thrift 0.23.0 → 0.25.0

data/spec/uuid_validation_spec.rb

@@ -1,238 +0,0 @@
-#
-# Licensed to the Apache Software Foundation (ASF) under one
-# or more contributor license agreements. See the NOTICE file
-# distributed with this work for additional information
-# regarding copyright ownership. The ASF licenses this file
-# to you under the Apache License, Version 2.0 (the
-# "License"); you may not use this file except in compliance
-# with the License. You may obtain a copy of the License at
-#
-#   http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing,
-# software distributed under the License is distributed on an
-# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
-# KIND, either express or implied. See the License for the
-# specific language governing permissions and limitations
-# under the License.
-#
-
-require 'spec_helper'
-
-describe 'UUID Validation' do
-  protocols = [
-    ['BinaryProtocol', Thrift::BinaryProtocol],
-  ]
-
-  if defined?(Thrift::BinaryProtocolAccelerated)
-    protocols << ['BinaryProtocolAccelerated', Thrift::BinaryProtocolAccelerated]
-  end
-
-  protocols << ['CompactProtocol', Thrift::CompactProtocol]
-  protocols << ['JsonProtocol', Thrift::JsonProtocol]
-
-  protocols.each do |protocol_name, protocol_class|
-    describe protocol_name do
-      before(:each) do
-        @trans = Thrift::MemoryBufferTransport.new
-        @prot = protocol_class.new(@trans)
-      end
-
-      context 'valid UUIDs' do
-        it 'should accept lowercase UUIDs' do
-          uuid = '550e8400-e29b-41d4-a716-446655440000'
-          expect { @prot.write_uuid(uuid) }.not_to raise_error
-          result = @prot.read_uuid
-          expect(result).to eq(uuid)
-        end
-
-        it 'should accept uppercase UUIDs' do
-          uuid = '550E8400-E29B-41D4-A716-446655440000'
-          expect { @prot.write_uuid(uuid) }.not_to raise_error
-          result = @prot.read_uuid
-          # Result should be lowercase
-          expect(result).to eq('550e8400-e29b-41d4-a716-446655440000')
-        end
-
-        it 'should accept mixed case UUIDs' do
-          uuid = '550e8400-E29B-41d4-A716-446655440000'
-          expect { @prot.write_uuid(uuid) }.not_to raise_error
-          result = @prot.read_uuid
-          expect(result).to eq('550e8400-e29b-41d4-a716-446655440000')
-        end
-
-        it 'should accept all zeros' do
-          uuid = '00000000-0000-0000-0000-000000000000'
-          expect { @prot.write_uuid(uuid) }.not_to raise_error
-          result = @prot.read_uuid
-          expect(result).to eq(uuid)
-        end
-
-        it 'should accept all fs' do
-          uuid = 'ffffffff-ffff-ffff-ffff-ffffffffffff'
-          expect { @prot.write_uuid(uuid) }.not_to raise_error
-          result = @prot.read_uuid
-          expect(result).to eq(uuid)
-        end
-      end
-
-      context 'invalid UUIDs' do
-        def expect_invalid_uuid(value, message)
-          expect { @prot.write_uuid(value) }.to raise_error(Thrift::ProtocolException) do |error|
-            expect(error.type).to eq(Thrift::ProtocolException::INVALID_DATA)
-            expect(error.message).to eq(message)
-          end
-        end
-
-        it 'should reject nil' do
-          expect_invalid_uuid(nil, 'UUID must be a string')
