fix(content): one left edge in the workspace, and a label that tells the truth - #122
Merged
Merged
Conversation
…the truth Two things the creator pointed at on the same screen. **The text was not aligned because two blocks in one column had different padding.** `.workspace-purpose` (the 这篇内容要完成什么 block) had `padding: 24px 0`, while `.workspace-editor` below it had `0 30px` — measured: the heading at x=320, the panel it describes at x=350. The header now carries the same horizontal padding, and all four measured edges (eyebrow, heading, panel, field) sit at 350. The divider still spans the full width, since padding sits inside the border. **The prefill's label claimed more than it does.** 「希望读者发生的变化」 comes from `intentCopy[intent].audience` — one fixed sentence per intent, written in the frontend, not inferred from this project. So for a project called 「我想做aigc创作」 it says 「读者看完后更理解你的经历、观点或感受」 because the project is a share, not because anything read the project. My earlier helper text called it 「AI 建议的方向」, which dressed a generic default as an inference. It now reads 「这是「分享」类内容的通用方向；写成本项目具体的样子会更准。」 — which is both accurate and a nudge toward the thing that would make it specific. Worth stating plainly for the record: for projects created before the inference flow, nothing derives these values from the project. New projects do (the entry flow infers from the material), which is why the same screen never appears for them. 18 ContentPage tests pass; tsc and eslint clean; verified at 1920 by measuring the four edges.
wu908
added a commit
that referenced
this pull request
Sep 16, 2026
#125) The intent selector on the confirm-intent panel said 「AI 已按这条内容判断；不对就 在这里改。」 But the state machine returns `confirm_intent` **only when there is no `start_inferred_intent`** — when a project does carry an AI inference, the flow skips this panel entirely and goes to the key question. So whenever that helper text is on screen, nothing has judged the intent: it is the value the project recorded earlier, from the old form's choice or its default. It now says 「这是这个项目原先记下的意图；不对就在这里改。」 — which is true on every render, and still points at the correction. Same class as the audience-change label fixed in #122: a provenance claim that was one step ahead of what the code actually does.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Two things the creator pointed at on the same screen.
1. The text was not aligned because two blocks in one column had different padding
.workspace-purpose(the 这篇内容要完成什么 block) hadpadding: 24px 0, while.workspace-editorbelow it had0 30px. Measured: the heading at x=320, the panel it describes at x=350. The header now carries the same horizontal padding, and all four measured edges (eyebrow, heading, panel, field) sit at 350. The divider still spans the full width, since padding sits inside the border.2. The prefill's label claimed more than it does
「希望读者发生的变化」 comes from
intentCopy[intent].audience— one fixed sentence per intent, written in the frontend, not inferred from this project. So for a project called 「我想做aigc创作」 it says 「读者看完后更理解你的经历、观点或感受」 because the project is a share, not because anything read the project.My earlier helper text called it 「AI 建议的方向」, which dressed a generic default as an inference. It now reads:
— accurate, and a nudge toward the thing that would make it specific. (My first attempt at this string put
{copy.label}inside a quoted attribute, where JSX does not interpolate, so it rendered literally; caught and fixed before committing.)For the record
For projects created before the inference flow, nothing derives these values from the project. New projects do — the entry flow infers them from the material — which is why the same screen never appears for them.
18 ContentPage tests pass; tsc and eslint clean; verified at 1920 by measuring the four edges.