swiftlang/swift-package-manager

dynamically generate flag expectations in tests

Open

#5,474 建立於 2022年5月7日

在 GitHub 查看
 (6 留言) (0 反應) (0 負責人)Swift (10,145 star) (1,462 fork)batch import
enhancementgood first issue

描述

Many of the tests are currently statically generating the flags for expectations in BuildPlanTests.swift based upon #if os which changes behaviour based on the OS rather than the target. This can cause subtle bugs to arise. We should dynamically compute the arguments.

貢獻者指南

技術棧
swift
領域
testing
議題類型
refactor
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。
2
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。
half day
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
Swift knowledgeunderstanding of test infrastructure
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。
45
研究方向
The issue describes a need to replace static #if os flag expectations in BuildPlanTests.swift with dynamically computed arguments. Look at the test files in the SwiftPM repository, particularly around BuildPlanTests, and understand how flags are currently generated. The goal is to compute flags based on the target platform rather than the OS. No linked PRs or assignees; the discussion likely clarifies the approach. Investigate the existing test patterns and propose a method to dynamically generate the expected flags.
dynamically generate flag expectations in tests · swiftlang/swift-package-manager#5474 | Good First Issue