-        end
-
-        it 'should reject non-string' do
-          expect_invalid_uuid(12345, 'UUID must be a string')
-        end
-
-        it 'should reject wrong length' do
-          expect_invalid_uuid('550e8400-e29b-41d4-a716', 'Invalid UUID format')
-        end
-
-        it 'should reject missing hyphens' do
-          expect_invalid_uuid('550e8400e29b41d4a716446655440000', 'Invalid UUID format')
-        end
-
-        it 'should reject hyphens in wrong positions' do
-          expect_invalid_uuid('550e840-0e29b-41d4-a716-446655440000', 'Invalid UUID format')
-        end
-
-        it 'should reject invalid hex characters (g)' do
-          expect_invalid_uuid('550e8400-e29b-41d4-a716-44665544000g', 'Invalid UUID format')
-        end
-
-        it 'should reject invalid hex characters (z)' do
-          expect_invalid_uuid('z50e8400-e29b-41d4-a716-446655440000', 'Invalid UUID format')
-        end
-
-        it 'should reject invalid hex characters (space)' do
-          expect_invalid_uuid('550e8400-e29b-41d4-a716-44665544000 ', 'Invalid UUID format')
-        end
-
-        it 'should reject empty string' do
-          expect_invalid_uuid('', 'Invalid UUID format')
-        end
-
-        it 'should reject UUID with extra characters' do
-          expect_invalid_uuid('550e8400-e29b-41d4-a716-446655440000x', 'Invalid UUID format')
-        end
-
-        it 'should reject trailing hyphen' do
-          expect_invalid_uuid('550e8400-e29b-41d4-a716-44665544000-', 'Invalid UUID format')
-        end
-
-        it 'should reject hyphen inside hex pair' do
-          expect_invalid_uuid('550e8400-e29b-41d4-a716-4466-5544000', 'Invalid UUID format')
-        end
-      end
-
-      context 'malformed binary data on read' do
-        it 'should raise error on truncated data' do
-          @trans = Thrift::MemoryBufferTransport.new
-          @prot = protocol_class.new(@trans)
-
-          # Write only 10 bytes instead of 16
-          if protocol_class == Thrift::JsonProtocol
-            @trans.write('"00000000-0000-0000-0000"')
-          else
-            @trans.write("\x00" * 10)
-          end
-
-          expect { @prot.read_uuid }.to raise_error(EOFError)
-        end
-
-        it 'should raise error on 15 bytes (one byte short)' do
-          @trans = Thrift::MemoryBufferTransport.new
-          @prot = protocol_class.new(@trans)
-
-          if protocol_class == Thrift::JsonProtocol
-            @trans.write('"00000000-0000-0000-0000-000000000"')
-          else
-            @trans.write("\x00" * 15)
-          end
-
-          expect { @prot.read_uuid }.to raise_error(EOFError)
-        end
-
-        it 'should raise error on empty buffer' do
-          @trans = Thrift::MemoryBufferTransport.new
-          @prot = protocol_class.new(@trans)
-
-          expect { @prot.read_uuid }.to raise_error(EOFError)
-        end
-      end
-
-      context 'multiple UUIDs in sequence' do
-        it 'should handle 10 UUIDs in sequence' do
-          uuids = 10.times.map { |i| sprintf('%08x-0000-0000-0000-000000000000', i) }
-
-          @trans = Thrift::MemoryBufferTransport.new
-          @prot = protocol_class.new(@trans)
-
-          uuids.each { |uuid| @prot.write_uuid(uuid) }
-
