Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Editions: Migrate string_field_validation to a C++ feature #13553

Closed
wants to merge 1 commit into from
Closed

Editions: Migrate string_field_validation to a C++ feature #13553

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 0 additions & 2 deletions src/google/protobuf/compiler/command_line_interface_unittest.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1560,7 +1560,6 @@ TEST_F(CommandLineInterfaceTest, Plugin_RuntimeFeatures) {
EqualsProto(R"pb(field_presence: IMPLICIT
enum_type: OPEN
repeated_field_encoding: PACKED
string_field_validation: MANDATORY
message_encoding: LENGTH_PREFIXED
json_format: ALLOW
raw_features { field_presence: IMPLICIT }
Expand All @@ -1577,7 +1576,6 @@ TEST_F(CommandLineInterfaceTest, Plugin_RuntimeFeatures) {
EqualsProto(R"pb(field_presence: IMPLICIT
enum_type: OPEN
repeated_field_encoding: PACKED
string_field_validation: MANDATORY
message_encoding: LENGTH_PREFIXED
json_format: ALLOW
raw_features { field_presence: IMPLICIT }
Expand Down
48 changes: 42 additions & 6 deletions src/google/protobuf/compiler/cpp/generator.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -101,6 +101,17 @@ absl::flat_hash_map<absl::string_view, std::string> CommonVars(
};
}

static bool IsStringMapType(const FieldDescriptor& field) {
if (!field.is_map()) return false;
for (int i = 0; i < field.message_type()->field_count(); ++i) {
if (field.message_type()->field(i)->type() ==
FieldDescriptor::TYPE_STRING) {
return true;
}
}
return false;
}

} // namespace

bool CppGenerator::Generate(const FileDescriptor* file,
Expand Down Expand Up @@ -356,19 +367,44 @@ absl::Status CppGenerator::ValidateFeatures(const FileDescriptor* file) const {
google::protobuf::internal::VisitDescriptors(*file, [&](const FieldDescriptor& field) {
const FeatureSet& source_features = GetSourceFeatures(field);
const FeatureSet& raw_features = GetSourceRawFeatures(field);
if (raw_features.GetExtension(::pb::cpp).has_legacy_closed_enum() &&
field.cpp_type() != FieldDescriptor::CPPTYPE_ENUM) {
status = absl::FailedPreconditionError(absl::StrCat(
"Field ", field.full_name(),
" specifies the legacy_closed_enum feature but has non-enum type."));
}
if (field.enum_type() != nullptr &&
source_features.GetExtension(::pb::cpp).legacy_closed_enum() &&
source_features.field_presence() == FeatureSet::IMPLICIT) {
status = absl::FailedPreconditionError(
absl::StrCat("Field ", field.full_name(),
" has a closed enum type with implicit presence."));
}
if (source_features.GetExtension(::pb::cpp).utf8_validation() ==
pb::CppFeatures::UTF8_VALIDATION_UNKNOWN) {
status = absl::FailedPreconditionError(absl::StrCat(
"Field ", field.full_name(),
" has an unknown value for the utf8_validation feature. ",
source_features.DebugString(), "\nRawFeatures: ", raw_features));
}

if (field.containing_type() == nullptr ||
!field.containing_type()->options().map_entry()) {
// Skip validation of explicit features on generated map fields. These
// will be blindly propagated from the original map field, and may violate
// a lot of these conditions. Note: we do still validate the
// user-specified map field.
if (raw_features.GetExtension(::pb::cpp).has_legacy_closed_enum() &&
field.cpp_type() != FieldDescriptor::CPPTYPE_ENUM) {
status = absl::FailedPreconditionError(
absl::StrCat("Field ", field.full_name(),
" specifies the legacy_closed_enum feature but has "
"non-enum type."));
}
if (field.type() != FieldDescriptor::TYPE_STRING &&
!IsStringMapType(field) &&
raw_features.GetExtension(::pb::cpp).has_utf8_validation()) {
status = absl::FailedPreconditionError(
absl::StrCat("Field ", field.full_name(),
" specifies the utf8_validation feature but is not of "
"string type."));
}
}

