plugin/main.py \
--custom_opt=INCLUDE_GOOGLE \ --custom_opt=INCLUDE_GOOGLE \
--custom_out=betterproto/lib \ --custom_out=src/betterproto/lib \
-I /usr/local/include/ \ -I /usr/local/include/ \
/usr/local/include/google/protobuf/*.proto /usr/local/include/google/protobuf/*.proto
``` ```

18
src/betterproto/__init__.py
View File

@ -1172,6 +1172,7 @@ from .lib.google.protobuf import ( # noqa
BytesValue, BytesValue,
DoubleValue, DoubleValue,
Duration, Duration,
EnumValue,
FloatValue, FloatValue,
Int32Value, Int32Value,
Int64Value, Int64Value,
@ -1238,14 +1239,17 @@ class _WrappedMessage(Message):
def _get_wrapper(proto_type: str) -> Type: def _get_wrapper(proto_type: str) -> Type:
"""Get the wrapper message class for a wrapped type.""" """Get the wrapper message class for a wrapped type."""
# TODO: include ListValue and NullValue?
return { return {
TYPE_BOOL: BoolValue, TYPE_BOOL: BoolValue,
TYPE_INT32: Int32Value,
TYPE_UINT32: UInt32Value,
TYPE_INT64: Int64Value,
TYPE_UINT64: UInt64Value,
TYPE_FLOAT: FloatValue,
TYPE_DOUBLE: DoubleValue,
TYPE_STRING: StringValue,
TYPE_BYTES: BytesValue, TYPE_BYTES: BytesValue,
TYPE_DOUBLE: DoubleValue,
TYPE_FLOAT: FloatValue,
TYPE_ENUM: EnumValue,
TYPE_INT32: Int32Value,
TYPE_INT64: Int64Value,
TYPE_STRING: StringValue,
TYPE_UINT32: UInt32Value,
TYPE_UINT64: UInt64Value,
}[proto_type] }[proto_type]

2133
src/betterproto/lib/google/protobuf/__init__.py
View File

File diff suppressed because it is too large Load Diff

125
src/betterproto/lib/google/protobuf/compiler/__init__.py Normal file
View File

@ -0,0 +1,125 @@
# Generated by the protocol buffer compiler. DO NOT EDIT!
# sources: google/protobuf/compiler/plugin.proto
# plugin: python-betterproto
from dataclasses import dataclass
from typing import List
import betterproto
@dataclass(eq=False, repr=False)
class Version(betterproto.Message):
"""The version number of protocol compiler."""
major: int = betterproto.int32_field(1)
minor: int = betterproto.int32_field(2)
patch: int = betterproto.int32_field(3)
# A suffix for alpha, beta or rc release, e.g., "alpha-1", "rc2". It should
# be empty for mainline stable releases.
suffix: str = betterproto.string_field(4)
def __post_init__(self) -> None:
super().__post_init__()
@dataclass(eq=False, repr=False)
class CodeGeneratorRequest(betterproto.Message):
"""An encoded CodeGeneratorRequest is written to the plugin's stdin."""
# The .proto files that were explicitly listed on the command-line. The code
# generator should generate code only for these files. Each file's
# descriptor will be included in proto_file, below.
file_to_generate: List[str] = betterproto.string_field(1)
# The generator parameter passed on the command-line.
parameter: str = betterproto.string_field(2)
# FileDescriptorProtos for all files in files_to_generate and everything they
# import. The files will appear in topological order, so each file appears
# before any file that imports it. protoc guarantees that all proto_files
# will be written after the fields above, even though this is not technically
# guaranteed by the protobuf wire format. This theoretically could allow a
# plugin to stream in the FileDescriptorProtos and handle them one by one
# rather than read the entire set into memory at once. However, as of this
# writing, this is not similarly optimized on protoc's end -- it will store
# all fields in memory at once before sending them to the plugin. Type names
# of fields and extensions in the FileDescriptorProto are always fully
# qualified.
proto_file: List[
"betterproto_lib_google_protobuf.FileDescriptorProto"
] = betterproto.message_field(15)
# The version number of protocol compiler.
compiler_version: "Version" = betterproto.message_field(3)
def __post_init__(self) -> None:
super().__post_init__()
@dataclass(eq=False, repr=False)
class CodeGeneratorResponse(betterproto.Message):
"""The plugin writes an encoded CodeGeneratorResponse to stdout."""
# Error message. If non-empty, code generation failed. The plugin process
# should exit with status code zero even if it reports an error in this way.
# This should be used to indicate errors in .proto files which prevent the
# code generator from generating correct code. Errors which indicate a
# problem in protoc itself -- such as the input CodeGeneratorRequest being
# unparseable -- should be reported by writing a message to stderr and
# exiting with a non-zero status code.
error: str = betterproto.string_field(1)
file: List["CodeGeneratorResponseFile"] = betterproto.message_field(15)
def __post_init__(self) -> None:
super().__post_init__()
@dataclass(eq=False, repr=False)
class CodeGeneratorResponseFile(betterproto.Message):
"""Represents a single generated file."""
# The file name, relative to the output directory. The name must not contain
# "." or ".." components and must be relative, not be absolute (so, the file
# cannot lie outside the output directory). "/" must be used as the path
# separator, not "\". If the name is omitted, the content will be appended to
# the previous file. This allows the generator to break large files into
# small chunks, and allows the generated text to be streamed back to protoc
# so that large files need not reside completely in memory at one time. Note
# that as of this writing protoc does not optimize for this -- it will read
# the entire CodeGeneratorResponse before writing files to disk.
name: str = betterproto.string_field(1)
# If non-empty, indicates that the named file should already exist, and the
# content here is to be inserted into that file at a defined insertion point.
# This feature allows a code generator to extend the output produced by
# another code generator. The original generator may provide insertion
# points by placing special annotations in the file that look like:
# @@protoc_insertion_point(NAME) The annotation can have arbitrary text
# before and after it on the line, which allows it to be placed in a comment.
# NAME should be replaced with an identifier naming the point -- this is what
# other generators will use as the insertion_point. Code inserted at this
# point will be placed immediately above the line containing the insertion
# point (thus multiple insertions to the same point will come out in the
# order they were added). The double-@ is intended to make it unlikely that
# the generated code could contain things that look like insertion points by
# accident. For example, the C++ code generator places the following line in
# the .pb.h files that it generates: //
# @@protoc_insertion_point(namespace_scope) This line appears within the
# scope of the file's package namespace, but outside of any particular class.
# Another plugin can then specify the insertion_point "namespace_scope" to
# generate additional classes or other declarations that should be placed in
# this scope. Note that if the line containing the insertion point begins
# with whitespace, the same whitespace will be added to every line of the
# inserted text. This is useful for languages like Python, where indentation
# matters. In these languages, the insertion point comment should be
# indented the same amount as any inserted code will need to be in order to
# work correctly in that context. The code generator that generates the
# initial file and the one which inserts into it must both run as part of a
# single invocation of protoc. Code generators are executed in the order in
# which they appear on the command line. If |insertion_point| is present,
# |name| must also be present.
insertion_point: str = betterproto.string_field(2)