-          results = 10.times.map { @prot.read_uuid }
-          expect(results).to eq(uuids)
-        end
-
-        it 'should handle UUIDs interleaved with other types' do
-          @trans = Thrift::MemoryBufferTransport.new
-          @prot = protocol_class.new(@trans)
-
-          @prot.write_message_begin('testMessage', Thrift::MessageTypes::CALL, 0)
-          @prot.write_i32(42)
-          @prot.write_uuid('550e8400-e29b-41d4-a716-446655440000')
-          @prot.write_string('test')
-          @prot.write_uuid('6ba7b810-9dad-11d1-80b4-00c04fd430c8')
-          @prot.write_i64(123456789)
-          @prot.write_message_end
-
-          @prot.read_message_begin
-          expect(@prot.read_i32).to eq(42)
-          expect(@prot.read_uuid).to eq('550e8400-e29b-41d4-a716-446655440000')
-          expect(@prot.read_string).to eq('test')
-          expect(@prot.read_uuid).to eq('6ba7b810-9dad-11d1-80b4-00c04fd430c8')
-          expect(@prot.read_i64).to eq(123456789)
-          @prot.read_message_end
-        end
-
-        it 'should handle UUIDs in struct fields context' do
-          @trans = Thrift::MemoryBufferTransport.new
-          @prot = protocol_class.new(@trans)
-
-          # Simulate struct field headers
-          @prot.write_struct_begin('test')
-          @prot.write_field_begin('uuid1', Thrift::Types::UUID, 1)
-          @prot.write_uuid('550e8400-e29b-41d4-a716-446655440000')
-          @prot.write_field_end
-          @prot.write_field_begin('uuid2', Thrift::Types::UUID, 2)
-          @prot.write_uuid('6ba7b810-9dad-11d1-80b4-00c04fd430c8')
-          @prot.write_field_end
-          @prot.write_field_stop
-          @prot.write_struct_end
-
-          @prot.read_struct_begin
-          name, type, id = @prot.read_field_begin
-          expect(type).to eq(Thrift::Types::UUID)
-          expect(@prot.read_uuid).to eq('550e8400-e29b-41d4-a716-446655440000')
-          @prot.read_field_end
-
-          name, type, id = @prot.read_field_begin
-          expect(type).to eq(Thrift::Types::UUID)
-          expect(@prot.read_uuid).to eq('6ba7b810-9dad-11d1-80b4-00c04fd430c8')
-          @prot.read_field_end
-
-          name, type, id = @prot.read_field_begin
-          expect(type).to eq(Thrift::Types::STOP)
-        end
-      end
-    end
-  end
-end

data/test/fuzz/Makefile.am

@@ -1,173 +0,0 @@
-#
-# Licensed to the Apache Software Foundation (ASF) under one
-# or more contributor license agreements. See the NOTICE file
-# distributed with this work for additional information
-# regarding copyright ownership. The ASF licenses this file
-# to you under the Apache License, Version 2.0 (the
-# "License"); you may not use this file except in compliance
-# with the License. You may obtain a copy of the License at
-#
-#   http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing,
-# software distributed under the License is distributed on an
-# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
-# KIND, either express or implied. See the License for the
-# specific language governing permissions and limitations
-# under the License.
-#
-
-THRIFT = $(top_builddir)/compiler/cpp/thrift
-FUZZ_RB_ROOT = $(top_srcdir)/lib/rb
-FUZZ_GEN_DIR = $(FUZZ_RB_ROOT)/test/fuzz
-FUZZ_STUBS = \
-	$(FUZZ_GEN_DIR)/gen-rb/fuzz_test_constants.rb \
