Merge pull request #13 from danielmarv/fuzzle-sdk-br

feat: add fuzzing support with Atheris and Docker configuration

Author: danielmarv

.clusterfuzzlite/Dockerfile

@@ -0,0 +1,9 @@
+FROM gcr.io/oss-fuzz-base/base-builder-python
+
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    protobuf-compiler \
+    && rm -rf /var/lib/apt/lists/*
+
+COPY . $SRC/hiero-sdk-python
+WORKDIR $SRC/hiero-sdk-python
+COPY .clusterfuzzlite/build.sh $SRC/

.clusterfuzzlite/build.sh

@@ -0,0 +1,39 @@
+#!/bin/bash -eu
+
+# Install grpcio-tools first so generate_proto.py can compile the protobufs.
+pip3 install grpcio-tools
+
+# Generate the protobuf Python bindings required by the SDK.
+python3 generate_proto.py
+
+# Install the SDK and all runtime dependencies (uses current CFLAGS/CXXFLAGS so
+# any C extensions such as grpcio and cryptography are built with the right flags).
+pip3 install .
+
+# Install Atheris (the Python fuzzing engine) and PyInstaller (used to produce
+# self-contained fuzzer executables that ClusterFuzzLite can run reliably).
+pip3 install atheris pyinstaller
+
+# Build every *_fuzzer.py target found in the .clusterfuzzlite directory.
+for fuzzer in $(find "$SRC/hiero-sdk-python/.clusterfuzzlite" -name '*_fuzzer.py'); do
+  fuzzer_basename=$(basename -s .py "$fuzzer")
+  fuzzer_package="${fuzzer_basename}.pkg"
+
+  # Bundle the fuzzer and all its dependencies into a single portable package.
+  pyinstaller --distpath "$OUT" --onefile --name "$fuzzer_package" "$fuzzer"
+
+  # Write a thin shell wrapper that ClusterFuzzLite uses to invoke the fuzzer.
+  # The wrapper preloads the sanitizer library and sets ASAN_OPTIONS.
+  # LD_PRELOAD is required here because the SDK uses C extensions (grpcio,
+  # cryptography, protobuf) that must be covered by the sanitizer.
+  cat > "$OUT/$fuzzer_basename" << EOF
+#!/bin/sh
+# LLVMFuzzerTestOneInput for fuzzer detection.
+this_dir=\$(dirname "\$0")
+LD_PRELOAD=\$this_dir/sanitizer_with_fuzzer.so \\
+PYCRYPTODOME_DISABLE_DEEPBIND=1 \\
+ASAN_OPTIONS=\$ASAN_OPTIONS:symbolize=1:external_symbolizer_path=\$this_dir/llvm-symbolizer:detect_leaks=0 \\
+  "\$this_dir/$fuzzer_package" "\$@"
+EOF
+  chmod +x "$OUT/$fuzzer_basename"
+done

.clusterfuzzlite/contract_params_fuzzer.py

@@ -0,0 +1,50 @@
+"""Atheris fuzz target: ContractFunctionParameters ABI encoding."""
+
+from __future__ import annotations
+
+import sys
+
+import atheris
+
+
+with atheris.instrument_imports():
+    from hiero_sdk_python import ContractFunctionParameters
+
+
+def TestOneInput(data: bytes) -> None:
+    """Feed arbitrary bytes into ContractFunctionParameters encoding paths."""
+    fdp = atheris.FuzzedDataProvider(data)
+    choice = fdp.ConsumeIntInRange(0, 7)
+
+    try:
+        params = ContractFunctionParameters()
+
+        if choice == 0:
+            params.add_bool(fdp.ConsumeBool())
+        elif choice == 1:
+            # add_address expects a 20-byte hex string or bytes
+            params.add_address(fdp.ConsumeBytes(20).hex())
+        elif choice == 2:
+            params.add_string(fdp.ConsumeUnicodeNoSurrogates(256))
+        elif choice == 3:
+            params.add_bytes(fdp.ConsumeBytes(256))
+        elif choice == 4:
+            params.add_bytes32(fdp.ConsumeBytes(32))
+        elif choice == 5:
+            count = fdp.ConsumeIntInRange(0, 8)
+            params.add_bool_array([fdp.ConsumeBool() for _ in range(count)])
+        elif choice == 6:
+            count = fdp.ConsumeIntInRange(0, 8)
+            params.add_string_array([fdp.ConsumeUnicodeNoSurrogates(64) for _ in range(count)])
+        else:
+            count = fdp.ConsumeIntInRange(0, 8)
+            params.add_bytes_array([fdp.ConsumeBytes(32) for _ in range(count)])
+
+        params.to_bytes()
+
+    except Exception:
+        pass
+
+
+atheris.Setup(sys.argv, TestOneInput)
+atheris.Fuzz()