# The file contents.
content: str = betterproto.string_field(15)
def __post_init__(self) -> None:
super().__post_init__()
import betterproto.lib.google.protobuf as betterproto_lib_google_protobuf

17
src/betterproto/plugin/main.py Normal file → Executable file
View File

@ -3,9 +3,13 @@
import os import os
import sys import sys
from google.protobuf.compiler import plugin_pb2 as plugin from betterproto.lib.google.protobuf.compiler import (
CodeGeneratorRequest,
CodeGeneratorResponse,
)
from betterproto.plugin.parser import generate_code from betterproto.plugin.parser import generate_code
from betterproto.plugin.models import monkey_patch_oneof_index
def main() -> None: def main() -> None:
@ -13,16 +17,19 @@ def main() -> None:
# Read request message from stdin # Read request message from stdin
data = sys.stdin.buffer.read() data = sys.stdin.buffer.read()
# Apply Work around for proto2/3 difference in protoc messages
monkey_patch_oneof_index()
# Parse request # Parse request
request = plugin.CodeGeneratorRequest() request = CodeGeneratorRequest()
request.ParseFromString(data) request.parse(data)
dump_file = os.getenv("BETTERPROTO_DUMP") dump_file = os.getenv("BETTERPROTO_DUMP")
if dump_file: if dump_file:
dump_request(dump_file, request) dump_request(dump_file, request)
# Create response # Create response
response = plugin.CodeGeneratorResponse() response = CodeGeneratorResponse()
# Generate code # Generate code
generate_code(request, response) generate_code(request, response)
@ -34,7 +41,7 @@ def main() -> None:
sys.stdout.buffer.write(output) sys.stdout.buffer.write(output)
def dump_request(dump_file: str, request: plugin.CodeGeneratorRequest) -> None: def dump_request(dump_file: str, request: CodeGeneratorRequest) -> None:
""" """
For developers: Supports running plugin.py standalone so its possible to debug it. For developers: Supports running plugin.py standalone so its possible to debug it.
Run protoc (or generate.py) with BETTERPROTO_DUMP="yourfile.bin" to write the request to a file. Run protoc (or generate.py) with BETTERPROTO_DUMP="yourfile.bin" to write the request to a file.

192
src/betterproto/plugin/models.py
View File

@ -29,12 +29,37 @@ instantiating field `A` with parent message `B` should add a
reference to `A` to `B`'s `fields` attribute. reference to `A` to `B`'s `fields` attribute.
""" """
import betterproto
from betterproto import which_one_of
from betterproto.casing import sanitize_name
from betterproto.compile.importing import (
get_type_reference,
parse_source_type_name,
)
from betterproto.compile.naming import (
pythonize_class_name,
pythonize_field_name,
pythonize_method_name,
)
from betterproto.lib.google.protobuf import (
DescriptorProto,
EnumDescriptorProto,
FileDescriptorProto,
MethodDescriptorProto,
Field,
FieldDescriptorProto,
FieldDescriptorProtoType,
FieldDescriptorProtoLabel,
)
from betterproto.lib.google.protobuf.compiler import CodeGeneratorRequest
import re import re
import textwrap import textwrap
from dataclasses import dataclass, field from dataclasses import dataclass, field
from typing import Dict, Iterator, List, Optional, Set, Text, Type, Union from typing import Dict, Iterator, List, Optional, Set, Text, Type, Union
import sys
import betterproto
from ..casing import sanitize_name from ..casing import sanitize_name
from ..compile.importing import get_type_reference, parse_source_type_name from ..compile.importing import get_type_reference, parse_source_type_name
@ -44,26 +69,6 @@ from ..compile.naming import (
pythonize_method_name, pythonize_method_name,
) )
try:
# betterproto[compiler] specific dependencies
from google.protobuf.compiler import plugin_pb2 as plugin
from google.protobuf.descriptor_pb2 import (
DescriptorProto,
EnumDescriptorProto,
FieldDescriptorProto,
FileDescriptorProto,
MethodDescriptorProto,
)
except ImportError as err:
print(
"\033[31m"
f"Unable to import `{err.name}` from betterproto plugin! "
"Please ensure that you've installed betterproto as "
'`pip install "betterproto[compiler]"` so that compiler dependencies '
"are included."
"\033[0m"
)
raise SystemExit(1)
# Create a unique placeholder to deal with # Create a unique placeholder to deal with
# https://stackoverflow.com/questions/51575931/class-inheritance-in-python-3-7-dataclasses # https://stackoverflow.com/questions/51575931/class-inheritance-in-python-3-7-dataclasses
@ -71,54 +76,75 @@ PLACEHOLDER = object()
# Organize proto types into categories # Organize proto types into categories
PROTO_FLOAT_TYPES = ( PROTO_FLOAT_TYPES = (
FieldDescriptorProto.TYPE_DOUBLE, # 1 FieldDescriptorProtoType.TYPE_DOUBLE, # 1
FieldDescriptorProto.TYPE_FLOAT, # 2 FieldDescriptorProtoType.TYPE_FLOAT, # 2
) )
PROTO_INT_TYPES = ( PROTO_INT_TYPES = (
FieldDescriptorProto.TYPE_INT64, # 3 FieldDescriptorProtoType.TYPE_INT64, # 3
FieldDescriptorProto.TYPE_UINT64, # 4 FieldDescriptorProtoType.TYPE_UINT64, # 4
FieldDescriptorProto.TYPE_INT32, # 5 FieldDescriptorProtoType.TYPE_INT32, # 5
FieldDescriptorProto.TYPE_FIXED64, # 6 FieldDescriptorProtoType.TYPE_FIXED64, # 6
FieldDescriptorProto.TYPE_FIXED32, # 7 FieldDescriptorProtoType.TYPE_FIXED32, # 7
FieldDescriptorProto.TYPE_UINT32, # 13 FieldDescriptorProtoType.TYPE_UINT32, # 13
FieldDescriptorProto.TYPE_SFIXED32, # 15 FieldDescriptorProtoType.TYPE_SFIXED32, # 15
FieldDescriptorProto.TYPE_SFIXED64, # 16 FieldDescriptorProtoType.TYPE_SFIXED64, # 16
FieldDescriptorProto.TYPE_SINT32, # 17 FieldDescriptorProtoType.TYPE_SINT32, # 17
FieldDescriptorProto.TYPE_SINT64, # 18 FieldDescriptorProtoType.TYPE_SINT64, # 18
) )
PROTO_BOOL_TYPES = (FieldDescriptorProto.TYPE_BOOL,) # 8 PROTO_BOOL_TYPES = (FieldDescriptorProtoType.TYPE_BOOL,) # 8
PROTO_STR_TYPES = (FieldDescriptorProto.TYPE_STRING,) # 9 PROTO_STR_TYPES = (FieldDescriptorProtoType.TYPE_STRING,) # 9
PROTO_BYTES_TYPES = (FieldDescriptorProto.TYPE_BYTES,) # 12 PROTO_BYTES_TYPES = (FieldDescriptorProtoType.TYPE_BYTES,) # 12
PROTO_MESSAGE_TYPES = ( PROTO_MESSAGE_TYPES = (
FieldDescriptorProto.TYPE_MESSAGE, # 11 FieldDescriptorProtoType.TYPE_MESSAGE, # 11
FieldDescriptorProto.TYPE_ENUM, # 14 FieldDescriptorProtoType.TYPE_ENUM, # 14
) )
PROTO_MAP_TYPES = (FieldDescriptorProto.TYPE_MESSAGE,) # 11 PROTO_MAP_TYPES = (FieldDescriptorProtoType.TYPE_MESSAGE,) # 11
PROTO_PACKED_TYPES = ( PROTO_PACKED_TYPES = (
FieldDescriptorProto.TYPE_DOUBLE, # 1 FieldDescriptorProtoType.TYPE_DOUBLE, # 1
FieldDescriptorProto.TYPE_FLOAT, # 2 FieldDescriptorProtoType.TYPE_FLOAT, # 2
FieldDescriptorProto.TYPE_INT64, # 3 FieldDescriptorProtoType.TYPE_INT64, # 3
FieldDescriptorProto.TYPE_UINT64, # 4 FieldDescriptorProtoType.TYPE_UINT64, # 4
FieldDescriptorProto.TYPE_INT32, # 5 FieldDescriptorProtoType.TYPE_INT32, # 5
FieldDescriptorProto.TYPE_FIXED64, # 6 FieldDescriptorProtoType.TYPE_FIXED64, # 6
FieldDescriptorProto.TYPE_FIXED32, # 7 FieldDescriptorProtoType.TYPE_FIXED32, # 7
FieldDescriptorProto.TYPE_BOOL, # 8 FieldDescriptorProtoType.TYPE_BOOL, # 8
FieldDescriptorProto.TYPE_UINT32, # 13 FieldDescriptorProtoType.TYPE_UINT32, # 13
FieldDescriptorProto.TYPE_SFIXED32, # 15 FieldDescriptorProtoType.TYPE_SFIXED32, # 15
FieldDescriptorProto.TYPE_SFIXED64, # 16 FieldDescriptorProtoType.TYPE_SFIXED64, # 16
FieldDescriptorProto.TYPE_SINT32, # 17 FieldDescriptorProtoType.TYPE_SINT32, # 17
FieldDescriptorProto.TYPE_SINT64, # 18 FieldDescriptorProtoType.TYPE_SINT64, # 18
) )
def monkey_patch_oneof_index():
"""
The compiler message types are written for proto2, but we read them as proto3.
