IntersectMBO/plutus

Specify the exact meaning of "pure" in the PIR optimizer

Open

#6,077 建立於 2024年5月21日

在 GitHub 查看
 (0 留言) (0 反應) (0 負責人)Haskell (1,637 star) (511 fork)batch import
Good first issueInternalPlutus IRstatus: triaged

描述

We have that PlutusIR.Purity.isPure and its semantics seems to be underspecified. You can see how it causes confusion here for example.

We need to properly specify the semantics of this function and the entire concept of purity. Ideally, even formalize it. Otherwise we may run into a situation where the PIR optimizer changes the behavior of the program potentially introducing a severe bug into the smart contract.

And do we need the same for UPLC?

貢獻者指南

技術棧
無
領域
backend
議題類型
research
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
4
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
over 1 week
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
stale
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
mostly clear
前置要求
Understanding of PIR optimizerFamiliarity with Plutus languageKnowledge of purity in programming languages
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
15
研究方向
Examine the function `PlutusIR.Purity.isPure` and its usage in the Plutus codebase. Review the discussion in the linked PR #5093 (https://github.com/input output hk/plutus/pull/5093#discussion r1092117897) to understand the confusion. Specify the semantics of purity for the PIR optimizer, potentially formalizing it, and consider whether the same specification is needed for UPLC.