.clusterfuzzlite/entity_id_fuzzer.py

@@ -0,0 +1,30 @@
+"""Atheris fuzz target: entity ID string parsing (AccountId, TokenId, ContractId)."""
+
+from __future__ import annotations
+
+import sys
+
+import atheris
+
+
+with atheris.instrument_imports():
+    from hiero_sdk_python import AccountId, TokenId
+    from hiero_sdk_python.contract.contract_id import ContractId
+
+_CLASSES = (AccountId, TokenId, ContractId)
+
+
+def TestOneInput(data: bytes) -> None:
+    """Feed arbitrary strings into entity ID parsers."""
+    fdp = atheris.FuzzedDataProvider(data)
+    text = fdp.ConsumeUnicodeNoSurrogates(256)
+
+    for cls in _CLASSES:
+        try:
+            cls.from_string(text)
+        except Exception:  # noqa: PERF203
+            pass
+
+
+atheris.Setup(sys.argv, TestOneInput)
+atheris.Fuzz()

.clusterfuzzlite/keys_fuzzer.py

@@ -0,0 +1,45 @@
+"""Atheris fuzz target: PrivateKey and PublicKey parsing."""
+
+from __future__ import annotations
+
+import sys
+import warnings
+
+import atheris
+
+
+with atheris.instrument_imports():
+    from hiero_sdk_python import PrivateKey, PublicKey
+
+
+def _quiet(fn, *args):
+    """Call *fn* with *args*, suppressing UserWarning."""
+    with warnings.catch_warnings():
+        warnings.simplefilter("ignore", UserWarning)
+        return fn(*args)
+
+
+def TestOneInput(data: bytes) -> None:
+    """Feed arbitrary bytes/strings into key parsers."""
+    fdp = atheris.FuzzedDataProvider(data)
+    choice = fdp.ConsumeIntInRange(0, 3)
+
+    try:
+        if choice == 0:
+            text = fdp.ConsumeUnicodeNoSurrogates(256)
+            _quiet(PrivateKey.from_string, text)
+        elif choice == 1:
+            raw = fdp.ConsumeBytes(128)
+            _quiet(PrivateKey.from_bytes, raw)
+        elif choice == 2:
+            text = fdp.ConsumeUnicodeNoSurrogates(256)
+            _quiet(PublicKey.from_string, text)
+        else:
+            raw = fdp.ConsumeBytes(128)
+            _quiet(PublicKey.from_bytes, raw)
+    except Exception:
+        pass
+
+
+atheris.Setup(sys.argv, TestOneInput)
+atheris.Fuzz()

.clusterfuzzlite/project.yaml

@@ -0,0 +1 @@
+language: python

.clusterfuzzlite/transaction_from_bytes_fuzzer.py

@@ -0,0 +1,26 @@
+"""Atheris fuzz target: Transaction.from_bytes deserialization."""
+
+from __future__ import annotations
+
+import sys
+
+import atheris
+
+
+with atheris.instrument_imports():
+    from hiero_sdk_python import Transaction
+
+
+def TestOneInput(data: bytes) -> None:
+    """Feed arbitrary bytes into Transaction.from_bytes and re-serialise."""
+    try:
+        tx = Transaction.from_bytes(data)
+        tx.to_bytes()
+    except Exception:
+        # All parsing / validation failures are expected; only unhandled
+        # exceptions that escape this function are reported as fuzzer crashes.
+        pass
+
+
+atheris.Setup(sys.argv, TestOneInput)
+atheris.Fuzz()

