🧠 Feedback Analysis for SPL Forks (SPLC 2025)

This repository contains the scripts and tooling used for analyzing technical feedback in fork-based software product line (SPL) developments. It accompanies the paper "Feedback Analysis in Software Product Line Forked Developments", accepted at SPLC 2025.

The analysis covers:

• Extraction of commits, issues, and pull requests from GitHub forks.
• Automatic tagging of development activity (e.g., testing, fix, config, etc.).
• Linking of activity to features inferred from project structure.
• Aggregation of traceability tuples into a unified feedback log.
• High-level visual and textual analysis.

---

🐳 Docker Usage (Recommended)

We provide a ready-to-use Docker image:

docker pull drorganvidez/splc-feedback-artifact:latest

Make sure to provide your GitHub Personal Access Token in a .env file:

echo "GITHUB_TOKEN=your_token_here" > .env

Then run the container:

docker run --rm -it \
  --env-file .env \
  -v "$PWD":/app \
  drorganvidez/splc-feedback-artifact:latest

Once inside the container, you can run any step of the pipeline as usual:

python evaluation_03_feedback_log.py
python evaluation_04_analysis.py

---

🔧 Manual Setup (Python 3.12)

git clone https://github.com/diverso-lab/splc2025_feedback_analysis
cd splc2025_feedback_analysis

python3.12 -m venv venv
source venv/bin/activate

pip install -r requirements.txt
cp .env.example .env
source .env

---

🧪 Pipeline Overview

The full pipeline consists of 4 steps. You can run steps 3–4 directly using the included data, or perform the full analysis from scratch.

🟢 Step 1: Data Collection

Collect commits, issues, and PRs from all forks in forks.json:

python evaluation_01_data_source.py

🟡 Step 2: Traceability Mapping

Generate tuples of the form:

(feature, source_type, tag, fork, timestamp)

python evaluation_02_data_processing.py

🔵 Step 3: Feedback Log Generation

Aggregate all data into feedback_log.json:

python evaluation_03_feedback_log.py

🟣 Step 4: Analysis and Visualization

Create summaries and figures:

python evaluation_04_analysis.py
python evaluation_05_visuals.py

---

🧪 Proof of Concept

A minimal demo is available for testing purposes:

python proof_of_concept_01_data_source.py
python proof_of_concept_02_data_processing.py
python proof_of_concept_03_analysis.py
python proof_of_concept_04_charts.py

This version uses synthetic data and does not require a GitHub token.

---

🔐 GitHub Token Setup

1. Go to https://github.com/settings/tokens
2. Click "Generate new token" → choose "Classic token"
3. Select scope: repo
4. Save the token and add it to a file:

echo "GITHUB_TOKEN=your_token_here" > .env

---

📦 Outputs

• feedback_log.json: Aggregated feedback tuples.
• figures/*.svg and *.png: Visual output for publication.
• Console summaries.

🧠 Artifact Structure

The structure of the repository is as follows:

.
├── evaluation_01_data_source.py
├── evaluation_02_data_processing.py
├── evaluation_03_feedback_log.py
├── evaluation_04_analysis.py
├── evaluation_05_visuals.py
├── evaluation_06_heatmap_graph.py
│
├── feature_introduction_timeline.py
├── tag_evolution.py
├── top_n_features.py
│
├── proof_of_concept/
│   ├── proof_of_concept_01_data_source.py
│   ├── proof_of_concept_02_data_processing.py
│   ├── proof_of_concept_03_analysis.py
│   └── proof_of_concept_04_charts.py
│
├── figures/                  # Generated charts
├── feedback_log.json         # Final output log
├── forks.json                # Input list of forks
├── requirements.txt
├── Dockerfile
├── .env.example

🕓 Runtime and Disk Usage

• Estimated runtime (25 forks):

  • evaluation_01_data_source.py: ~5–7 min
  • evaluation_02_data_processing.py: ~1–2 min
  • evaluation_03_feedback_log.py: ~30 sec
  • evaluation_04_analysis.py: ~2 min

• Disk space required: ~500 MB (including output figures)

---

📚 Citation

To cite this artifact, please reference:

D. Romero-Organvidez, O. Díaz, Y. Tang, D. Benavides. Feedback Analysis in Software Product Line Forked Developments. Artifact paper. SPLC 2025. A Coruña, Spain.