-	$(FUZZ_GEN_DIR)/gen-rb/fuzz_test_types.rb
-
-FUZZ_RUBY_FILES = \
-	$(srcdir)/fuzz_tracer.rb \
-	$(srcdir)/fuzz_common.rb \
-	$(srcdir)/fuzz_parse_binary_protocol.rb \
-	$(srcdir)/fuzz_parse_binary_protocol_harness.rb \
-	$(srcdir)/fuzz_parse_binary_protocol_accelerated.rb \
-	$(srcdir)/fuzz_parse_binary_protocol_accelerated_harness.rb \
-	$(srcdir)/fuzz_parse_compact_protocol.rb \
-	$(srcdir)/fuzz_parse_compact_protocol_harness.rb \
-	$(srcdir)/fuzz_parse_json_protocol.rb \
-	$(srcdir)/fuzz_parse_json_protocol_harness.rb \
-	$(srcdir)/fuzz_roundtrip_binary_protocol.rb \
-	$(srcdir)/fuzz_roundtrip_binary_protocol_harness.rb \
-	$(srcdir)/fuzz_roundtrip_binary_protocol_accelerated.rb \
-	$(srcdir)/fuzz_roundtrip_binary_protocol_accelerated_harness.rb \
-	$(srcdir)/fuzz_roundtrip_compact_protocol.rb \
-	$(srcdir)/fuzz_roundtrip_compact_protocol_harness.rb \
-	$(srcdir)/fuzz_roundtrip_json_protocol.rb \
-	$(srcdir)/fuzz_roundtrip_json_protocol_harness.rb
-
-FUZZ_ASAN_OPTIONS ?= allocator_may_return_null=1:detect_leaks=0:use_sigaltstack=0
-FUZZ_CC ?= clang
-FUZZ_CXX ?= clang++
-FUZZ_LDSHARED ?= $(FUZZ_CC) -shared
-FUZZ_LDSHAREDXX ?= $(FUZZ_CXX) -shared
-FUZZ_CFLAGS ?= -fsanitize=address,fuzzer-no-link -fno-omit-frame-pointer -fno-common -fPIC -g
-FUZZ_CXXFLAGS ?= -fsanitize=address,fuzzer-no-link -fno-omit-frame-pointer -fno-common -fPIC -g
-FUZZ_MAKE ?= make
-FUZZ_MAKEFLAGS ?= --environment-overrides V=1
-
-check: fuzz-prepare
-
-fuzz-prepare: $(FUZZ_STUBS)
-	@for script in $(FUZZ_RUBY_FILES); do \
-		$(RUBY) -c $$script; \
-	done
-	cd $(FUZZ_RB_ROOT) && \
-		RUBYLIB="lib:test/fuzz/gen-rb" $(RUBY) -e 'require "thrift"; require "fuzz_test_types"; abort("Fuzz::FuzzTest is missing") unless defined?(Fuzz::FuzzTest)'
-
-$(THRIFT):
-	$(MAKE) -C $(top_builddir)/compiler/cpp all
-
-$(FUZZ_STUBS): $(top_srcdir)/test/FuzzTest.thrift $(THRIFT)
-	$(MKDIR_P) $(FUZZ_GEN_DIR)
-	$(THRIFT) --gen rb -o $(FUZZ_GEN_DIR) $(top_srcdir)/test/FuzzTest.thrift
-
-fuzz-build-ext:
-	cd $(FUZZ_RB_ROOT)/ext && \
-		env MAKE="$(FUZZ_MAKE) $(FUZZ_MAKEFLAGS)" \
-			CC="$(FUZZ_CC)" \
-			CXX="$(FUZZ_CXX)" \
-			LDSHARED="$(FUZZ_LDSHARED)" \
-			LDSHAREDXX="$(FUZZ_LDSHAREDXX)" \
-			CFLAGS="$(FUZZ_CFLAGS)" \
-			CXXFLAGS="$(FUZZ_CXXFLAGS)" \
-			$(RUBY) extconf.rb && \
-		env MAKE="$(FUZZ_MAKE) $(FUZZ_MAKEFLAGS)" \
-			CC="$(FUZZ_CC)" \
-			CXX="$(FUZZ_CXX)" \
-			LDSHARED="$(FUZZ_LDSHARED)" \
-			LDSHAREDXX="$(FUZZ_LDSHAREDXX)" \
-			CFLAGS="$(FUZZ_CFLAGS)" \
-			CXXFLAGS="$(FUZZ_CXXFLAGS)" \
-			$(FUZZ_MAKE) $(FUZZ_MAKEFLAGS) clean all
-
-fuzz-run: fuzz-prepare
-	@test -n "$(TARGET)" || { echo 'Set TARGET=<fuzz script name>'; exit 1; }
-	cd $(FUZZ_RB_ROOT) && \
-		ASAN_OPTIONS="$(FUZZ_ASAN_OPTIONS)" \
-		LD_PRELOAD=$$($(RUBY) -e 'require "ruzzy"; print Ruzzy::ASAN_PATH') \
-		$(RUBY) test/fuzz/$(TARGET) $(CORPUS) $(FUZZ_ARGS)
-
-fuzz-parse-binary:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-run TARGET=fuzz_parse_binary_protocol.rb CORPUS="$(CORPUS)" FUZZ_ARGS="$(FUZZ_ARGS)"
-
-fuzz-parse-binary-accelerated: fuzz-build-ext
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-run TARGET=fuzz_parse_binary_protocol_accelerated.rb CORPUS="$(CORPUS)" FUZZ_ARGS="$(FUZZ_ARGS)"
-
-fuzz-parse-compact:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-run TARGET=fuzz_parse_compact_protocol.rb CORPUS="$(CORPUS)" FUZZ_ARGS="$(FUZZ_ARGS)"