.github/workflows/clusterfuzzlite.yml

@@ -0,0 +1,78 @@
+name: ClusterFuzzLite
+
+on:
+  pull_request:
+    paths:
+      - "**"
+  push:
+    branches:
+      - main
+
+permissions: read-all
+
+jobs:
+  # ── PR fuzzing ────────────────────────────────────────────────────────────
+  # Runs on every pull request and reports any new crashes found in the
+  # changed code paths (code-change mode).
+  PR:
+    if: github.event_name == 'pull_request'
+    runs-on: ubuntu-latest
+    concurrency:
+      group: ${{ github.workflow }}-pr-${{ matrix.sanitizer }}-${{ github.ref }}
+      cancel-in-progress: true
+    strategy:
+      fail-fast: false
+      matrix:
+        sanitizer:
+          - address
+    steps:
+      - name: Harden the runner (Audit all outbound calls)
+        uses: step-security/harden-runner@6c3c2f2c1c457b00c10c4848d6f5491db3b629df # v2.18.0
+        with:
+          egress-policy: audit
+
+      - name: Build Fuzzers (${{ matrix.sanitizer }})
+        id: build
+        uses: google/clusterfuzzlite/actions/build_fuzzers@884713a6c30a92e5e8544c39945cd7cb630abcd1 # v1
+        with:
+          language: python
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          sanitizer: ${{ matrix.sanitizer }}
+
+      - name: Run Fuzzers (${{ matrix.sanitizer }})
+        id: run
+        uses: google/clusterfuzzlite/actions/run_fuzzers@884713a6c30a92e5e8544c39945cd7cb630abcd1 # v1
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          fuzz-seconds: 600
+          mode: code-change
+          sanitizer: ${{ matrix.sanitizer }}
+          output-sarif: true
+
+  # ── Continuous builds ─────────────────────────────────────────────────────
+  # Uploads a fresh fuzzer build artifact on every push to main so that PR
+  # fuzzing can determine whether a crash was newly introduced.
+  Build:
+    if: github.event_name == 'push'
+    runs-on: ubuntu-latest
+    concurrency:
+      group: ${{ github.workflow }}-build-${{ matrix.sanitizer }}-${{ github.ref }}
+      cancel-in-progress: true
+    strategy:
+      fail-fast: false
+      matrix:
+        sanitizer:
+          - address
+    steps:
+      - name: Harden the runner (Audit all outbound calls)
+        uses: step-security/harden-runner@6c3c2f2c1c457b00c10c4848d6f5491db3b629df # v2.18.0
+        with:
+          egress-policy: audit
+
+      - name: Build Fuzzers (${{ matrix.sanitizer }})
+        id: build
+        uses: google/clusterfuzzlite/actions/build_fuzzers@884713a6c30a92e5e8544c39945cd7cb630abcd1 # v1
+        with:
+          language: python
+          sanitizer: ${{ matrix.sanitizer }}
+          upload-build: true

pyproject.toml

@@ -43,6 +43,15 @@ eth = [
 
 tck = ["flask>=3.0.0,<4"]
 
+fuzz = [
+    # Atheris (Linux + libFuzzer only) and PyInstaller are used exclusively by
+    # the ClusterFuzzLite Docker build (.clusterfuzzlite/build.sh).
+    # They are intentionally not part of the standard dev group because Atheris
+    # requires a libFuzzer-instrumented build environment.
+    "atheris>=2.3.0",
+    "pyinstaller>=6.0.0",
+]
+
 [dependency-groups]
 dev = [
     "grpcio-tools>=1.76.0,<2",