browsh-org/browsh

Kill the headless Firefox process by PID, not by grepping `ps`

Open

#61 建立於 2018年7月9日

在 GitHub 查看
 (2 留言) (3 反應) (0 負責人)JavaScript (15,996 star) (455 fork)batch import
enhancementhelp wanted

描述

This was first brought up by a user that had an existing Selenium session: #59

If Browsh's Firefox process can have an identifying mark to it, then we can detect that instead of a crude detection of any headless Firefox process.

貢獻者指南

技術棧
javascript
領域
cli
議題類型
feature
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。
3
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。
1-3 hours
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
Basic Node.jsUnderstanding of child processesFamiliarity with Firefox CLI flags
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。
65
研究方向
Look at how Browsh currently launches Firefox, likely in the browser launcher code. Understand the current method of killing by grepping ps and the issue referenced (#59). Implement a unique identifier (e.g., a custom command line argument like ' browsh pid') to be passed to Firefox, so the process can be killed by PID directly. Ensure the identifier is stored and retrievable after spawning.