For this to work in the case of the oneof_index fields, which depend on being able
to tell whether they were set, we have to treat them as oneof fields. This method
monkey patches the generated classes after the fact to force this behaviour.
"""
object.__setattr__(
FieldDescriptorProto.__dataclass_fields__["oneof_index"].metadata[
"betterproto"
],
"group",
"oneof_index",
)
object.__setattr__(
Field.__dataclass_fields__["oneof_index"].metadata["betterproto"],
"group",
"oneof_index",
)
def get_comment( def get_comment(
proto_file: "FileDescriptorProto", path: List[int], indent: int = 4 proto_file: "FileDescriptorProto", path: List[int], indent: int = 4
) -> str: ) -> str:
pad = " " * indent pad = " " * indent
for sci in proto_file.source_code_info.location: for sci_loc in proto_file.source_code_info.location:
if list(sci.path) == path and sci.leading_comments: if list(sci_loc.path) == path and sci_loc.leading_comments:
lines = textwrap.wrap( lines = textwrap.wrap(
sci.leading_comments.strip().replace("\n", ""), width=79 - indent sci_loc.leading_comments.strip().replace("\n", ""), width=79 - indent
) )
if path[-2] == 2 and path[-4] != 6: if path[-2] == 2 and path[-4] != 6:
@ -139,6 +165,7 @@ def get_comment(
class ProtoContentBase: class ProtoContentBase:
"""Methods common to MessageCompiler, ServiceCompiler and ServiceMethodCompiler.""" """Methods common to MessageCompiler, ServiceCompiler and ServiceMethodCompiler."""
source_file: FileDescriptorProto
path: List[int] path: List[int]
comment_indent: int = 4 comment_indent: int = 4
parent: Union["betterproto.Message", "OutputTemplate"] parent: Union["betterproto.Message", "OutputTemplate"]
@ -156,13 +183,6 @@ class ProtoContentBase:
current = current.parent current = current.parent
return current return current
@property
def proto_file(self) -> FieldDescriptorProto:
current = self
while not isinstance(current, OutputTemplate):
current = current.parent
return current.package_proto_obj
@property @property
def request(self) -> "PluginRequestCompiler": def request(self) -> "PluginRequestCompiler":
current = self current = self
@ -176,14 +196,14 @@ class ProtoContentBase:
for this object. for this object.
""" """
return get_comment( return get_comment(
proto_file=self.proto_file, path=self.path, indent=self.comment_indent proto_file=self.source_file, path=self.path, indent=self.comment_indent
) )
@dataclass @dataclass
class PluginRequestCompiler: class PluginRequestCompiler:
plugin_request_obj: plugin.CodeGeneratorRequest plugin_request_obj: CodeGeneratorRequest
output_packages: Dict[str, "OutputTemplate"] = field(default_factory=dict) output_packages: Dict[str, "OutputTemplate"] = field(default_factory=dict)
@property @property
@ -253,6 +273,7 @@ class OutputTemplate:
class MessageCompiler(ProtoContentBase): class MessageCompiler(ProtoContentBase):
"""Representation of a protobuf message.""" """Representation of a protobuf message."""
source_file: FileDescriptorProto
parent: Union["MessageCompiler", OutputTemplate] = PLACEHOLDER parent: Union["MessageCompiler", OutputTemplate] = PLACEHOLDER
proto_obj: DescriptorProto = PLACEHOLDER proto_obj: DescriptorProto = PLACEHOLDER
path: List[int] = PLACEHOLDER path: List[int] = PLACEHOLDER
@ -296,7 +317,7 @@ def is_map(
proto_field_obj: FieldDescriptorProto, parent_message: DescriptorProto proto_field_obj: FieldDescriptorProto, parent_message: DescriptorProto
) -> bool: ) -> bool:
"""True if proto_field_obj is a map, otherwise False.""" """True if proto_field_obj is a map, otherwise False."""
if proto_field_obj.type == FieldDescriptorProto.TYPE_MESSAGE: if proto_field_obj.type == FieldDescriptorProtoType.TYPE_MESSAGE:
# This might be a map... # This might be a map...
message_type = proto_field_obj.type_name.split(".").pop().lower() message_type = proto_field_obj.type_name.split(".").pop().lower()
map_entry = f"{proto_field_obj.name.replace('_', '').lower()}entry" map_entry = f"{proto_field_obj.name.replace('_', '').lower()}entry"
@ -311,8 +332,20 @@ def is_map(
def is_oneof(proto_field_obj: FieldDescriptorProto) -> bool: def is_oneof(proto_field_obj: FieldDescriptorProto) -> bool:
"""True if proto_field_obj is a OneOf, otherwise False.""" """
return proto_field_obj.HasField("oneof_index") True if proto_field_obj is a OneOf, otherwise False.
.. warning::
Becuase the message from protoc is defined in proto2, and betterproto works with
proto3, and interpreting the FieldDescriptorProto.oneof_index field requires
distinguishing between default and unset values (which proto3 doesn't support),
we have to hack the generated FieldDescriptorProto class for this to work.
The hack consists of setting group="oneof_index" in the field metadata,
essentially making oneof_index the sole member of a one_of group, which allows
us to tell whether it was set, via the which_one_of interface.
"""
return which_one_of(proto_field_obj, "oneof_index")[0] == "oneof_index"
@dataclass @dataclass
@ -377,7 +410,7 @@ class FieldCompiler(MessageCompiler):
def field_wraps(self) -> Optional[str]: def field_wraps(self) -> Optional[str]:
"""Returns betterproto wrapped field type or None.""" """Returns betterproto wrapped field type or None."""
match_wrapper = re.match( match_wrapper = re.match(
r"\.google\.protobuf\.(.+)Value", self.proto_obj.type_name r"\.google\.protobuf\.(.+)Value$", self.proto_obj.type_name
) )
if match_wrapper: if match_wrapper:
wrapped_type = "TYPE_" + match_wrapper.group(1).upper() wrapped_type = "TYPE_" + match_wrapper.group(1).upper()
@ -388,7 +421,7 @@ class FieldCompiler(MessageCompiler):
@property @property
def repeated(self) -> bool: def repeated(self) -> bool:
return ( return (
self.proto_obj.label == FieldDescriptorProto.LABEL_REPEATED self.proto_obj.label == FieldDescriptorProtoLabel.LABEL_REPEATED
and not is_map(self.proto_obj, self.parent) and not is_map(self.proto_obj, self.parent)
) )
@ -401,7 +434,9 @@ class FieldCompiler(MessageCompiler):
def field_type(self) -> str: def field_type(self) -> str:
"""String representation of proto field type.""" """String representation of proto field type."""
