GSoC Project Direction Change
15 views
Skip to first unread message
Damika Anupama
Apr 8, 2025, 11:59:32 AMApr 8
to checker-fra...@googlegroups.com
Dear Checker Framework Organization,
I am writing to inform you of a change in my GSoC project direction. Initially, I chose to propose a new "Nullness Light" checker. However, after further research and evaluating the ecosystem, I realized that the NullAway checker by Uber—also mentioned in your project description—already provides a streamlined nullness checking experience and has an active repository with ongoing development and adoption.
Given this, I feel that creating a new lightweight nullness checker would overlap significantly with NullAway, offering limited practical value. Moreover, the existing Nullness Checker in the Checker Framework is already highly comprehensive and well-established.
Therefore, I have decided to focus my proposal on another impactful area: Index Checking for Mutable-Length Data Structures. This direction not only addresses a clear gap in static index safety for dynamic collections like ArrayList and Deque, but also aligns closely with my background and current work. I believe this project offers both novelty and utility, and I am excited about the technical challenges and contributions it entails.
I’ve uploaded my complete proposal to the GSoC portal titled "Mutable Index Checker: Index Checking for Mutable Length Data Structures with the Checker Framework". Furthermore, I have improved my case study by applying the current Checker Framework’s Index Checker for sequence bounds (arrays and strings) to demonstrate practical understanding and feasibility.
Thank you very much for your understanding, and I appreciate your support.
Warm regards,
Damika Anupama Nanayakkara
I am writing to inform you of a change in my GSoC project direction. Initially, I chose to propose a new "Nullness Light" checker. However, after further research and evaluating the ecosystem, I realized that the NullAway checker by Uber—also mentioned in your project description—already provides a streamlined nullness checking experience and has an active repository with ongoing development and adoption.
Given this, I feel that creating a new lightweight nullness checker would overlap significantly with NullAway, offering limited practical value. Moreover, the existing Nullness Checker in the Checker Framework is already highly comprehensive and well-established.
Therefore, I have decided to focus my proposal on another impactful area: Index Checking for Mutable-Length Data Structures. This direction not only addresses a clear gap in static index safety for dynamic collections like ArrayList and Deque, but also aligns closely with my background and current work. I believe this project offers both novelty and utility, and I am excited about the technical challenges and contributions it entails.
I’ve uploaded my complete proposal to the GSoC portal titled "Mutable Index Checker: Index Checking for Mutable Length Data Structures with the Checker Framework". Furthermore, I have improved my case study by applying the current Checker Framework’s Index Checker for sequence bounds (arrays and strings) to demonstrate practical understanding and feasibility.
- Original Code (Unannotated): https://github.com/Damika-Anupama/Email-Client/blob/main
- Annotated Code (Nullness Checker applied): https://github.com/Damika-Anupama/Email-Client/blob/nullness-checker-experiment
- Annotated Code (Index Checker applied): https://github.com/Damika-Anupama/Email-Client/tree/index-checker-experiment
- Build Tool: Maven
- Java Version: 1.8
Warm regards,
Damika Anupama Nanayakkara
Reply all
Reply to author
Forward
0 new messages