-
-fuzz-parse-json:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-run TARGET=fuzz_parse_json_protocol.rb CORPUS="$(CORPUS)" FUZZ_ARGS="$(FUZZ_ARGS)"
-
-fuzz-roundtrip-binary:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-run TARGET=fuzz_roundtrip_binary_protocol.rb CORPUS="$(CORPUS)" FUZZ_ARGS="$(FUZZ_ARGS)"
-
-fuzz-roundtrip-binary-accelerated: fuzz-build-ext
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-run TARGET=fuzz_roundtrip_binary_protocol_accelerated.rb CORPUS="$(CORPUS)" FUZZ_ARGS="$(FUZZ_ARGS)"
-
-fuzz-roundtrip-compact:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-run TARGET=fuzz_roundtrip_compact_protocol.rb CORPUS="$(CORPUS)" FUZZ_ARGS="$(FUZZ_ARGS)"
-
-fuzz-roundtrip-json:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-run TARGET=fuzz_roundtrip_json_protocol.rb CORPUS="$(CORPUS)" FUZZ_ARGS="$(FUZZ_ARGS)"
-
-fuzz-smoke-parse-binary:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-parse-binary CORPUS="$(CORPUS)" FUZZ_ARGS="-runs=100 $(FUZZ_ARGS)"
-
-fuzz-smoke-parse-binary-accelerated:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-parse-binary-accelerated CORPUS="$(CORPUS)" FUZZ_ARGS="-runs=100 $(FUZZ_ARGS)"
-
-fuzz-smoke-parse-compact:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-parse-compact CORPUS="$(CORPUS)" FUZZ_ARGS="-runs=100 $(FUZZ_ARGS)"
-
-fuzz-smoke-parse-json:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-parse-json CORPUS="$(CORPUS)" FUZZ_ARGS="-runs=100 $(FUZZ_ARGS)"
-
-fuzz-smoke-roundtrip-binary:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-roundtrip-binary CORPUS="$(CORPUS)" FUZZ_ARGS="-runs=100 $(FUZZ_ARGS)"
-
-fuzz-smoke-roundtrip-binary-accelerated:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-roundtrip-binary-accelerated CORPUS="$(CORPUS)" FUZZ_ARGS="-runs=100 $(FUZZ_ARGS)"
-
-fuzz-smoke-roundtrip-compact:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-roundtrip-compact CORPUS="$(CORPUS)" FUZZ_ARGS="-runs=100 $(FUZZ_ARGS)"
-
-fuzz-smoke-roundtrip-json:
-	$(MAKE) $(AM_MAKEFLAGS) fuzz-roundtrip-json CORPUS="$(CORPUS)" FUZZ_ARGS="-runs=100 $(FUZZ_ARGS)"
-
-clean-local:
-	-$(RM) -r $(FUZZ_GEN_DIR)/gen-rb
-
-distdir:
-	$(MAKE) $(AM_MAKEFLAGS) distdir-am
-
-EXTRA_DIST = \
-	.gitignore \
-	README.md \
-	fuzz_common.rb \
-	fuzz_tracer.rb \
-	fuzz_parse_binary_protocol.rb \
-	fuzz_parse_binary_protocol_harness.rb \
-	fuzz_parse_binary_protocol_accelerated.rb \
-	fuzz_parse_binary_protocol_accelerated_harness.rb \
-	fuzz_parse_compact_protocol.rb \
-	fuzz_parse_compact_protocol_harness.rb \
-	fuzz_parse_json_protocol.rb \
-	fuzz_parse_json_protocol_harness.rb \
-	fuzz_roundtrip_binary_protocol.rb \
-	fuzz_roundtrip_binary_protocol_harness.rb \
-	fuzz_roundtrip_binary_protocol_accelerated.rb \
-	fuzz_roundtrip_binary_protocol_accelerated_harness.rb \
-	fuzz_roundtrip_compact_protocol.rb \
-	fuzz_roundtrip_compact_protocol_harness.rb \
-	fuzz_roundtrip_json_protocol.rb \
-	fuzz_roundtrip_json_protocol_harness.rb

data/test/fuzz/README.md

@@ -1,149 +0,0 @@
-# Ruby Fuzzing README
-
-The Ruby Thrift implementation uses [Ruzzy](https://github.com/trailofbits/ruzzy) for fuzzing. Ruzzy is a coverage-guided fuzzer for pure Ruby code and Ruby C extensions.