return ( return (
self.proto_obj.Type.Name(self.proto_obj.type).lower().replace("type_", "") FieldDescriptorProtoType(self.proto_obj.type)
.name.lower()
.replace("type_", "")
) )
@property @property
@ -495,14 +530,19 @@ class MapEntryCompiler(FieldCompiler):
): ):
# Get Python types # Get Python types
self.py_k_type = FieldCompiler( self.py_k_type = FieldCompiler(
parent=self, proto_obj=nested.field[0] # key source_file=self.source_file,
parent=self,
proto_obj=nested.field[0], # key
).py_type ).py_type
self.py_v_type = FieldCompiler( self.py_v_type = FieldCompiler(
parent=self, proto_obj=nested.field[1] # value source_file=self.source_file,
parent=self,
proto_obj=nested.field[1], # value
).py_type ).py_type
# Get proto types # Get proto types
self.proto_k_type = self.proto_obj.Type.Name(nested.field[0].type) self.proto_k_type = FieldDescriptorProtoType(nested.field[0].type).name
self.proto_v_type = self.proto_obj.Type.Name(nested.field[1].type) self.proto_v_type = FieldDescriptorProtoType(nested.field[1].type).name
super().__post_init__() # call FieldCompiler-> MessageCompiler __post_init__ super().__post_init__() # call FieldCompiler-> MessageCompiler __post_init__
@property @property
@ -544,7 +584,7 @@ class EnumDefinitionCompiler(MessageCompiler):
name=sanitize_name(entry_proto_value.name), name=sanitize_name(entry_proto_value.name),
value=entry_proto_value.number, value=entry_proto_value.number,
comment=get_comment( comment=get_comment(
proto_file=self.proto_file, path=self.path + [2, entry_number] proto_file=self.source_file, path=self.path + [2, entry_number]
), ),
) )
for entry_number, entry_proto_value in enumerate(self.proto_obj.value) for entry_number, entry_proto_value in enumerate(self.proto_obj.value)

87
src/betterproto/plugin/parser.py
View File

@ -1,28 +1,19 @@
from betterproto.lib.google.protobuf import (
DescriptorProto,
EnumDescriptorProto,
FieldDescriptorProto,
FileDescriptorProto,
ServiceDescriptorProto,
)
from betterproto.lib.google.protobuf.compiler import (
CodeGeneratorRequest,
CodeGeneratorResponse,
CodeGeneratorResponseFile,
)
import itertools import itertools
import pathlib import pathlib
import sys import sys
from typing import TYPE_CHECKING, Iterator, List, Tuple, Union, Set from typing import Iterator, List, Tuple, TYPE_CHECKING, Union
try:
# betterproto[compiler] specific dependencies
from google.protobuf.compiler import plugin_pb2 as plugin
from google.protobuf.descriptor_pb2 import (
DescriptorProto,
EnumDescriptorProto,
FieldDescriptorProto,
ServiceDescriptorProto,
)
except ImportError as err:
print(
"\033[31m"
f"Unable to import `{err.name}` from betterproto plugin! "
"Please ensure that you've installed betterproto as "
'`pip install "betterproto[compiler]"` so that compiler dependencies '
"are included."
"\033[0m"
)
raise SystemExit(1)
from .compiler import outputfile_compiler from .compiler import outputfile_compiler
from .models import ( from .models import (
EnumDefinitionCompiler, EnumDefinitionCompiler,
@ -70,7 +61,7 @@ def traverse(
def generate_code( def generate_code(
request: plugin.CodeGeneratorRequest, response: plugin.CodeGeneratorResponse request: CodeGeneratorRequest, response: CodeGeneratorResponse
) -> None: ) -> None:
plugin_options = request.parameter.split(",") if request.parameter else [] plugin_options = request.parameter.split(",") if request.parameter else []
@ -100,7 +91,12 @@ def generate_code(
for output_package_name, output_package in request_data.output_packages.items(): for output_package_name, output_package in request_data.output_packages.items():
for proto_input_file in output_package.input_files: for proto_input_file in output_package.input_files:
for item, path in traverse(proto_input_file): for item, path in traverse(proto_input_file):
read_protobuf_type(item=item, path=path, output_package=output_package) read_protobuf_type(
source_file=proto_input_file,
item=item,
path=path,
output_package=output_package,
)
# Read Services # Read Services
for output_package_name, output_package in request_data.output_packages.items(): for output_package_name, output_package in request_data.output_packages.items():
@ -116,11 +112,13 @@ def generate_code(
output_path = pathlib.Path(*output_package_name.split("."), "__init__.py") output_path = pathlib.Path(*output_package_name.split("."), "__init__.py")
output_paths.add(output_path) output_paths.add(output_path)
f: response.File = response.file.add() response.file.append(
f.name = str(output_path) CodeGeneratorResponseFile(
name=str(output_path),
# Render and then format the output file # Render and then format the output file
f.content = outputfile_compiler(output_file=output_package) content=outputfile_compiler(output_file=output_package),
)
)
# Make each output directory a package with __init__ file # Make each output directory a package with __init__ file
init_files = { init_files = {
@ -130,38 +128,53 @@ def generate_code(
} - output_paths } - output_paths
for init_file in init_files: for init_file in init_files:
init = response.file.add() response.file.append(CodeGeneratorResponseFile(name=str(init_file)))
init.name = str(init_file)
for output_package_name in sorted(output_paths.union(init_files)): for output_package_name in sorted(output_paths.union(init_files)):
print(f"Writing {output_package_name}", file=sys.stderr) print(f"Writing {output_package_name}", file=sys.stderr)
def read_protobuf_type( def read_protobuf_type(
item: DescriptorProto, path: List[int], output_package: OutputTemplate item: DescriptorProto,
path: List[int],
source_file: "FileDescriptorProto",
output_package: OutputTemplate,
) -> None: ) -> None:
if isinstance(item, DescriptorProto): if isinstance(item, DescriptorProto):
if item.options.map_entry: if item.options.map_entry:
# Skip generated map entry messages since we just use dicts # Skip generated map entry messages since we just use dicts
return return
# Process Message # Process Message
message_data = MessageCompiler(parent=output_package, proto_obj=item, path=path) message_data = MessageCompiler(
source_file=source_file, parent=output_package, proto_obj=item, path=path
)
for index, field in enumerate(item.field): for index, field in enumerate(item.field):
if is_map(field, item): if is_map(field, item):
MapEntryCompiler( MapEntryCompiler(
parent=message_data, proto_obj=field, path=path + [2, index] source_file=source_file,
parent=message_data,
proto_obj=field,
path=path + [2, index],
) )
elif is_oneof(field): elif is_oneof(field):
OneOfFieldCompiler( OneOfFieldCompiler(
parent=message_data, proto_obj=field, path=path + [2, index] source_file=source_file,
parent=message_data,
proto_obj=field,
path=path + [2, index],
) )
else: else:
FieldCompiler( FieldCompiler(
parent=message_data, proto_obj=field, path=path + [2, index] source_file=source_file,
parent=message_data,
proto_obj=field,
path=path + [2, index],
) )
elif isinstance(item, EnumDescriptorProto): elif isinstance(item, EnumDescriptorProto):
# Enum # Enum
EnumDefinitionCompiler(parent=output_package, proto_obj=item, path=path) EnumDefinitionCompiler(
source_file=source_file, parent=output_package, proto_obj=item, path=path
)
def read_protobuf_service( def read_protobuf_service(

911
tests/inputs/example/example.proto
View File

@ -1,8 +1,909 @@
syntax = "proto3"; // Protocol Buffers - Google's data interchange format
// Copyright 2008 Google Inc. All rights reserved.
