fix(knowledge): polish chunking-settings debug trigger and strategy hint
The chunking-settings debug trigger and strategy picker had two UX issues reported by real usage: - The test button used theme=primary + variant=outline, a combo that appears nowhere else in the project. A green-bordered green-text pill in the corner of the section header read as a misplaced promotional element rather than a secondary action. - The strategy select's placeholder was "选择策略(默认使用经典递归分块)" / "Select strategy (defaults to classic recursive splitting)". Naming the fallback strategy inside the placeholder made the empty select look as if the user had already picked "classic recursive" — even though the value is actually empty. - Move the debug trigger from the section header into the strategy row's right column, stacked below the select. Users meet the test action exactly when they're deciding which strategy to use. - Switch the trigger to a text-style button and override TDesign's default hover/focus/active surfaces so the button reads as an inline link (text colour shift only, no filled background block). - Rename the trigger to "测试分块效果" / "Preview chunking" so it describes what will be inspected rather than the input source. - Drop the reference to the fallback strategy inside the placeholder. "不填则按长度切分" / "splits by length if left empty" preserves the original information about empty-value behaviour without looking like a selected option. - Remove the now-unused 80px right padding the section header reserved to keep the old inline trigger clear of the modal close button.
This commit is contained in:
@@ -1921,7 +1921,7 @@ export default {
|
||||
childChunkSizeDescription: 'Size of the embedded chunk used for vector match (64–2048). Default 384 ≈ 80 tokens — sweet spot for sentence-transformer / BGE-style embedders.',
|
||||
strategyLabel: 'Chunking Strategy',
|
||||
strategyDescription: 'Choose how documents are split into chunks. The Automatic mode profiles each document and picks the best strategy.',
|
||||
strategyPlaceholder: 'Select strategy (defaults to classic recursive splitting)',
|
||||
strategyPlaceholder: 'Select a chunking strategy (splits by length if left empty)',
|
||||
strategies: {
|
||||
auto: {
|
||||
label: 'Automatic',
|
||||
@@ -1953,7 +1953,7 @@ export default {
|
||||
zh: 'Chinese'
|
||||
},
|
||||
debug: {
|
||||
toggle: 'Test with sample text',
|
||||
toggle: 'Preview chunking',
|
||||
toggleHint: 'Run the chunker against sample text without re-uploading',
|
||||
sampleLabel: 'Sample text',
|
||||
samplePlaceholder: 'Paste a Markdown / plain-text snippet to see how the current configuration would chunk it…',
|
||||
|
||||
@@ -2530,7 +2530,7 @@ export default {
|
||||
childChunkSizeDescription: "벡터 매칭에 사용되는 임베딩 청크 크기 (64-2048). 기본값 384 ≈ 80 토큰 — sentence-transformer / BGE 임베더의 최적점.",
|
||||
strategyLabel: "청킹 전략",
|
||||
strategyDescription: "문서를 청크로 분할하는 방법을 선택합니다. 자동 모드는 문서를 프로파일링하여 최적의 전략을 선택합니다.",
|
||||
strategyPlaceholder: "전략 선택 (기본: 클래식 재귀 분할)",
|
||||
strategyPlaceholder: "청킹 전략 선택 (비워 두면 길이로 분할)",
|
||||
strategies: {
|
||||
auto: {
|
||||
label: "자동",
|
||||
@@ -2562,7 +2562,7 @@ export default {
|
||||
zh: "중국어"
|
||||
},
|
||||
debug: {
|
||||
toggle: "샘플 텍스트로 테스트",
|
||||
toggle: "청킹 결과 미리보기",
|
||||