-
-We currently have several fuzz targets that test different aspects of the Thrift implementation:
-
-- `fuzz_parse_binary_protocol.rb` -- fuzzes deserialization of the Binary protocol
-- `fuzz_parse_binary_protocol_accelerated.rb` -- fuzzes deserialization of the accelerated Binary protocol
-- `fuzz_parse_compact_protocol.rb` -- fuzzes deserialization of the Compact protocol
-- `fuzz_parse_json_protocol.rb` -- fuzzes JSON protocol messages
-- `fuzz_roundtrip_binary_protocol.rb` -- fuzzes Binary roundtrips (deserialize, serialize, deserialize, compare)
-- `fuzz_roundtrip_binary_protocol_accelerated.rb` -- fuzzes accelerated Binary roundtrips
-- `fuzz_roundtrip_compact_protocol.rb` -- fuzzes Compact roundtrips
-- `fuzz_roundtrip_json_protocol.rb` -- fuzzes JSON message roundtrips
-
-The runnable files in this directory are tracer entrypoints. Ruzzy requires that pure Ruby fuzzing starts from a tracer script which then loads a separate harness, so do not invoke the matching `_harness.rb` files directly.
-
-The fuzzers use Ruzzy's mutation engine to generate test cases. Each target uses common testing code from `fuzz_common.rb`.
-
-For more information about Ruzzy and its options, see the [Ruzzy documentation](https://github.com/trailofbits/ruzzy).
-
-You can also use the corpus generator from the Rust implementation to generate initial Binary and Compact corpora that can be reused by the Ruby fuzzers, since those wire formats are identical between implementations.
-
-## Setup
-
-Run these commands from the repository root:
-
-```bash
-apt install -y clang-19
-
-# from https://github.com/trailofbits/ruzzy?tab=readme-ov-file#installing
-MAKE="make --environment-overrides V=1" \
-     CC="/usr/bin/clang-19" \
-     CXX="/usr/bin/clang++-19" \
-     LDSHARED="/usr/bin/clang-19 -shared" \
-     LDSHAREDXX="/usr/bin/clang++-19 -shared" \
-     gem install ruzzy
-
-# Validate the fuzz directory.
-# This generates test/fuzz/gen-rb, syntax-checks the fuzz scripts,
-# and verifies that Fuzz::FuzzTest loads correctly.
-make -C lib/rb/test/fuzz check
-```
-
-`make -C lib/rb check` now recurses into `lib/rb/test/fuzz`, so the same validation also runs as part of the normal Ruby `make check` flow.
-
-## Running Fuzzers
-
-The Makefile in this directory hides the `LD_PRELOAD` and `ASAN_OPTIONS` setup needed by Ruzzy.
