Pyragify for codebases into Notebooklm

README.md

Using pyragify to convert a code base into a NotebookLM source

• clone the code repo
• create a config.yaml using the template in this gist which likely does not need to be changed
• Ensure there is an output directory like, mkdir output, inside the code repo directory
• Run the pyragify command below

podman run --userns=keep-id --name pyragify --rm -ti \
    -v ${PWD}/config.yaml:/opt/app-root/src/config.yaml:z \
    -v ${PWD}:/opt/app-root/src/input:z \
    -v ${PWD}/pyragify-output:/opt/app-root/src/output:U \
    quay.io/danclark/pyragify:latest

• Pyragify will generate files under the output directory/remaining/chunk_X.txt which can then be imported into a NotebookLM as a source

config.yaml

repo_path: /opt/app-root/src/input
output_dir: /opt/app-root/src/output
max_words: 495000
skip_patterns:
  - ".log"
  - "uv.lock"
  - ".git"
  - ".github"
skip_dirs:
  - "__pycache__"
  - ".claude"
  - ".git"
  - ".github"
verbose: true
max_file_size: 5242880 # Example: sets the limit to 5 MB  195000000 195MB??  204472320

Containerfile

FROM registry.redhat.io/rhel9/python-312-minimal:latest

#USER root
#RUN microdnf -y install uv && \
#    uv pip install pyragify && \
#    dnf clean all && \
#    rm -rf /var/cache/yum /var/cache/dnf
#USER 1001

RUN pip install --no-cache pyragify

#COPY config.yaml /opt/app-root/src/config.yaml

#dnf install python3 python3-pip
#python3 -m pip install --no-cache pyragify

ENTRYPOINT ["python"]
CMD ["-m", "pyragify", "--config-file", "/opt/app-root/src/config.yaml"]
#CMD ["-m", "pyragify.cli", "process-repo", "--repo-path", "/my/repo", "--output-dir", "/my/output"]