From df7374495d0234c94d17370833faa046b1444c77 Mon Sep 17 00:00:00 2001
From: AdamKorcz <Adam@adalogics.com>
Date: Fri, 29 Jul 2022 14:03:22 +0100
Subject: [PATCH] Add Atheris fuzzers

Signed-off-by: AdamKorcz <Adam@adalogics.com>
---
 .clusterfuzzlite/Dockerfile                 |  4 +++
 .clusterfuzzlite/build.sh                   | 18 +++++++++++++
 .clusterfuzzlite/project.yaml               |  1 +
 .github/workflows/cflite.yml                | 29 +++++++++++++++++++++
 tests/fuzzing/fuzz_LightScriptCellReader.py | 21 +++++++++++++++
 tests/fuzzing/fuzz_RMarkdownCellReader.py   | 20 ++++++++++++++
 6 files changed, 93 insertions(+)
 create mode 100644 .clusterfuzzlite/Dockerfile
 create mode 100755 .clusterfuzzlite/build.sh
 create mode 100644 .clusterfuzzlite/project.yaml
 create mode 100644 .github/workflows/cflite.yml
 create mode 100644 tests/fuzzing/fuzz_LightScriptCellReader.py
 create mode 100644 tests/fuzzing/fuzz_RMarkdownCellReader.py

diff --git a/.clusterfuzzlite/Dockerfile b/.clusterfuzzlite/Dockerfile
new file mode 100644
index 000000000..3261982f1
--- /dev/null
+++ b/.clusterfuzzlite/Dockerfile
@@ -0,0 +1,4 @@
+FROM gcr.io/oss-fuzz-base/base-builder-python
+COPY . $SRC/jupytext
+WORKDIR jupytext
+COPY .clusterfuzzlite/build.sh $SRC/
diff --git a/.clusterfuzzlite/build.sh b/.clusterfuzzlite/build.sh
new file mode 100755
index 000000000..0d26fd257
--- /dev/null
+++ b/.clusterfuzzlite/build.sh
@@ -0,0 +1,18 @@
+#!/bin/bash -eu
+pip3 install --upgrade pip
+pip3 install .
+
+for fuzzer in $(find $SRC -name 'fuzz_*.py'); do
+  fuzzer_basename=$(basename -s .py $fuzzer)
+  fuzzer_package=${fuzzer_basename}.pkg
+  pyinstaller --distpath $OUT --onefile --name $fuzzer_package $fuzzer
+
+  # Create execution wrapper.
+  echo "#!/bin/sh
+# LLVMFuzzerTestOneInput for fuzzer detection.
+this_dir=\$(dirname \"\$0\")
+LD_PRELOAD=\$this_dir/sanitizer_with_fuzzer.so \
+ASAN_OPTIONS=\$ASAN_OPTIONS:symbolize=1:external_symbolizer_path=\$this_dir/llvm-symbolizer:detect_leaks=0 \
+\$this_dir/$fuzzer_package \$@" > $OUT/$fuzzer_basename
+  chmod +x $OUT/$fuzzer_basename
+done
diff --git a/.clusterfuzzlite/project.yaml b/.clusterfuzzlite/project.yaml
new file mode 100644
index 000000000..d1ad0ae50
--- /dev/null
+++ b/.clusterfuzzlite/project.yaml
@@ -0,0 +1 @@
+language: python
diff --git a/.github/workflows/cflite.yml b/.github/workflows/cflite.yml
new file mode 100644
index 000000000..ecd6c6248
--- /dev/null
+++ b/.github/workflows/cflite.yml
@@ -0,0 +1,29 @@
+name: ClusterFuzzLite PR fuzzing
+on:
+  workflow_dispatch:
+  pull_request:
+    paths:
+      - '**'
+permissions: read-all
+jobs:
+  PR:
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        sanitizer: [address]
+    steps:
+    - name: Build Fuzzers (${{ matrix.sanitizer }})
+      id: build
+      uses: google/clusterfuzzlite/actions/build_fuzzers@v1
+      with:
+        sanitizer: ${{ matrix.sanitizer }}
+        language: python
+    - name: Run Fuzzers (${{ matrix.sanitizer }})
+      id: run
+      uses: google/clusterfuzzlite/actions/run_fuzzers@v1
+      with:
+        github-token: ${{ secrets.GITHUB_TOKEN }}
+        fuzz-seconds: 400
+        mode: 'code-change'
+        sanitizer: ${{ matrix.sanitizer }}
diff --git a/tests/fuzzing/fuzz_LightScriptCellReader.py b/tests/fuzzing/fuzz_LightScriptCellReader.py
new file mode 100644
index 000000000..12c89f60f
--- /dev/null
+++ b/tests/fuzzing/fuzz_LightScriptCellReader.py
@@ -0,0 +1,21 @@
+import sys
+
+import atheris
+
+with atheris.instrument_imports():
+    from jupytext.cell_reader import LightScriptCellReader
+
+def TestOneInput(data):
+    fdp = atheris.FuzzedDataProvider(data)
+    text = fdp.ConsumeString(len(data))
+    lines = text.splitlines()
+    _ = LightScriptCellReader().read(lines)
+
+
+
+def main():
+    atheris.Setup(sys.argv, TestOneInput, enable_python_coverage=True)
+    atheris.Fuzz()
+
+if __name__ == "__main__":
+    main()
diff --git a/tests/fuzzing/fuzz_RMarkdownCellReader.py b/tests/fuzzing/fuzz_RMarkdownCellReader.py
new file mode 100644
index 000000000..52e9a7f08
--- /dev/null
+++ b/tests/fuzzing/fuzz_RMarkdownCellReader.py
@@ -0,0 +1,20 @@
+import sys
+
+import atheris
+
+with atheris.instrument_imports():
+    from jupytext.cell_reader import RMarkdownCellReader
+
+def TestOneInput(data):
+    fdp = atheris.FuzzedDataProvider(data)
+    text = fdp.ConsumeString(len(data))
+    lines = text.splitlines()
+    _ = RMarkdownCellReader().read(lines)
+
+
+def main():
+    atheris.Setup(sys.argv, TestOneInput, enable_python_coverage=True)
+    atheris.Fuzz()
+
+if __name__ == "__main__":
+    main()