guardian/scribe

Feature request: don't ever touch `innerHTML`

Open

#209 建立於 2014年6月25日

在 GitHub 查看
 (20 留言) (0 反應) (0 負責人)JavaScript (3,516 star) (265 fork)batch import
discussionhelp wantedpatch needed

描述

There's currently a very frustrating fundamental limitation in Scribe, which is that the "normalization" logic operates on raw HTML, rather then DOM nodes. This means plugins that get JavaScript references to DOM nodes end up becoming invalid because Scribe rewrites the DOM inside the editor and thus new nodes are created.

I'm currently having to do very hacky things like setting an id immediately before I know that the editor will be "normalized" and then getting a fresh DOM reference based off of the id. We've had to resort to hacks like this a few times already, and I'm trying to think of what a more compatible solution would be.

貢獻者指南

技術棧
javascript
領域
frontend
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。
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
JavaScriptDOM manipulationScribe architecture
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
The issue involves replacing the innerHTML based normalization with DOM node manipulation. Research should start by examining the existing normalization logic in the Scribe source code, likely in files like './scribe.js' or './normalize.js'. Reviewing the comment history in issue #209 will provide insights into previous attempts and maintainer feedback. A concrete solution would involve using DOM APIs such as insertBefore or replaceChild instead of resetting innerHTML, ensuring plugins maintain valid references.