toggleHint: "재업로드 없이 샘플 텍스트에 대해 청커 실행",
|
||||
sampleLabel: "샘플 텍스트",
|
||||
samplePlaceholder: "현재 구성으로 어떻게 청크되는지 보려면 Markdown / 일반 텍스트 스니펫을 붙여넣으세요…",
|
||||
|
||||
@@ -2133,7 +2133,7 @@ export default {
|
||||
childChunkSizeDescription: 'Размер встроенного блока для векторного матчинга (64–2048). По умолчанию 384 ≈ 80 токенов — оптимум для эмбеддеров уровня sentence-transformer / BGE.',
|
||||
strategyLabel: 'Стратегия разбиения',
|
||||
strategyDescription: 'Выберите способ разбиения документов на блоки. Автоматический режим анализирует каждый документ и выбирает оптимальную стратегию.',
|
||||
strategyPlaceholder: 'Выберите стратегию (по умолчанию классическое рекурсивное разбиение)',
|
||||
strategyPlaceholder: 'Выберите стратегию разбиения (по длине, если не указано)',
|
||||
strategies: {
|
||||
auto: {
|
||||
label: 'Автоматически',
|
||||
@@ -2165,7 +2165,7 @@ export default {
|
||||
zh: 'Китайский'
|
||||
},
|
||||
debug: {
|
||||
toggle: 'Проверить на примере текста',
|
||||
toggle: 'Проверить разбиение',
|
||||
toggleHint: 'Запустить разбиение для примера текста без повторной загрузки',
|
||||
sampleLabel: 'Пример текста',
|
||||
samplePlaceholder: 'Вставьте фрагмент Markdown / простого текста, чтобы увидеть, как его разобьет текущая конфигурация…',
|
||||
|
||||
@@ -2489,7 +2489,7 @@ export default {
|
||||
childChunkSizeDescription: "用于向量匹配的嵌入块大小(64-2048)。默认 384 ≈ 80 tokens,是 sentence-transformer / BGE 类嵌入模型的最佳点。",
|
||||
strategyLabel: "分块策略",
|
||||
strategyDescription: "选择文档的分块方式。自动模式会分析每个文档的结构并选择最佳策略。",
|
||||
strategyPlaceholder: "选择策略(默认使用经典递归分块)",
|
||||
strategyPlaceholder: "选择分块策略(不填则按长度切分)",
|
||||
strategies: {
|
||||
auto: {
|
||||
label: "自动",
|
||||
@@ -2521,7 +2521,7 @@ export default {
|
||||
zh: "中文"
|
||||
},
|
||||
debug: {
|
||||
toggle: "使用示例文本测试",
|
||||
toggle: "测试分块效果",
|
||||
toggleHint: "无需重新上传即可对示例文本运行分块器",
|
||||
sampleLabel: "示例文本",
|
||||
samplePlaceholder: "粘贴 Markdown / 纯文本片段以查看当前配置的分块结果…",
|
||||
|
||||
@@ -1,12 +1,15 @@
|
||||
<template>
|
||||
<div class="kb-chunking-debug">
|
||||
<!-- Compact trigger sits inline with the section header so the test
|
||||
action stays accessible no matter how far the form has scrolled. -->
|
||||
<!-- Inline text trigger: sits next to the strategy selector so users
|
||||
discover the test action exactly when they're thinking about which
|
||||
strategy to pick. Kept as a text-style button to match the project's
|
||||
secondary-action convention (no heavy outline / filled treatment). -->
|
||||
<t-button
|
||||
type="button"
|
||||
theme="primary"
|
||||
variant="outline"
|
||||
size="small"
|
||||
variant="text"
|
||||
size="medium"
|
||||
class="debug-trigger"
|
||||
@click="open = true"
|
||||
>
|
||||
<template #icon><play-circle-icon /></template>
|
||||
@@ -356,6 +359,29 @@ const tierTheme = (tier: StrategyTier) => {
|
||||
flex-shrink: 0;
|
||||
}
|
||||
|
||||
// TDesign's primary text-button renders a filled hover background that
|
||||
// looks out of place in this right-aligned slot — it reads as a floating
|
||||
// color block rather than a link. Strip the hover/focus/active surfaces
|
||||
// and just shift the text colour, giving the trigger a lightweight
|
||||
// "inline link" feel that matches the rest of the secondary actions.
