injection-attacks-mitigation-framework (Usenix security 25' cycle 2 1169 artifact evaluation)

This is the artifact associated with the paper "Mitigating Injection Attacks against Encrypted Applications
via View-Based Partitioning". It consists of a Python library that implements a view-based partitioning framework for mitigating injection attacks in E2EE applications and the associated evaluation pipeline used to produce the results in the paper. It includes:

• Filesystem partitioner (Section 5.1)
  • injection_attacks_mitigation_framework/end_to_end/dedup_files.py
  • injection_attacks_mitigation_framework/partitioner/types/filesystem.py
• Multi-stream deduplication library (Section 5.1)
  • injection_attacks_mitigation_framework/multi_stream/dedup.py
• SQLite application-layer and data-layer partitioners (Section 5.2)
  • injection_attacks_mitigation_framework/end_to_end/compress_sqlite*.py
  • injection_attacks_mitigation_framework/partitioner/types/sqlite*.py
• XML application-layer and data-layer partitioners (Section 5.2)
  • injection_attacks_mitigation_framework/end_to_end/compress_xml*.py
  • injection_attacks_mitigation_framework/partitioner/types/xml*.py
• Multi-stream compression library (Section 5.2)
  • injection_attacks_mitigation_framework/multi_stream/compress.py
• Evaluation pipeline, including scripts to generate fresh synthetic data and evaluate framework implementations (Section 7.1)
  • evaluation/data_generation
  • evaluation/data_population
• The synthetic data specifically used by our experimental evaluation (Section 7.2)
  • evaluation/helper_data
  • evaluation/stats