-
-Pure Ruby targets only need `fuzz-prepare`, which is the same work as `check`:
-
-```bash
-make -C lib/rb/test/fuzz fuzz-parse-binary
-make -C lib/rb/test/fuzz fuzz-roundtrip-compact
-make -C lib/rb/test/fuzz fuzz-parse-json
-```
-
-Accelerated targets rebuild `thrift_native` with sanitizer flags first:
-
-```bash
-make -C lib/rb/test/fuzz fuzz-parse-binary-accelerated
-make -C lib/rb/test/fuzz fuzz-roundtrip-binary-accelerated
-```
-
-Use `CORPUS=...` to point at an input corpus. The path is resolved from `lib/rb`, which matches the old manual commands:
-
-```bash
-make -C lib/rb/test/fuzz fuzz-parse-binary \
-    CORPUS=../rs/test/fuzz/corpus/binary
-
-make -C lib/rb/test/fuzz fuzz-parse-compact \
-    CORPUS=../rs/test/fuzz/corpus/compact
-```
-
-Use `FUZZ_ARGS=...` for extra libFuzzer-style arguments. For a short local run, the Makefile also provides bounded smoke targets:
-
-```bash
-make -C lib/rb/test/fuzz fuzz-smoke-parse-binary
-make -C lib/rb/test/fuzz fuzz-smoke-parse-binary-accelerated \
-    FUZZ_CC=/usr/bin/clang-19 \
-    FUZZ_CXX=/usr/bin/clang++-19
-```
-
-If the default `clang` and `clang++` names are not correct for your system, override these variables for accelerated targets:
-
-```bash
-make -C lib/rb/test/fuzz fuzz-build-ext \
-    FUZZ_CC=/usr/bin/clang-19 \
-    FUZZ_CXX=/usr/bin/clang++-19
-```
-
-The underlying manual command from `lib/rb` is still:
-
-```bash
-# Memory allocation failures are common and low impact (DoS), so skip them for now.
-# Like Python, the Ruby interpreter leaks data, so ignore these for now.
-# Ruby recommends disabling sigaltstack.
-export ASAN_OPTIONS="allocator_may_return_null=1:detect_leaks=0:use_sigaltstack=0"
-
-LD_PRELOAD=$(ruby -e 'require "ruzzy"; print Ruzzy::ASAN_PATH') \
-    ruby test/fuzz/fuzz_parse_binary_protocol.rb
-```
-
-## Rust Corpus Generator
-
-We can use the Rust corpus generator to prepare corpora for the Binary and Compact protocol targets. Run it from the repository root:
-
-```bash
-cargo run --manifest-path lib/rs/test/fuzz/Cargo.toml --bin corpus_generator -- \
-    --output-dir <output_dir> \
-    --protocol <binary|compact> \
-    --generate <num_files> \
-    --buffer-size <buffer_size> \
-    --random-size <random_size>
-```
-
-Reasonable values (determined empirically):
-
-```bash
-cargo run --manifest-path lib/rs/test/fuzz/Cargo.toml --bin corpus_generator -- \
-    --output-dir ./lib/rs/test/fuzz/corpus/binary \
-    --protocol binary \
-    --generate 1000 \
-    --buffer-size 65536 \
-    --random-size 16384
-
-cargo run --manifest-path lib/rs/test/fuzz/Cargo.toml --bin corpus_generator -- \
-    --output-dir ./lib/rs/test/fuzz/corpus/compact \
-    --protocol compact \
-    --generate 1000 \
-    --buffer-size 16384 \
-    --random-size 16384
-```
-
-Then run a matching Ruby target:
-
-```bash
-make -C lib/rb/test/fuzz fuzz-parse-binary \
-    CORPUS=../rs/test/fuzz/corpus/binary
-```
-
-The Rust corpus generator does not emit JSON protocol inputs, so use it only with the Binary and Compact Ruby targets.
-
-## Troubleshooting
-
-If libFuzzer prints `WARNING: no interesting inputs were found so far. Is the code instrumented for coverage?` and quickly shrinks the seed corpus to `corp: 1/1b`, coverage tracing is not active. That usually means the tracer script was bypassed or `ruzzy` was not installed with the expected sanitizer-enabled toolchain. Run the `.rb` files in this directory, not the `_harness.rb` files.
-
-If an accelerated target starts using the pure Ruby implementation, rebuild the extension with `make -C lib/rb/test/fuzz fuzz-build-ext ...` and then rerun the accelerated target.

data/test/fuzz/fuzz_common.rb

@@ -1,95 +0,0 @@
-#
-# Licensed to the Apache Software Foundation (ASF) under one
-# or more contributor license agreements. See the NOTICE file
-# distributed with this work for additional information
-# regarding copyright ownership. The ASF licenses this file
-# to you under the Apache License, Version 2.0 (the
-# "License"); you may not use this file except in compliance
-# with the License. You may obtain a copy of the License at
-#
-#   http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing,
-# software distributed under the License is distributed on an
-# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
-# KIND, either express or implied. See the License for the
-# specific language governing permissions and limitations
-# under the License.