// https://developers.google.com/protocol-buffers/
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
// * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
// * Redistributions in binary form must reproduce the above
// copyright notice, this list of conditions and the following disclaimer
// in the documentation and/or other materials provided with the
// distribution.
// * Neither the name of Google Inc. nor the names of its
// contributors may be used to endorse or promote products derived from
// this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
package hello; // Author: kenton@google.com (Kenton Varda)
// Based on original Protocol Buffers design by
// Sanjay Ghemawat, Jeff Dean, and others.
//
// The messages in this file describe the definitions found in .proto files.
// A valid .proto file can be translated directly to a FileDescriptorProto
// without any other information (e.g. without reading its imports).
// Greeting represents a message you can tell a user.
message Greeting { syntax = "proto2";
string message = 1;
// package google.protobuf;
option go_package = "google.golang.org/protobuf/types/descriptorpb";
option java_package = "com.google.protobuf";
option java_outer_classname = "DescriptorProtos";
option csharp_namespace = "Google.Protobuf.Reflection";
option objc_class_prefix = "GPB";
option cc_enable_arenas = true;
// descriptor.proto must be optimized for speed because reflection-based
// algorithms don't work during bootstrapping.
option optimize_for = SPEED;
// The protocol compiler can output a FileDescriptorSet containing the .proto
// files it parses.
message FileDescriptorSet {
repeated FileDescriptorProto file = 1;
}
// Describes a complete .proto file.
message FileDescriptorProto {
optional string name = 1; // file name, relative to root of source tree
optional string package = 2; // e.g. "foo", "foo.bar", etc.
// Names of files imported by this file.
repeated string dependency = 3;
// Indexes of the public imported files in the dependency list above.
repeated int32 public_dependency = 10;
// Indexes of the weak imported files in the dependency list.
// For Google-internal migration only. Do not use.
repeated int32 weak_dependency = 11;
// All top-level definitions in this file.
repeated DescriptorProto message_type = 4;
repeated EnumDescriptorProto enum_type = 5;
repeated ServiceDescriptorProto service = 6;
repeated FieldDescriptorProto extension = 7;
optional FileOptions options = 8;
// This field contains optional information about the original source code.
// You may safely remove this entire field without harming runtime
// functionality of the descriptors -- the information is needed only by
// development tools.
optional SourceCodeInfo source_code_info = 9;
// The syntax of the proto file.
// The supported values are "proto2" and "proto3".
optional string syntax = 12;
}
// Describes a message type.
message DescriptorProto {
optional string name = 1;
repeated FieldDescriptorProto field = 2;
repeated FieldDescriptorProto extension = 6;
repeated DescriptorProto nested_type = 3;
repeated EnumDescriptorProto enum_type = 4;
message ExtensionRange {
optional int32 start = 1; // Inclusive.
optional int32 end = 2; // Exclusive.
optional ExtensionRangeOptions options = 3;
}
repeated ExtensionRange extension_range = 5;
repeated OneofDescriptorProto oneof_decl = 8;
optional MessageOptions options = 7;
// Range of reserved tag numbers. Reserved tag numbers may not be used by
// fields or extension ranges in the same message. Reserved ranges may
// not overlap.
message ReservedRange {
optional int32 start = 1; // Inclusive.
optional int32 end = 2; // Exclusive.
}
repeated ReservedRange reserved_range = 9;
// Reserved field names, which may not be used by fields in the same message.
// A given name may only be reserved once.
repeated string reserved_name = 10;
}
message ExtensionRangeOptions {
// The parser stores options it doesn't recognize here. See above.
repeated UninterpretedOption uninterpreted_option = 999;
// Clients can define custom options in extensions of this message. See above.
extensions 1000 to max;
}
// Describes a field within a message.
message FieldDescriptorProto {
enum Type {
// 0 is reserved for errors.
// Order is weird for historical reasons.
TYPE_DOUBLE = 1;
TYPE_FLOAT = 2;
// Not ZigZag encoded. Negative numbers take 10 bytes. Use TYPE_SINT64 if
// negative values are likely.
TYPE_INT64 = 3;
TYPE_UINT64 = 4;
// Not ZigZag encoded. Negative numbers take 10 bytes. Use TYPE_SINT32 if
// negative values are likely.
TYPE_INT32 = 5;
TYPE_FIXED64 = 6;
TYPE_FIXED32 = 7;
TYPE_BOOL = 8;
TYPE_STRING = 9;
// Tag-delimited aggregate.
// Group type is deprecated and not supported in proto3. However, Proto3
// implementations should still be able to parse the group wire format and
// treat group fields as unknown fields.
TYPE_GROUP = 10;
TYPE_MESSAGE = 11; // Length-delimited aggregate.
// New in version 2.
TYPE_BYTES = 12;
TYPE_UINT32 = 13;
TYPE_ENUM = 14;
TYPE_SFIXED32 = 15;
TYPE_SFIXED64 = 16;
TYPE_SINT32 = 17; // Uses ZigZag encoding.
TYPE_SINT64 = 18; // Uses ZigZag encoding.
}
enum Label {
// 0 is reserved for errors
LABEL_OPTIONAL = 1;
LABEL_REQUIRED = 2;
LABEL_REPEATED = 3;
}
optional string name = 1;
optional int32 number = 3;
optional Label label = 4;
// If type_name is set, this need not be set. If both this and type_name
// are set, this must be one of TYPE_ENUM, TYPE_MESSAGE or TYPE_GROUP.
optional Type type = 5;
// For message and enum types, this is the name of the type. If the name
// starts with a '.', it is fully-qualified. Otherwise, C++-like scoping
// rules are used to find the type (i.e. first the nested types within this
// message are searched, then within the parent, on up to the root
// namespace).
optional string type_name = 6;
// For extensions, this is the name of the type being extended. It is
// resolved in the same manner as type_name.
optional string extendee = 2;
// For numeric types, contains the original text representation of the value.
// For booleans, "true" or "false".
// For strings, contains the default text contents (not escaped in any way).
// For bytes, contains the C escaped value. All bytes >= 128 are escaped.
// TODO(kenton): Base-64 encode?
optional string default_value = 7;
// If set, gives the index of a oneof in the containing type's oneof_decl
// list. This field is a member of that oneof.
optional int32 oneof_index = 9;
// JSON name of this field. The value is set by protocol compiler. If the
// user has set a "json_name" option on this field, that option's value
// will be used. Otherwise, it's deduced from the field's name by converting
// it to camelCase.
optional string json_name = 10;
optional FieldOptions options = 8;
// If true, this is a proto3 "optional". When a proto3 field is optional, it
// tracks presence regardless of field type.
//
// When proto3_optional is true, this field must be belong to a oneof to
// signal to old proto3 clients that presence is tracked for this field. This
// oneof is known as a "synthetic" oneof, and this field must be its sole
// member (each proto3 optional field gets its own synthetic oneof). Synthetic
// oneofs exist in the descriptor only, and do not generate any API. Synthetic
// oneofs must be ordered after all "real" oneofs.
//
// For message fields, proto3_optional doesn't create any semantic change,
// since non-repeated message fields always track presence. However it still
// indicates the semantic detail of whether the user wrote "optional" or not.
// This can be useful for round-tripping the .proto file. For consistency we
// give message fields a synthetic oneof also, even though it is not required
// to track presence. This is especially important because the parser can't
// tell if a field is a message or an enum, so it must always create a
// synthetic oneof.