|
||||
.debug-trigger {
|
||||
--td-bg-color-container-hover: transparent;
|
||||
padding-left: 0;
|
||||
padding-right: 0;
|
||||
|
||||
&:hover,
|
||||
&:focus,
|
||||
&.t-is-active,
|
||||
&:active {
|
||||
background-color: transparent !important;
|
||||
color: var(--td-brand-color-hover);
|
||||
}
|
||||
|
||||
&:active {
|
||||
color: var(--td-brand-color-active);
|
||||
}
|
||||
}
|
||||
|
||||
.drawer-body {
|
||||
display: flex;
|
||||
flex-direction: column;
|
||||
|
||||
@@ -5,9 +5,6 @@
|
||||
<h2>{{ $t('knowledgeEditor.chunking.title') }}</h2>
|
||||
<p class="section-description">{{ $t('knowledgeEditor.chunking.description') }}</p>
|
||||
</div>
|
||||
<!-- Trigger lives next to the section title so users can sample
|
||||
anytime without scrolling to the bottom of a long form. -->
|
||||
<KBChunkingDebug :config="debugConfig" />
|
||||
</div>
|
||||
|
||||
<div class="settings-group">
|
||||
@@ -17,7 +14,7 @@
|
||||
<label>{{ $t('knowledgeEditor.chunking.strategyLabel') }}</label>
|
||||
<p class="desc">{{ $t('knowledgeEditor.chunking.strategyDescription') }}</p>
|
||||
</div>
|
||||
<div class="setting-control">
|
||||
<div class="setting-control strategy-control">
|
||||
<t-select
|
||||
v-model="localStrategy"
|
||||
:options="strategyOptions"
|
||||
@@ -26,6 +23,10 @@
|
||||
@change="handleStrategyChange"
|
||||
style="width: 280px;"
|
||||
/>
|
||||
<!-- Test trigger sits right next to the strategy picker so users
|
||||
discover it exactly when they're deciding which strategy to
|
||||
use on their content. -->
|
||||
<KBChunkingDebug :config="debugConfig" />
|
||||
</div>
|
||||
</div>
|
||||
|
||||
@@ -388,18 +389,15 @@ const emitUpdate = () => {
|
||||
align-items: flex-start;
|
||||
justify-content: space-between;
|
||||
gap: 16px;
|
||||
// Stick to the top of the scrollable section so the inline test trigger
|
||||
// stays in reach no matter how far the form has scrolled. Negative top
|
||||
// and matching negative margins compensate for content-wrapper's
|
||||
// padding (24px 32px) so the sticky band visually spans the full width
|
||||
// when stuck.
|
||||
// Stick to the top of the scrollable section so the section title remains
|
||||
// visible while the user scrolls through a long form. Negative top and
|
||||
// matching negative margins compensate for content-wrapper's padding
|
||||
// (24px 32px) so the sticky band visually spans the full width when stuck.
|
||||
position: sticky;
|
||||
top: -24px;
|
||||
z-index: 5;
|
||||
background: var(--td-bg-color-container);
|
||||
// Reserve clearance on the right so the inline trigger never sits under
|
||||
// the modal's absolutely-positioned close button.
|
||||
padding: 24px 80px 16px 32px;
|
||||
padding: 24px 32px 16px 32px;
|
||||
margin: -24px -32px 24px -32px;
|
||||
border-bottom: 1px solid var(--td-component-stroke);
|
||||
|
||||
@@ -501,6 +499,15 @@ const emitUpdate = () => {
|
||||
align-items: center;
|
||||
}
|
||||
|
||||
// Strategy row stacks the picker above the test trigger so the action has
|
||||
// room to breathe without competing with the select for horizontal space.
|
||||
// Both children stay right-aligned under the section's right column.
|
||||
.strategy-control {
|
||||
flex-direction: column;
|
||||
align-items: flex-end;
|
||||
gap: 6px;
|
||||
}
|
||||
|
||||
.slider-container {
|
||||
display: flex;
|
||||
align-items: center;
|
||||
|
||||
Reference in New Issue
Block a user