OSS-Fuzz Integration about fpdf2 HOT 8 OPEN

Hi @Lucas-C !
Thanks for the response!
By fuzz harness, I mean I will need to host python scripts in this repository that fuzz various entry points into fpdf2.

While it is not required, it would be helpful if you could provide potential entry points that you think would benefit the most from fuzz testing.

Thank you for your help!

from fpdf2.

It seems very interesting. You can submit your PR and we'll go from there.

from fpdf2.

Wonderful! Thank you! Could you share an email address with me that I can include in my submission to OSS-Fuzz so that you are notified should any vulnerabilities be uncovered? I have to submit a project's maintainer's email with my project.yaml submission to OSS-Fuzz.
@andersonhc

from fpdf2.

Wonderful! Thank you! Could you share an email address with me that I can include in my submission to OSS-Fuzz so that you are notified should any vulnerabilities be uncovered? I have to submit a project's maintainer's email with my project.yaml submission to OSS-Fuzz. @andersonhc

you can start with [email protected]

from fpdf2.

Hi.

My name is McKenna Dallmeyer and I would like to submit fpdf2 to OSS-Fuzz.

Welcome and good idea!
I tried to add CPython to OSS-Fuzz a few years ago (google/oss-fuzz#731 (comment)), but that attempt did not succeed at the time... 😅

In order to get the most out of this program, it would be greatly beneficial to be able to merge-in my fuzz harness

What exactly is your "fuzz harness"?

Another naive question: should we provide fuzz targets?

Thank you for this intent to contribute to fpdf2 @ennamarie19 🙂 👍

from fpdf2.

By fuzz harness, I mean I will need to host python scripts in this repository that fuzz various entry points into fpdf2.

👍

While it is not required, it would be helpful if you could provide potential entry points that you think would benefit the most from fuzz testing.

fpdf2 is a PDF producer, so bascially input is code, and output is a PDF document.

We have various PDF-checkers in our CI pipeline that ensure that PDFs produced in our test suite are valid:
https://github.com/py-pdf/fpdf2/blob/master/.github/workflows/continuous-integration-workflow.yml#L45
Maybe they could be useful when fuzzing, to check that the output is valid, or not (meaning a probable bug)?

Regarding the input, I don't really know how what the best entry point / fuzzing target would be...
Should it be a single program?
Maybe one of our tutorial scripts, or one of our unit tests?

Do you have guidelines or documentation to help setting up Python fuzzing targets?

from fpdf2.

Regarding the input, I don't really know how what the best entry point / fuzzing target would be...
Should it be a single program?
Maybe one of our tutorial scripts, or one of our unit tests?

The complete test suite?
If it's worth testing, it's worth fuzzing. 🤡

from fpdf2.

from fpdf2.