This change add parsing editions specific descriptor parts to
internal/filedesc which makes it possible to load file descriptors of
protos using editions at runtime.
Change-Id: I7a0c23303acddd2cff115859d4cf82bf0102b14c
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/563615
Reviewed-by: Michael Stapelberg <stapelberg@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
I'll add the runtime implementations and tests in a followup change to
make it easier to review.
Change-Id: I2917e0d40a99e81799d89a66584a680ed7e8dbf7
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/561939
Auto-Submit: Lasse Folger <lassefolger@google.com>
Reviewed-by: Michael Stapelberg <stapelberg@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
This change adds the first Go specific editions feature and the
associated proto. The feature is used to control if the legacy
UnmarshalJSON method should be generated for enums. This change only
affects protos using editions.
More tests will be added in a followup change.
Change-Id: Ifb62454d7568bd6d90d0b93f8953adcfe46c45fd
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/561938
Reviewed-by: Michael Stapelberg <stapelberg@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Lasse Folger <lassefolger@google.com>
+ The previous commit forgot to resolve the LegacyRequired feature
Change-Id: I05a2847be3d52277c65df004363ea3c87944364e
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/555976
Reviewed-by: Michael Stapelberg <stapelberg@google.com>
Auto-Submit: Lasse Folger <lassefolger@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Before this CL we only used features where they were applicable, e.g.
file features were only set on the file descriptor and field features
only on the fields descriptor. After this CL all descriptors (file,
message and field) contain always all features and when initializing
either of them we merge the features specified by options with the
features of the parent descriptor and use the result as the effective
feature. This way we initialize all the descriptors the same way and the
proto spec can change on which level features can be specified without
us having to change the code (at the moment most features can only be
specified on file or field level, in the future message might also be
possible).
Change-Id: I822202bf6a2bba5167bf9622c91a65c4523e78f9
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/555975
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Michael Stapelberg <stapelberg@google.com>
Before this change, calling dynamicpb.NewMessage() on a
protoreflect.MessageDescriptor that uses Editions would fail with:
message field "[…].relatedqueries" is not packable
(While Go Protobuf does not yet support Editions, C++ programs
can save descriptors to a file that is later loaded by a Go program.)
This was an oversight in commit e8baad6b6c9e2bb1c48e4963866248d4f35d4fd7.
Change-Id: I45d2b07a85ee40cb7c028098b81c4e76bf2ff555
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/554896
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Cassondra Foesch <cfoesch@gmail.com>
Reviewed-by: Christian Höppner <hoeppi@google.com>
Auto-Submit: Michael Stapelberg <stapelberg@google.com>
Before this change, calling dynamicpb.NewMessage() on a
protoreflect.MessageDescriptor that uses Editions would fail with:
message field "[…].relatedqueries" is an invalid group: proto: invalid under
proto2 semantics
(While Go Protobuf does not yet support Editions, C++ programs
can save descriptors to a file that is later loaded by a Go program.)
Given that groups are only supported in proto2, but not in proto3
(see https://protobuf.dev/editions/features/#message_encoding),
I think the conditional meant to say "proto3 semantics".
Change-Id: I00368e4433ca60bb883b726e34ee6b8862cfd1e7
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/554955
Reviewed-by: Cassondra Foesch <cfoesch@gmail.com>
Reviewed-by: Christian Höppner <hoeppi@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
This is the first of multiple changes to add protobuf editions support
to the Go implementation.
This change includes the feature resolutions for protobuf editions.
The plugin does not yet report editions support and protoc would fail
if it invoked this plugin with a proto that uses editions.
Change-Id: I7d31909366c3433b21abab74ec92263e08145434
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/547315
Auto-Submit: Lasse Folger <lassefolger@google.com>
Reviewed-by: Michael Stapelberg <stapelberg@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Commit-Queue: Lasse Folger <lassefolger@google.com>
Reviewed-by: Lasse Folger <lassefolger@google.com>
Fixesgolang/protobuf#1575
When converting a protoreflect.FileDescriptor
to a FileDescriptorProto, don't emit an invalid syntax.
This can happen for placeholder files, which represent
missing dependencies of a file.
Change-Id: I1d2ae1bd9f89c8a92e751871ce5c02727505b5e2
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/540455
Reviewed-by: Lasse Folger <lassefolger@google.com>
Reviewed-by: Michael Stapelberg <stapelberg@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Damien Neil <dneil@google.com>
Modernize the documentation across the entire module
to make use of the newer ability to linkify declarations.