//
// Proto2 optional fields do not set this flag, because they already indicate
// optional with `LABEL_OPTIONAL`.
optional bool proto3_optional = 17;
}
// Describes a oneof.
message OneofDescriptorProto {
optional string name = 1;
optional OneofOptions options = 2;
}
// Describes an enum type.
message EnumDescriptorProto {
optional string name = 1;
repeated EnumValueDescriptorProto value = 2;
optional EnumOptions options = 3;
// Range of reserved numeric values. Reserved values may not be used by
// entries in the same enum. Reserved ranges may not overlap.
//
// Note that this is distinct from DescriptorProto.ReservedRange in that it
// is inclusive such that it can appropriately represent the entire int32
// domain.
message EnumReservedRange {
optional int32 start = 1; // Inclusive.
optional int32 end = 2; // Inclusive.
}
// Range of reserved numeric values. Reserved numeric values may not be used
// by enum values in the same enum declaration. Reserved ranges may not
// overlap.
repeated EnumReservedRange reserved_range = 4;
// Reserved enum value names, which may not be reused. A given name may only
// be reserved once.
repeated string reserved_name = 5;
}
// Describes a value within an enum.
message EnumValueDescriptorProto {
optional string name = 1;
optional int32 number = 2;
optional EnumValueOptions options = 3;
}
// Describes a service.
message ServiceDescriptorProto {
optional string name = 1;
repeated MethodDescriptorProto method = 2;
optional ServiceOptions options = 3;
}
// Describes a method of a service.
message MethodDescriptorProto {
optional string name = 1;
// Input and output type names. These are resolved in the same way as
// FieldDescriptorProto.type_name, but must refer to a message type.
optional string input_type = 2;
optional string output_type = 3;
optional MethodOptions options = 4;
// Identifies if client streams multiple client messages
optional bool client_streaming = 5 [default = false];
// Identifies if server streams multiple server messages
optional bool server_streaming = 6 [default = false];
}
// ===================================================================
// Options
// Each of the definitions above may have "options" attached. These are
// just annotations which may cause code to be generated slightly differently
// or may contain hints for code that manipulates protocol messages.
//
// Clients may define custom options as extensions of the *Options messages.
// These extensions may not yet be known at parsing time, so the parser cannot
// store the values in them. Instead it stores them in a field in the *Options
// message called uninterpreted_option. This field must have the same name
// across all *Options messages. We then use this field to populate the
// extensions when we build a descriptor, at which point all protos have been
// parsed and so all extensions are known.
//
// Extension numbers for custom options may be chosen as follows:
// * For options which will only be used within a single application or
// organization, or for experimental options, use field numbers 50000
// through 99999. It is up to you to ensure that you do not use the
// same number for multiple options.
// * For options which will be published and used publicly by multiple
// independent entities, e-mail protobuf-global-extension-registry@google.com
// to reserve extension numbers. Simply provide your project name (e.g.
// Objective-C plugin) and your project website (if available) -- there's no
// need to explain how you intend to use them. Usually you only need one
// extension number. You can declare multiple options with only one extension
// number by putting them in a sub-message. See the Custom Options section of
// the docs for examples:
// https://developers.google.com/protocol-buffers/docs/proto#options
// If this turns out to be popular, a web service will be set up
// to automatically assign option numbers.
message FileOptions {
// Sets the Java package where classes generated from this .proto will be
// placed. By default, the proto package is used, but this is often
// inappropriate because proto packages do not normally start with backwards
// domain names.
optional string java_package = 1;
// If set, all the classes from the .proto file are wrapped in a single
// outer class with the given name. This applies to both Proto1
// (equivalent to the old "--one_java_file" option) and Proto2 (where
// a .proto always translates to a single class, but you may want to
// explicitly choose the class name).
optional string java_outer_classname = 8;
// If set true, then the Java code generator will generate a separate .java
// file for each top-level message, enum, and service defined in the .proto
// file. Thus, these types will *not* be nested inside the outer class
// named by java_outer_classname. However, the outer class will still be
// generated to contain the file's getDescriptor() method as well as any
// top-level extensions defined in the file.
optional bool java_multiple_files = 10 [default = false];
// This option does nothing.
optional bool java_generate_equals_and_hash = 20 [deprecated=true];
// If set true, then the Java2 code generator will generate code that
// throws an exception whenever an attempt is made to assign a non-UTF-8
// byte sequence to a string field.
// Message reflection will do the same.
// However, an extension field still accepts non-UTF-8 byte sequences.
// This option has no effect on when used with the lite runtime.
optional bool java_string_check_utf8 = 27 [default = false];
// Generated classes can be optimized for speed or code size.
enum OptimizeMode {
SPEED = 1; // Generate complete code for parsing, serialization,
// etc.
CODE_SIZE = 2; // Use ReflectionOps to implement these methods.
LITE_RUNTIME = 3; // Generate code using MessageLite and the lite runtime.
}
optional OptimizeMode optimize_for = 9 [default = SPEED];
// Sets the Go package where structs generated from this .proto will be
// placed. If omitted, the Go package will be derived from the following:
// - The basename of the package import path, if provided.
// - Otherwise, the package statement in the .proto file, if present.
// - Otherwise, the basename of the .proto file, without extension.
optional string go_package = 11;
// Should generic services be generated in each language? "Generic" services
// are not specific to any particular RPC system. They are generated by the
// main code generators in each language (without additional plugins).
// Generic services were the only kind of service generation supported by
// early versions of google.protobuf.
//
// Generic services are now considered deprecated in favor of using plugins
// that generate code specific to your particular RPC system. Therefore,
// these default to false. Old code which depends on generic services should
// explicitly set them to true.
optional bool cc_generic_services = 16 [default = false];
optional bool java_generic_services = 17 [default = false];
optional bool py_generic_services = 18 [default = false];
optional bool php_generic_services = 42 [default = false];
// Is this file deprecated?
// Depending on the target platform, this can emit Deprecated annotations
// for everything in the file, or it will be completely ignored; in the very
// least, this is a formalization for deprecating files.
optional bool deprecated = 23 [default = false];
// Enables the use of arenas for the proto messages in this file. This applies
// only to generated classes for C++.
optional bool cc_enable_arenas = 31 [default = true];
// Sets the objective c class prefix which is prepended to all objective c
// generated classes from this .proto. There is no default.
optional string objc_class_prefix = 36;
// Namespace for generated classes; defaults to the package.
optional string csharp_namespace = 37;
// By default Swift generators will take the proto package and CamelCase it
// replacing '.' with underscore and use that to prefix the types/symbols
// defined. When this options is provided, they will use this value instead
// to prefix the types/symbols defined.
optional string swift_prefix = 39;
// Sets the php class prefix which is prepended to all php generated classes
// from this .proto. Default is empty.
optional string php_class_prefix = 40;
// Use this option to change the namespace of php generated classes. Default
// is empty. When this option is empty, the package name will be used for
// determining the namespace.
optional string php_namespace = 41;
// Use this option to change the namespace of php generated metadata classes.
// Default is empty. When this option is empty, the proto file name will be
// used for determining the namespace.
optional string php_metadata_namespace = 44;
// Use this option to change the package of ruby generated classes. Default
// is empty. When this option is not set, the package name will be used for
// determining the ruby package.
optional string ruby_package = 45;
// The parser stores options it doesn't recognize here.
// See the documentation for the "Options" section above.
repeated UninterpretedOption uninterpreted_option = 999;
// Clients can define custom options in extensions of this message.
// See the documentation for the "Options" section above.
extensions 1000 to max;
reserved 38;
}
message MessageOptions {
// Set true to use the old proto1 MessageSet wire format for extensions.
// This is provided for backwards-compatibility with the MessageSet wire
// format. You should not use this for any other reason: It's less
// efficient, has fewer features, and is more complicated.
//
// The message must be defined exactly as follows:
// message Foo {
// option message_set_wire_format = true;
// extensions 4 to max;
// }
// Note that the message cannot have any defined fields; MessageSets only
// have extensions.