#ifdef PROTOBUF_FUTURE_REMOVE_WRONG_CTYPE
if (field.options().has_ctype()) {
if (field.cpp_type() != FieldDescriptor::CPPTYPE_STRING) {
Expand Down
89 changes: 85 additions & 4 deletions src/google/protobuf/compiler/cpp/generator_unittest.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -87,7 +87,6 @@ TEST_F(CppGeneratorTest, BasicError) {
"foo.proto:4:7: Expected \"required\", \"optional\", or \"repeated\"");
}


TEST_F(CppGeneratorTest, LegacyClosedEnumOnNonEnumField) {
CreateTempFile("foo.proto",
R"schema(
Expand Down Expand Up @@ -150,8 +149,7 @@ TEST_F(CppGeneratorTest, LegacyClosedEnumInherited) {
}

TEST_F(CppGeneratorTest, LegacyClosedEnumImplicit) {
CreateTempFile("foo.proto",
R"schema(
CreateTempFile("foo.proto", R"schema(
edition = "2023";
import "google/protobuf/cpp_features.proto";
option features.(pb.cpp).legacy_closed_enum = true;
Expand All @@ -162,7 +160,8 @@ TEST_F(CppGeneratorTest, LegacyClosedEnumImplicit) {
message Foo {
TestEnum bar = 1 [features.field_presence = IMPLICIT];
int32 baz = 2;
})schema");
}
)schema");

RunProtoc(
"protocol_compiler --proto_path=$tmpdir --cpp_out=$tmpdir "
Expand All @@ -172,6 +171,88 @@ TEST_F(CppGeneratorTest, LegacyClosedEnumImplicit) {
"Field Foo.bar has a closed enum type with implicit presence.");
}

TEST_F(CppGeneratorTest, Utf8ValidationMap) {
CreateTempFile("foo.proto", R"schema(
edition = "2023";
import "google/protobuf/cpp_features.proto";

message Foo {
map<string, string> map_field = 1 [
features.(pb.cpp).utf8_validation = NONE
];
map<int32, string> map_field_value = 2 [
features.(pb.cpp).utf8_validation = VERIFY_PARSE
];
map<string, int32> map_field_key = 3 [
features.(pb.cpp).utf8_validation = VERIFY_DLOG
];
}
)schema");

RunProtoc(
"protocol_compiler --proto_path=$tmpdir --cpp_out=$tmpdir "
"--experimental_editions foo.proto");

ExpectNoErrors();
}

TEST_F(CppGeneratorTest, Utf8ValidationNonString) {
CreateTempFile("foo.proto", R"schema(
edition = "2023";
import "google/protobuf/cpp_features.proto";

message Foo {
int64 bar = 1 [
features.(pb.cpp).utf8_validation = NONE
];
}
)schema");

RunProtoc(
"protocol_compiler --proto_path=$tmpdir --cpp_out=$tmpdir "
"--experimental_editions foo.proto");

ExpectErrorSubstring("Field Foo.bar specifies the utf8_validation feature");
}

TEST_F(CppGeneratorTest, Utf8ValidationNonStringMap) {
CreateTempFile("foo.proto", R"schema(
edition = "2023";
import "google/protobuf/cpp_features.proto";

message Foo {
map<int64, int64> bar = 1 [
features.(pb.cpp).utf8_validation = VERIFY_PARSE
];
}
)schema");

RunProtoc(
"protocol_compiler --proto_path=$tmpdir --cpp_out=$tmpdir "
"--experimental_editions foo.proto");

ExpectErrorSubstring("Field Foo.bar specifies the utf8_validation feature");
}

TEST_F(CppGeneratorTest, Utf8ValidationUnknownValue) {
CreateTempFile("foo.proto", R"schema(
edition = "2023";
import "google/protobuf/cpp_features.proto";

message Foo {
string bar = 1 [
features.(pb.cpp).utf8_validation = UTF8_VALIDATION_UNKNOWN
];
}
)schema");

RunProtoc(
"protocol_compiler --proto_path=$tmpdir --cpp_out=$tmpdir "
"--experimental_editions foo.proto");