Change-Id: I440f9a3f025ec6fadfd9cba59b1dfb57028bf3f1
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/309430
Reviewed-by: Michael Stapelberg <stapelberg@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
Apply go1.19 gofmt to non-generated files.
Generated .pb.go files are created by generate.bash using Go 1.18,
so leave them unchanged for now.
Change-Id: Ied36c83cf99704988d059bf0412e677f0fbc71b0
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/418676
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: Lasse Folger <lassefolger@google.com>
If a .proto file does not have a package statement,
then protoc will not populate the package field.
Emulate the same behavior for ToFileDescriptorProto.
Fixesgolang/protobuf#1195
Change-Id: I617102e50c96f0fa1bac6a6bc4de1cc485750bb9
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/259899
Reviewed-by: Damien Neil <dneil@google.com>
Trust: Joe Tsai <thebrokentoaster@gmail.com>
Added API:
SourceLocations.ByPath
SourceLocations.ByDescriptor
SourceLocation.Next
SourcePath.String
SourcePath.Equal
We modify compiler/protogen to use SourceLocations.ByDescriptor.
In retrospect, if this had existed during the development of protogen,
we would not have exposed protogen.Location and related fields.
Change-Id: I58f17e59f90b9ba16f0982c4b71c2542e4ff6e75
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/238000
Reviewed-by: Damien Neil <dneil@google.com>
Add a new TextName accessor that returns the field name that should
be used for the text format. It is usually just the field name, except:
1) it uses the inlined message name for groups,
2) uses the full name surrounded by brackets for extensions, and
3) strips the "message_set_extension" for well-formed extensions
to the proto1 MessageSet.
We make similar adjustments to the JSONName accessor so that it applies
similar semantics for extensions.
The two changes simplifies all logic that wants the humanly readable
name for a field.
Change-Id: I524b6e017fb955146db81819270fe197f8f97980
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/239838
Reviewed-by: Herbie Ong <herbie@google.com>
The genid package unifies the genname, fieldnum, and detectknown
packages into a single package.
Whenever possible use the generated constants rather than
hard-coded literals. This makes it easier to search the entire
module for special logic that deal with well-known types.
Change-Id: I13beff1f4149444a0c0b9e607ebf759657f000f4
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/235301
Reviewed-by: Herbie Ong <herbie@google.com>
In the upcoming 3.12.x release of protoc, the proto3 language will be
amended to support true presence for scalars. This CL adds support
to both the generator and runtime to support these semantics.
Newly added public API:
protogen.Plugin.SupportedFeatures
protoreflect.FieldDescriptor.HasPresence
protoreflect.FieldDescriptor.HasOptionalKeyword
protoreflect.OneofDescriptor.IsSynthetic
Change-Id: I7c86bf66d0ae56642109beb5f2132184593747ad
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/230698
Reviewed-by: Damien Neil <dneil@google.com>
Add a function that takes a protoreflect.FileDescriptorSet and returns
a protoregistry.Files.
Updates golang/protobuf##1065.
Change-Id: I2715d042053ef7d3f1bfcee1866c20cac423e327
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/226237
Reviewed-by: Joe Tsai <thebrokentoaster@gmail.com>
The FileOptions type provides the ability to specify specialized options
for how a file descriptor is constructed. It follows the same optional
arguments pattern as used in the proto package.
The resolver is not an option since it almost always necessary
when constructing a file descriptor.
Change-Id: Ib98ac6289881ad8402dd615f6c895da5899cb8d9
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/218940
Reviewed-by: Damien Neil <dneil@google.com>
TODOs that we do not intend to address have been deleted.
Those that are blocking v2 release are marked with "blocks".
Change-Id: I7efa9e546d0637b562101d0edc7009893d762722
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/218878
Reviewed-by: Damien Neil <dneil@google.com>
Add a sentinel proto.Error error which matches all errors returned by
packages in this module.
Document that protoregistry.NotFound is an exact sentinel value for
performance reasons.
Add a Wrap function to the internal/errors package and use it to wrap
errors from outside sources (resolvers). Wrapped errors match
proto.Error.
Fixesgolang/protobuf#1021.