//
// All extensions of your type must be singular messages; e.g. they cannot
// be int32s, enums, or repeated messages.
//
// Because this is an option, the above two restrictions are not enforced by
// the protocol compiler.
optional bool message_set_wire_format = 1 [default = false];
// Disables the generation of the standard "descriptor()" accessor, which can
// conflict with a field of the same name. This is meant to make migration
// from proto1 easier; new code should avoid fields named "descriptor".
optional bool no_standard_descriptor_accessor = 2 [default = false];
// Is this message deprecated?
// Depending on the target platform, this can emit Deprecated annotations
// for the message, or it will be completely ignored; in the very least,
// this is a formalization for deprecating messages.
optional bool deprecated = 3 [default = false];
// Whether the message is an automatically generated map entry type for the
// maps field.
//
// For maps fields:
// map<KeyType, ValueType> map_field = 1;
// The parsed descriptor looks like:
// message MapFieldEntry {
// option map_entry = true;
// optional KeyType key = 1;
// optional ValueType value = 2;
// }
// repeated MapFieldEntry map_field = 1;
//
// Implementations may choose not to generate the map_entry=true message, but
// use a native map in the target language to hold the keys and values.
// The reflection APIs in such implementations still need to work as
// if the field is a repeated message field.
//
// NOTE: Do not set the option in .proto files. Always use the maps syntax
// instead. The option should only be implicitly set by the proto compiler
// parser.
optional bool map_entry = 7;
reserved 8; // javalite_serializable
reserved 9; // javanano_as_lite
// The parser stores options it doesn't recognize here. See above.
repeated UninterpretedOption uninterpreted_option = 999;
// Clients can define custom options in extensions of this message. See above.
extensions 1000 to max;
}
message FieldOptions {
// The ctype option instructs the C++ code generator to use a different
// representation of the field than it normally would. See the specific
// options below. This option is not yet implemented in the open source
// release -- sorry, we'll try to include it in a future version!
optional CType ctype = 1 [default = STRING];
enum CType {
// Default mode.
STRING = 0;
CORD = 1;
STRING_PIECE = 2;
}
// The packed option can be enabled for repeated primitive fields to enable
// a more efficient representation on the wire. Rather than repeatedly
// writing the tag and type for each element, the entire array is encoded as
// a single length-delimited blob. In proto3, only explicit setting it to
// false will avoid using packed encoding.
optional bool packed = 2;
// The jstype option determines the JavaScript type used for values of the
// field. The option is permitted only for 64 bit integral and fixed types
// (int64, uint64, sint64, fixed64, sfixed64). A field with jstype JS_STRING
// is represented as JavaScript string, which avoids loss of precision that
// can happen when a large value is converted to a floating point JavaScript.
// Specifying JS_NUMBER for the jstype causes the generated JavaScript code to
// use the JavaScript "number" type. The behavior of the default option
// JS_NORMAL is implementation dependent.
//
// This option is an enum to permit additional types to be added, e.g.
// goog.math.Integer.
optional JSType jstype = 6 [default = JS_NORMAL];
enum JSType {
// Use the default type.
JS_NORMAL = 0;
// Use JavaScript strings.
JS_STRING = 1;
// Use JavaScript numbers.
JS_NUMBER = 2;
}
// Should this field be parsed lazily? Lazy applies only to message-type
// fields. It means that when the outer message is initially parsed, the
// inner message's contents will not be parsed but instead stored in encoded
// form. The inner message will actually be parsed when it is first accessed.
//
// This is only a hint. Implementations are free to choose whether to use
// eager or lazy parsing regardless of the value of this option. However,
// setting this option true suggests that the protocol author believes that
// using lazy parsing on this field is worth the additional bookkeeping
// overhead typically needed to implement it.
//
// This option does not affect the public interface of any generated code;
// all method signatures remain the same. Furthermore, thread-safety of the
// interface is not affected by this option; const methods remain safe to
// call from multiple threads concurrently, while non-const methods continue
// to require exclusive access.
//
//
// Note that implementations may choose not to check required fields within
// a lazy sub-message. That is, calling IsInitialized() on the outer message
// may return true even if the inner message has missing required fields.
// This is necessary because otherwise the inner message would have to be
// parsed in order to perform the check, defeating the purpose of lazy
// parsing. An implementation which chooses not to check required fields
// must be consistent about it. That is, for any particular sub-message, the
// implementation must either *always* check its required fields, or *never*
// check its required fields, regardless of whether or not the message has
// been parsed.
optional bool lazy = 5 [default = false];
// Is this field deprecated?
// Depending on the target platform, this can emit Deprecated annotations
// for accessors, or it will be completely ignored; in the very least, this
// is a formalization for deprecating fields.
optional bool deprecated = 3 [default = false];
// For Google-internal migration only. Do not use.
optional bool weak = 10 [default = false];
// The parser stores options it doesn't recognize here. See above.
repeated UninterpretedOption uninterpreted_option = 999;
// Clients can define custom options in extensions of this message. See above.
extensions 1000 to max;
reserved 4; // removed jtype
}
message OneofOptions {
// The parser stores options it doesn't recognize here. See above.
repeated UninterpretedOption uninterpreted_option = 999;
// Clients can define custom options in extensions of this message. See above.
extensions 1000 to max;
}
message EnumOptions {
// Set this option to true to allow mapping different tag names to the same
// value.
optional bool allow_alias = 2;
// Is this enum deprecated?
// Depending on the target platform, this can emit Deprecated annotations
// for the enum, or it will be completely ignored; in the very least, this
// is a formalization for deprecating enums.
optional bool deprecated = 3 [default = false];
reserved 5; // javanano_as_lite
// The parser stores options it doesn't recognize here. See above.
repeated UninterpretedOption uninterpreted_option = 999;
// Clients can define custom options in extensions of this message. See above.
extensions 1000 to max;
}
message EnumValueOptions {
// Is this enum value deprecated?
// Depending on the target platform, this can emit Deprecated annotations
// for the enum value, or it will be completely ignored; in the very least,
// this is a formalization for deprecating enum values.
optional bool deprecated = 1 [default = false];
// The parser stores options it doesn't recognize here. See above.
repeated UninterpretedOption uninterpreted_option = 999;
// Clients can define custom options in extensions of this message. See above.
extensions 1000 to max;
}
message ServiceOptions {
// Note: Field numbers 1 through 32 are reserved for Google's internal RPC
// framework. We apologize for hoarding these numbers to ourselves, but
// we were already using them long before we decided to release Protocol
// Buffers.
// Is this service deprecated?
// Depending on the target platform, this can emit Deprecated annotations
// for the service, or it will be completely ignored; in the very least,
// this is a formalization for deprecating services.
optional bool deprecated = 33 [default = false];
// The parser stores options it doesn't recognize here. See above.
repeated UninterpretedOption uninterpreted_option = 999;
// Clients can define custom options in extensions of this message. See above.
extensions 1000 to max;
}
message MethodOptions {
// Note: Field numbers 1 through 32 are reserved for Google's internal RPC
// framework. We apologize for hoarding these numbers to ourselves, but
// we were already using them long before we decided to release Protocol
// Buffers.
// Is this method deprecated?
// Depending on the target platform, this can emit Deprecated annotations
// for the method, or it will be completely ignored; in the very least,
// this is a formalization for deprecating methods.
optional bool deprecated = 33 [default = false];
// Is this method side-effect-free (or safe in HTTP parlance), or idempotent,
// or neither? HTTP based RPC implementation may choose GET verb for safe
// methods, and PUT verb for idempotent methods instead of the default POST.
enum IdempotencyLevel {
IDEMPOTENCY_UNKNOWN = 0;
NO_SIDE_EFFECTS = 1; // implies idempotent
IDEMPOTENT = 2; // idempotent, but may have side effects
}
optional IdempotencyLevel idempotency_level = 34
[default = IDEMPOTENCY_UNKNOWN];
// The parser stores options it doesn't recognize here. See above.
repeated UninterpretedOption uninterpreted_option = 999;
// Clients can define custom options in extensions of this message. See above.
extensions 1000 to max;
}
// A message representing a option the parser does not recognize. This only
// appears in options protos created by the compiler::Parser class.