ExpectErrorSubstring(
"Field Foo.bar has an unknown value for the utf8_validation feature.");
}
#ifdef PROTOBUF_FUTURE_REMOVE_WRONG_CTYPE
TEST_F(CppGeneratorTest, CtypeOnNoneStringFieldTest) {
CreateTempFile("foo.proto",
Expand Down
2 changes: 1 addition & 1 deletion src/google/protobuf/compiler/parser_unittest.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4243,7 +4243,7 @@ TEST_F(ParseEditionsTest, InvalidMerge) {
string foo = 1 [
default = "hello",
features.field_presence = FIELD_PRESENCE_UNKNOWN,
features.string_field_validation = STRING_FIELD_VALIDATION_UNKNOWN
features.enum_type = ENUM_TYPE_UNKNOWN
];
})schema",
"5:17: Feature field google.protobuf.FeatureSet.field_presence must resolve to a "
Expand Down
27 changes: 7 additions & 20 deletions src/google/protobuf/descriptor.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1117,10 +1117,11 @@ FeatureSet* CreateProto2DefaultFeatures() {
features->set_field_presence(FeatureSet::EXPLICIT);
features->set_enum_type(FeatureSet::CLOSED);
features->set_repeated_field_encoding(FeatureSet::EXPANDED);
features->set_string_field_validation(FeatureSet::HINT);
features->set_message_encoding(FeatureSet::LENGTH_PREFIXED);
features->set_json_format(FeatureSet::LEGACY_BEST_EFFORT);
features->MutableExtension(pb::cpp)->set_legacy_closed_enum(true);
features->MutableExtension(pb::cpp)->set_utf8_validation(
pb::CppFeatures::VERIFY_DLOG);

return features;
}
Expand All @@ -1145,9 +1146,11 @@ const FeatureSet& GetProto3Features() {
features->set_field_presence(FeatureSet::IMPLICIT);
features->set_enum_type(FeatureSet::OPEN);
features->set_repeated_field_encoding(FeatureSet::PACKED);
features->set_string_field_validation(FeatureSet::MANDATORY);
features->set_message_encoding(FeatureSet::LENGTH_PREFIXED);
features->set_json_format(FeatureSet::ALLOW);
features->MutableExtension(pb::cpp)->set_legacy_closed_enum(false);
features->MutableExtension(pb::cpp)->set_utf8_validation(
pb::CppFeatures::VERIFY_PARSE);
return features;
}();
return *kProto3Features;
Expand Down Expand Up @@ -3817,7 +3820,8 @@ bool FieldDescriptor::is_packed() const {

static bool IsStrictUtf8(const FieldDescriptor* field) {
return internal::InternalFeatureHelper::GetFeatures(*field)
.string_field_validation() == FeatureSet::MANDATORY;
.GetExtension(pb::cpp)
.utf8_validation() == pb::CppFeatures::VERIFY_PARSE;
}

bool FieldDescriptor::requires_utf8_validation() const {
Expand Down Expand Up @@ -7838,17 +7842,6 @@ void DescriptorBuilder::ValidateOptions(const FieldDescriptor* field,

}

static bool IsStringMapType(const FieldDescriptor* field) {
if (!field->is_map()) return false;
for (int i = 0; i < field->message_type()->field_count(); ++i) {
if (field->message_type()->field(i)->type() ==
FieldDescriptor::TYPE_STRING) {
return true;
}
}
return false;
}

void DescriptorBuilder::ValidateFieldFeatures(
const FieldDescriptor* field, const FieldDescriptorProto& proto) {
// Rely on our legacy validation for proto2/proto3 files.
Expand Down Expand Up @@ -7895,12 +7888,6 @@ void DescriptorBuilder::ValidateFieldFeatures(
field->full_name(), proto, DescriptorPool::ErrorCollector::NAME,
"Only singular scalar fields can specify required field presence.");
}
if (field->type() != FieldDescriptor::TYPE_STRING &&
!IsStringMapType(field) &&
field->proto_features_->has_string_field_validation()) {
AddError(field->full_name(), proto, DescriptorPool::ErrorCollector::NAME,
"Only string fields can specify `string_field_validation`.");
}
if (!field->is_repeated() &&
field->proto_features_->has_repeated_field_encoding()) {
AddError(field->full_name(), proto, DescriptorPool::ErrorCollector::NAME,
Expand Down
Loading
Loading