Change-Id: I45567df3fd6c8dc9a5caafdb55654827f6fb1941
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/215338
Reviewed-by: Joe Tsai <joetsai@google.com>
We resisted adding Clone for a while since:
* It is a function that is perfectly suited for generics.
However, generics probably still won't be available in Go for some time
and it is impractical to block addition of this function when it is very
widely used and will be necessary for the v1 to v2 migration.
* In the past, there was no protoreflect.Message.IsValid, so there was
no proper API to detect invalid top-level messages and return them as such.
Since Clone relies on certain properties about proper round-tripping
of ProtoMessage.ProtoReflect <-> Message.Interface, we add a test
in testing/prototest to check for this.
Change-Id: Ic492b68f27b8b88322a6a3fa3a5e492228db79d9
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/213297
Reviewed-by: Damien Neil <dneil@google.com>
The JSONName constructor returns a struct value which shallow copies
a sync.Once within it; this is a dubious pattern.
Instead, add a jsonName.Init method to initialize the value.
Change-Id: I190a7239b1b62a8041ee7e4e09c0fe37b64ff623
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/213237
Reviewed-by: Damien Neil <dneil@google.com>
Avoid a deadlock when registering a legacy ExtensionType, caused by
initialization of the "internal/impl".ExtensionInfo calling IsMessageSet
on the MessageDescriptor of the type being extended.
We can avoid this deadlock either by initializing the ExtensionType
outside of the GlobalTypes mutex, or by moving IsMessageSet to L1
initialization of the MessageDescriptor so that it doesn't trigger lazy
init.
CL 204804 takes the former approach; this CL takes the latter.
Change-Id: Idfc1ed36a23a139839290ea32492142a17f68cf5
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/205957
Reviewed-by: Joe Tsai <thebrokentoaster@gmail.com>
Add type-safe methods to register message, enum, and extension types.
Deprecate the NewTypes function and the (*Types).Register method.
Add (*File).RegisterFile and deprecate the NewFiles function and
the (*File).Register method.
Updates golang/protobuf#963
Change-Id: Ie89e77526e0874539e9bd929ca0ba8d758e65a6e
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/199898
Reviewed-by: Joe Tsai <thebrokentoaster@gmail.com>
This CL:
* Make the meaning of impl/ExtensionInfo.goType consistent. Before,
it was sometimes a T and other times a []T depending on the current
state of initialization. Change it so that it is the constructor's
responsibility to pass in a []T if it is repeated.
* Make internal/filetype responsible for constructing a []T for
repeated extension fields.
* Makes filedesc/Extension.Cardinality one of the eagerly initialized
pieces of information since it is useful to internal/filetype.
* Unify ExtensionInfo.desc and ExtensionInfo.tdesc.ExtensionField,
which held the same information.
* Remove the internal implementation for impl.X.ExtensionDescFromType
since we are dropping support for this from v1.
Change-Id: Ie95c4de66cd674c1d886da4f63b133b7d763c7ef
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/195777
Reviewed-by: Damien Neil <dneil@google.com>
Use internal/detrand in the construction of our error messages.
This alters whether there is one or two spaces following the "proto:" prefix.
While it is easy for users to still work around this mutation,
sit at least forces them to write test infrastructure to more fuzzily
match on error strings.
Change-Id: I4ddca717526ee3fc4dbb1e0b36cfca8c6e0df36d
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/194038
Reviewed-by: Herbie Ong <herbie@google.com>
Rename build tag "proto1_legacy" -> "protolegacy"
to be consistent with the "protoreflect" tag.
Rename flag constant "Proto1Legacy" -> "ProtoLegacy" since
it covers more than simply proto1 legacy features.
For example, it covers alpha-features of proto3 that
were eventually removed from the final proto3 release.
Change-Id: I0f4fcbadd4b5a61c87645e2e5be11d187e59157c
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/189345
Reviewed-by: Damien Neil <dneil@google.com>
This implements generation of and reflection support for weak fields.
Weak fields are a proto1 feature where the "weak" option can be specified
on a singular message field. A weak reference results in generated code
that does not directly link in the dependency containing the weak message.
Weak field support is not added to any of the serialization logic.
Change-Id: I08ccfa72bc80b2ffb6af527a1677a0a81dcf33fb
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/185399
Reviewed-by: Damien Neil <dneil@google.com>
For proto3 messages with an enum field that could not be resolved,
do not check the syntax of that enum dependency.
