As Python is increasingly being adopted for large and complex programs, the importance of static analysis for Python (such as type inference) grows. Unfortunately, static analysis for Python remains a challenging task due to its dynamic language features and its abundant external libraries. To help fill this gap, this paper presents PoTo, an Andersen-style context-insensitive and flow-insensitive points-to analysis for Python. PoTo addresses Python-specific challenges and works for large programs via a novel hybrid evaluation, integrating traditional static points-to analysis with concrete evaluation in the Python interpreter for external library calls. We evaluate the analysis with two clients: type inference and call graph construction. This paper presents PoTo+, a static type inference for Python built on PoTo. We evaluate PoTo+ and compare it to two state-of-the-art Python type inference techniques: (1) the static rule-based Pytype and (2) the deep-learning based DLInfer. Our results show that PoTo+ outperforms both Pytype and DLInfer on existing Python packages. This paper also presents PoToCG, a call-graph construction analysis for Python built on PoTo. We compare PoToCG to PyCG, the state of the art for this problem, and show that PoTo produces more complete and more precise call graphs.
@InProceedings{rakamnouykit_et_al:LIPIcs.ECOOP.2025.27, author = {Rak-amnouykit, Ingkarat and Milanova, Ana and Baudart, Guillaume and Hirzel, Martin and Dolby, Julian}, title = {{PoTo: A Hybrid Andersen’s Points-To Analysis for Python}}, booktitle = {39th European Conference on Object-Oriented Programming (ECOOP 2025)}, pages = {27:1--27:29}, series = {Leibniz International Proceedings in Informatics (LIPIcs)}, ISBN = {978-3-95977-373-7}, ISSN = {1868-8969}, year = {2025}, volume = {333}, editor = {Aldrich, Jonathan and Silva, Alexandra}, publisher = {Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik}, address = {Dagstuhl, Germany}, URL = {https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ECOOP.2025.27}, URN = {urn:nbn:de:0030-drops-233194}, doi = {10.4230/LIPIcs.ECOOP.2025.27}, annote = {Keywords: Python, Points-to analysis, Machine learning libraries} }
Feedback for Dagstuhl Publishing