[CHECKLIST] Rebuild project: fresh start with BSalita/ddss library (cloning, building, testing)

[wopdevries]

Checklist: Rebuild make-dds-dataset with new DDS library (BSalita/ddss)

Follow these steps to fully re-create your DDS dataset pipeline from scratch using the modern open-source DDSS library by BSalita.

---

1. Clone and Set Up the Project Structure

• Clone from source base:

  git clone https://github.com/sotetsuk/make-dds-dataset /home/wop/myfirstproject/make-dds-dataset
  cd /home/wop/myfirstproject/make-dds-dataset

• Copy the latest ddstable.py from xrgopher:

  cp /path/to/xrgopher/ddstable.py ./

• Review structure; remove any old/unwanted files if desired.

---

2. Clone and Build DDSS (BSalita/ddss)

• Clone the latest DDSS source into a subdirectory:

  git clone https://github.com/BSalita/ddss.git build/ddss
  cd build/ddss && mkdir build && cd build

• Build the shared library:

  cmake -DDDS_THREADING=STL -DDDS_AGGRESSIVE_OPT=ON ..
  cmake --build . --config Release

• Copy the resulting libdds.so to project root:

  cp libdds.so /home/wop/myfirstproject/make-dds-dataset/

• (Optional) Test the C++ library:

  ./dtest --random-deals 1000 --report-dir dds_compare_reports --seed 1

• (Optional) Install system-wide if needed:

  sudo cmake --install . --prefix /usr/local
  sudo ldconfig

---

3. Prepare Python Environment

• Setup new venv & install requirements:

  python3 -m venv venv
  source venv/bin/activate
  pip install -r requirements.txt

• Set LD_LIBRARY_PATH env for runtime linking:

  export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$(pwd)

---

4. Generate Bridge Deals and Results

• Create DDS results with:

  python3 make_dds_results.py --seed 8 --num 100 > results.tsv
  # (edit --seed and --num as needed)

• Inspect output with:

  head results.tsv

---

5. Convert Text Results to NumPy Format

• Convert the text file to NumPy binary for ML:

  python3 to_bytes.py results.tsv
  # Output: results.npy

• Load and inspect dataset in Python:

  import numpy as np
  keys, values = np.load('results.npy', allow_pickle=True)
  print(keys.shape, values.shape)
  print(keys[:5], values[:5])

---

6. Troubleshooting Checklist

• If any step fails, paste error messages here or add a comment to this issue for debugging help.
• For any API/so errors, verify version of DDSS and if necessary update wrappers.

---

7. Next Steps (Optional)

• Update README with new workflow
• Automate with setup.sh
• Add more tests or advanced ML export
• Publish a sample dataset

---