Change-Id: I7c646539351edc35243ab950d335f4018cc4c0e9
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/186001
Reviewed-by: Damien Neil <dneil@google.com>
This adds minimal support for preserving the source context information.
Change-Id: I4b3cac9690b7469ecb4e5434251a809be4d7894c
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/183157
Reviewed-by: Damien Neil <dneil@google.com>
Usage of these is pervasive in code which works with proto2, and proto2
will be with us for a long, long time to come. Move them to the proto
package.
Change-Id: I1b2e57429fd5a8f107a848a4492d20c27f304bd7
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/185543
Reviewed-by: Joe Tsai <thebrokentoaster@gmail.com>
Create a new internal/strs package that unifies common functionality:
* Since protobuf itself pseudo-specifies at least 4 different camel-case
and snake-case conversion functions, we define all variants in one place.
* We move the internal/filedesc.nameBuilder function to this package.
We simplify its implementation to not depend on a strings.Builder fork
under the hood since the semantics we desire is simpler than what
strings.Builder provides.
* We use strs.Builder in reflect/protodesc in its construction of all
the full names. This is perfect use case of strs.Builder since all
full names within a file descriptor share the same lifetime.
* Add an UnsafeString and UnsafeBytes cast function that will be useful
in the near future for optimizing encoding/prototext and encoding/protojson.
Change-Id: I2cf07cbaf6f72e5f9fd6ae3d37b0d46f6af2ad59
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/185198
Reviewed-by: Damien Neil <dneil@google.com>
Hyrum's Law dictates that if we do not prevent naughty behavior,
people will rely on it. If we do not validate that the provided
file descriptor is correct today, it will be near impossible
to add proper validation checks later on.
The logic added validates that the provided file descriptor is
correct according to the same semantics as protoc,
which was reversed engineered to derive the set of rules implemented here.
The rules are unfortunately complicated because protobuf is a language
full of many non-orthogonal features. While our logic is complicated,
it is still 1/7th the size of the equivalent C++ code!
Change-Id: I6acc5dc3bd2e4c6bea6cd9e81214f8104402602a
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/184837
Reviewed-by: Damien Neil <dneil@google.com>
Overview of changes:
* Add an option that specifies whether to replace unresolvable references
with a placeholder instead of producing an error. Since the prior behavior
produced placeholders (not always), we default to that behavior for now,
but will enable strict resolving in a future CL.
* The option is not yet exported because there is concern about what the
public API should look like. This will be exposed in a future CL.
* Unlike before, we now permit placeholders for unresolvable enum values.
* We implement relative name resolution logic.
* We handle the case where the type is unknown, but type_name is specified.
In such a case, we populate both FieldDescriptor.{Enum,Message} and leave
the FieldDescriptor.Kind with the zero value. If the type_name happened
to resolve, we use that to determine the type.
* If a placeholder is used to represent a relative name,
the FullName reports an invalid full name with a "*." prefix.
Change-Id: Ifa8c750423c488fb9324eec4d033a2f251505fda
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/184317
Reviewed-by: Damien Neil <dneil@google.com>
There is little performance benefit to aliasing the input since we copy
every field except the options. Thus, just go all the way and copy the
options as well and document this as such.
Change-Id: If6ca5ce0ee03c9f76e528023b6056ad99d3ca209
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/184879
Reviewed-by: Damien Neil <dneil@google.com>
This fixes a bug introduced by CL/182360.
Overview of the problem:
* CL/182360 removes the internal/prototype package, such that
protodesc was re-implemented using internal/filedesc.
* As a result of that change, resolving internal dependencies became
the responsibility of protodesc.
* Dependency resolution used the following two-pass algorithm:
1) first pass derives the full name of all declarations
2) second pass fully initializes each descriptor declaration,
now being able to resolve local dependencies from the previous step.
* When the second pass looks up a local dependency, it is guaranteed to
find it, but it is not guaranteed that the dependency has been initialized
(since it may appear later on). This is problematic for default enum values
since it implies that the enum dependency may not be sufficiently
initialized to be able to query its set of values, leading to panics.
* CL/182360 recognized the problem and attempted to enforce an initialization
ordering where nested enums were always initialized before the body of the
message declaration itself.