// DescriptorPool resolves these when building Descriptor objects. Therefore,
// options protos in descriptor objects (e.g. returned by Descriptor::options(),
// or produced by Descriptor::CopyTo()) will never have UninterpretedOptions
// in them.
message UninterpretedOption {
// The name of the uninterpreted option. Each string represents a segment in
// a dot-separated name. is_extension is true iff a segment represents an
// extension (denoted with parentheses in options specs in .proto files).
// E.g.,{ ["foo", false], ["bar.baz", true], ["qux", false] } represents
// "foo.(bar.baz).qux".
message NamePart {
required string name_part = 1;
required bool is_extension = 2;
}
repeated NamePart name = 2;
// The value of the uninterpreted option, in whatever type the tokenizer
// identified it as during parsing. Exactly one of these should be set.
optional string identifier_value = 3;
optional uint64 positive_int_value = 4;
optional int64 negative_int_value = 5;
optional double double_value = 6;
optional bytes string_value = 7;
optional string aggregate_value = 8;
}
// ===================================================================
// Optional source code info
// Encapsulates information about the original source file from which a
// FileDescriptorProto was generated.
message SourceCodeInfo {
// A Location identifies a piece of source code in a .proto file which
// corresponds to a particular definition. This information is intended
// to be useful to IDEs, code indexers, documentation generators, and similar
// tools.
//
// For example, say we have a file like:
// message Foo {
// optional string foo = 1;
// }
// Let's look at just the field definition:
// optional string foo = 1;
// ^ ^^ ^^ ^ ^^^
// a bc de f ghi
// We have the following locations:
// span path represents
// [a,i) [ 4, 0, 2, 0 ] The whole field definition.
// [a,b) [ 4, 0, 2, 0, 4 ] The label (optional).
// [c,d) [ 4, 0, 2, 0, 5 ] The type (string).
// [e,f) [ 4, 0, 2, 0, 1 ] The name (foo).
// [g,h) [ 4, 0, 2, 0, 3 ] The number (1).
//
// Notes:
// - A location may refer to a repeated field itself (i.e. not to any
// particular index within it). This is used whenever a set of elements are
// logically enclosed in a single code segment. For example, an entire
// extend block (possibly containing multiple extension definitions) will
// have an outer location whose path refers to the "extensions" repeated
// field without an index.
// - Multiple locations may have the same path. This happens when a single
// logical declaration is spread out across multiple places. The most
// obvious example is the "extend" block again -- there may be multiple
// extend blocks in the same scope, each of which will have the same path.
// - A location's span is not always a subset of its parent's span. For
// example, the "extendee" of an extension declaration appears at the
// beginning of the "extend" block and is shared by all extensions within
// the block.
// - Just because a location's span is a subset of some other location's span
// does not mean that it is a descendant. For example, a "group" defines
// both a type and a field in a single declaration. Thus, the locations
// corresponding to the type and field and their components will overlap.
// - Code which tries to interpret locations should probably be designed to
// ignore those that it doesn't understand, as more types of locations could
// be recorded in the future.
repeated Location location = 1;
message Location {
// Identifies which part of the FileDescriptorProto was defined at this
// location.
//
// Each element is a field number or an index. They form a path from
// the root FileDescriptorProto to the place where the definition. For
// example, this path:
// [ 4, 3, 2, 7, 1 ]
// refers to:
// file.message_type(3) // 4, 3
// .field(7) // 2, 7
// .name() // 1
// This is because FileDescriptorProto.message_type has field number 4:
// repeated DescriptorProto message_type = 4;
// and DescriptorProto.field has field number 2:
// repeated FieldDescriptorProto field = 2;
// and FieldDescriptorProto.name has field number 1:
// optional string name = 1;
//
// Thus, the above path gives the location of a field name. If we removed
// the last element:
// [ 4, 3, 2, 7 ]
// this path refers to the whole field declaration (from the beginning
// of the label to the terminating semicolon).
repeated int32 path = 1 [packed = true];
// Always has exactly three or four elements: start line, start column,
// end line (optional, otherwise assumed same as start line), end column.
// These are packed into a single field for efficiency. Note that line
// and column numbers are zero-based -- typically you will want to add
// 1 to each before displaying to a user.
repeated int32 span = 2 [packed = true];
// If this SourceCodeInfo represents a complete declaration, these are any
// comments appearing before and after the declaration which appear to be
// attached to the declaration.
//
// A series of line comments appearing on consecutive lines, with no other
// tokens appearing on those lines, will be treated as a single comment.
//
// leading_detached_comments will keep paragraphs of comments that appear
// before (but not connected to) the current element. Each paragraph,
// separated by empty lines, will be one comment element in the repeated
// field.
//
// Only the comment content is provided; comment markers (e.g. //) are
// stripped out. For block comments, leading whitespace and an asterisk
// will be stripped from the beginning of each line other than the first.
// Newlines are included in the output.
//
// Examples:
//
// optional int32 foo = 1; // Comment attached to foo.
// // Comment attached to bar.
// optional int32 bar = 2;
//
// optional string baz = 3;
// // Comment attached to baz.
// // Another line attached to baz.
//
// // Comment attached to qux.
// //
// // Another line attached to qux.
// optional double qux = 4;
//
// // Detached comment for corge. This is not leading or trailing comments
// // to qux or corge because there are blank lines separating it from
// // both.
//
// // Detached comment for corge paragraph 2.
//
// optional string corge = 5;
// /* Block comment attached
// * to corge. Leading asterisks
// * will be removed. */
// /* Block comment attached to
// * grault. */
// optional int32 grault = 6;
//
// // ignored detached comments.
optional string leading_comments = 3;
optional string trailing_comments = 4;
repeated string leading_detached_comments = 6;
}
}
// Describes the relationship between generated code and its original source
// file. A GeneratedCodeInfo message is associated with only one generated
// source file, but may contain references to different source .proto files.
message GeneratedCodeInfo {
// An Annotation connects some span of text in generated code to an element
// of its generating .proto file.
repeated Annotation annotation = 1;
message Annotation {
// Identifies the element in the original source .proto file. This field
// is formatted the same as SourceCodeInfo.Location.path.
repeated int32 path = 1 [packed = true];
// Identifies the filesystem path to the original source .proto.
optional string source_file = 2;
// Identifies the starting offset in bytes in the generated code
// that relates to the identified object.
optional int32 begin = 3;
// Identifies the ending offset in bytes in the generated code that
// relates to the identified offset. The end offset should be one past
// the last relevant byte (so the length of the text = end - begin).
optional int32 end = 4;
}
} }

2
tests/inputs/oneof/oneof.json
View File

@ -1,3 +1,3 @@
{ {
"count": 100 "pitied": 100
} }

12
tests/inputs/oneof/oneof.proto
View File

@ -2,7 +2,15 @@ syntax = "proto3";
message Test { message Test {
oneof foo { oneof foo {
int32 count = 1; int32 pitied = 1;
string name = 2; string pitier = 2;
}
int32 just_a_regular_field = 3;
oneof bar {
int32 drinks = 11;
string bar_name = 12;
} }
} }

2
tests/inputs/oneof/oneof_name.json
View File

@ -1,3 +1,3 @@
{ {
"name": "foobar" "pitier": "Mr. T"
} }