-#
-
-$:.unshift File.expand_path('../../lib', __dir__)
-$:.unshift File.expand_path('../../ext', __dir__)
-require 'thrift'
-$:.unshift File.dirname(__FILE__) + "/gen-rb"
-require 'fuzz_test_constants'
-
-require 'coverage'
-Coverage.start(branches: true) unless Coverage.respond_to?(:running?) && Coverage.running?
-require 'ruzzy'
-# Ruzzy.enable_branch_coverage_hooks
-
-def ignorable_fuzz_exception?(error)
-  return true if error.is_a?(Thrift::ProtocolException) ||
-    error.is_a?(EOFError) ||
-    error.is_a?(Encoding::UndefinedConversionError)
-
-  [
-    /don't know what (?:c)?type/,
-    /Too many fields for union/,
-    /too big to convert to '(?:int|long)'/,
-    /bignum too big to convert into 'long'/,
-    /negative array size/,
-    /Union fields are not set/
-  ].any? { |pattern| error.message =~ pattern }
-end
-
-def read_fuzz_test(protocol, read_message_begin)
-  protocol.read_message_begin if read_message_begin
-  obj = Fuzz::FuzzTest.new
-  obj.read(protocol)
-  protocol.read_message_end if read_message_begin
-  obj
-end
-
-def write_fuzz_test(protocol, obj, write_message_begin)
-  if write_message_begin
-    protocol.write_message_begin('fuzz', Thrift::MessageTypes::CALL, 0)
-  end
-  obj.write(protocol)
-  protocol.write_message_end if write_message_begin
-end
-
-def create_parser_fuzzer(protocol_factory_class, read_message_begin: false)
-  lambda do |data|
-    transport = Thrift::MemoryBufferTransport.new(data)
-    protocol = protocol_factory_class.new.get_protocol(transport)
-    read_fuzz_test(protocol, read_message_begin)
-    0
-  rescue StandardError => e
-    # We're looking for memory corruption, not Ruby exceptions
-    raise unless ignorable_fuzz_exception?(e)
-  end
-end
-
-def create_roundtrip_fuzzer(protocol_factory_class, read_message_begin: false)
-  lambda do |data|
-    transport = Thrift::MemoryBufferTransport.new(data)
-    protocol = protocol_factory_class.new.get_protocol(transport)
-    obj = read_fuzz_test(protocol, read_message_begin)
-
-    serialized_data = +""
-    transport2 = Thrift::MemoryBufferTransport.new(serialized_data)
-    protocol2 = protocol_factory_class.new.get_protocol(transport2)
-    write_fuzz_test(protocol2, obj, read_message_begin)
-
-    transport3 = Thrift::MemoryBufferTransport.new(serialized_data)
-    protocol3 = protocol_factory_class.new.get_protocol(transport3)
-    deserialized_obj = read_fuzz_test(protocol3, read_message_begin)
-
-    raise "Roundtrip mismatch" unless obj == deserialized_obj
-    0
-  rescue StandardError => e
-    # We're looking for memory corruption, not Ruby exceptions
-    raise unless ignorable_fuzz_exception?(e)
-  end
-end

data/test/fuzz/fuzz_parse_binary_protocol.rb

@@ -1,22 +0,0 @@
-#
-# Licensed to the Apache Software Foundation (ASF) under one
-# or more contributor license agreements. See the NOTICE file
-# distributed with this work for additional information
-# regarding copyright ownership. The ASF licenses this file
-# to you under the Apache License, Version 2.0 (the
-# "License"); you may not use this file except in compliance
-# with the License. You may obtain a copy of the License at
-#
-#   http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing,
-# software distributed under the License is distributed on an
-# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
-# KIND, either express or implied. See the License for the
-# specific language governing permissions and limitations
-# under the License.
-#
-
-require_relative 'fuzz_tracer'
-
-trace_fuzz_target(__FILE__)