* However, that ordering fails to enforce that that enum declarations outside
the parent tree are initialized beforehand. For example, referring to an
enum value that is declared within a sibling of the parent message.
* This CL fixes the problem with a three-pass algorithm:
1) first pass derives the full name *and* fully initialize the
entire descriptor *except* for dependency references (i.e., type_name).
2) second pass only resolves dependency references,
where we do not need to worry about initialization ordering.
3) third pass validates the descriptors are well-formed.
This can now depend on all information being fully initialized.
* While a lot of code moves, this change is actually very mechanical.
Other than split things apart, no new logic is introduced nor removed.
Change-Id: Ia91d4aade8f6187c19d704d43ae96b3b9d276792
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/184297
Reviewed-by: Damien Neil <dneil@google.com>
This is a breaking change.
The replacement is the Files.FindDescriptorByName method,
which is more flexible as it handles all descriptor types.
Change-Id: I2ccd544a7630396a2428b1d41f836c5246070912
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/183700
Reviewed-by: Damien Neil <dneil@google.com>
This change makes it such that Files now functionally registers all
descriptors in a file (not just enums, messages, extensions, and services),
but also including enum values, messages fields/oneofs, and service methods.
The ability to look up any descriptor by full name is needed to:
1) properly detect namespace conflicts on enum values
2) properly implement the relative name lookup logic in reflect/protodesc
The approach taken:
1) Assumes that a FileDescriptor has no internal name conflicts.
This will (in a future CL) be guaranteed by reflect/protodesc and
is guaranteed today by protoc for generated descriptors.
2) Observes that the only declarations that can possibly conflict
with another file are top-level declarations (i.e., enums, enum values,
messages, extensions, and services). Enum values are annoying
since they live in the same scope as the parent enum, rather than
being under the enum.
For the internal data structure of Files, we only register the top-level
declarations. This is the bare minimum needed to detect whether the file
being registered has any namespace conflicts with previously registered files.
We shift the effort to lookups, where we now need to peel off the end fragments
of a full name until we find a match in the internal registry. If a match
is found, we may need to descend into that declaration to find a nested
declaration by name.
For initialization, we modify internal/filedesc to initialize the
enum values for all top-level enums. This performance cost is offsetted
by the fact that Files.Register now avoids internally registering
nested enums, messages, and extensions.
For lookup, the cost has shifted from O(1) to O(N),
where N is the number of segments in the full name.
Top-level descriptors still have O(1) lookup times.
Nested descriptors have O(M) lookup times,
where M is the level of nesting within a single file.
Change-Id: I950163423431f04a503b6201ddcc20a62ccba017
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/183697
Reviewed-by: Damien Neil <dneil@google.com>
The internal/fileinit package is split apart into two packages:
* internal/filedesc constructs descriptors from the raw proto.
It is very similar to the previous internal/fileinit package.
* internal/filetype wraps descriptors with Go type information
Overview:
* The internal/fileinit package will be deleted in a future CL.
It is kept around since the v1 repo currently depends on it.
* The internal/prototype package is deleted. All former usages of it
are now using internal/filedesc instead. Most significantly,
the reflect/protodesc package was almost entirely re-written.
* The internal/impl package drops support for messages that do not
have a Descriptor method (pre-2016). This removes a significant amount
of technical debt.
filedesc.Builder to parse raw descriptors.
* The internal/encoding/defval package now handles enum values by name.
Change-Id: I3957bcc8588a70470fd6c7de1122216b80615ab7
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/182360
Reviewed-by: Damien Neil <dneil@google.com>
The change to make protodesc.NewFile take an interface rather than a
concrete type means that NewFile(f, nil) now causes a panic. (A nil
*protoregistry.Files is valid.)
Fix this panic by using a default, empty registry when NewFile's second
parameter is nil.
Change-Id: I70a1f0759e7ea5b57fba5b6123ee85188f4d560c
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/182979
Reviewed-by: Joe Tsai <thebrokentoaster@gmail.com>
CL/177044 switches the serialization APIs to take in a resolver interface.
This does the moral equivalent for protodesc.
This is technically a breaking change since the signature of NewFile changes.
However, it is unlikely that anything is affected by this.
Change-Id: I7b44d5c3d5570a17c052add4d229550e4a0ad163
Reviewed-on: https://go-review.googlesource.com/c/protobuf/+/182638
Reviewed-by: Damien Neil <dneil@google.com>
