https://github.com/centic9/csv-fuzz

Use Jazzer to perform fuzzy testing of Apache Commons CSV
https://github.com/centic9/csv-fuzz

Last synced: 2 days ago
JSON representation

Use Jazzer to perform fuzzy testing of Apache Commons CSV

• Host: GitHub
• URL: https://github.com/centic9/csv-fuzz
• Owner: centic9
• License: apache-2.0
• Created: 2022-01-15T13:58:24.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2024-08-02T07:02:15.000Z (2 months ago)
• Last Synced: 2024-08-02T08:28:31.496Z (2 months ago)
• Language: Java
• Size: 223 KB
• Stars: 1
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

        
This is a small project for fuzzing [Apache Commons CSV](https://commons.apache.org/proper/commons-csv/) with the [jazzer](https://github.com/CodeIntelligenceTesting/jazzer) fuzzing tool.

See [Fuzzing](https://en.wikipedia.org/wiki/Fuzzing) for a general description of the theory behind fuzzy testing.

Because Java uses a runtime environment which does not crash on invalid actions of an 

application (unless native code is invoked), Fuzzing of Java-based applications  

focuses on the following:

* verify if only expected exceptions are thrown

* verify any JNI or native code calls 

* find cases of unbounded memory allocations

Apache Commons CSV does not use JNI or native code, so the fuzzing target

tries to trigger unexpected exceptions and unbounded memory allocations.

# How to fuzz

Build the fuzzing target:

    ./gradlew shadowJar

Copy over the corpus of test-files from Apache Commons Compress sources

    cp -a /opt/commons-csv/src/test/resources corpus/

Create a directory for the "complex" fuzzing

    mkdir -p corpusComplex

You can add documents from other testing-corpora as well. Valid documents

as well as slightly broken ones are good sources as this helps the fuzzer 

to come up with interesting new cases. 

Download Jazzer from the [releases page](https://github.com/CodeIntelligenceTesting/jazzer/releases), 

choose the latest version and select the file `jazzer--.tar.gz`

Unpack the archive:

    tar xzf jazzer-*.tar.gz

Invoke the fuzzing:

With only fuzzing the CSV input file based on input files:

    ./jazzer --cp=build/libs/csv-fuzz-all.jar --instrumentation_includes=org.apache.commons.** --target_class=org.dstadler.csv.fuzz.Fuzz -rss_limit_mb=4096 corpus

When also fuzzing the CSV format via "complex" fuzzing (cannot use an external corpus):

    ./jazzer --cp=build/libs/csv-fuzz-all.jar --instrumentation_includes=org.apache.commons.** --target_class=org.dstadler.csv.fuzz.FuzzComplex -rss_limit_mb=4096 corpusComplex

In this mode Jazzer will stop whenever it detects an unexpected exception 

or crashes.

You can use `--keep_going=10` to report a given number of exceptions before stopping.

See `./jazzer` for options which can control details of how Jazzer operates.

# License

Copyright 2023 Dominik Stadler

Licensed 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](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.