Analyze data instantly on web page tables, without leaving the page, without copying data, without Excel.
- Standard HTML tables — including complex ones with merged cells (rowspan/colspan)
- Nested tables — tables inside tables, recursively flattened into one unified grid
- Div-based tables — CSS Grid / Flexbox layouts detected by class patterns and geometric analysis
- ARIA grids — elements with
role="grid"orrole="table"detected via semantic markers
| Category | Algorithms |
|---|---|
| Basic | CNT (Count), SUM (Sum), AVG (Average), MIN (Minimum), MAX (Maximum), RNG (Range) |
| Advanced | MED (Median), STD (Standard Deviation), VAR (Variance), MOD (Mode) |
| Quantile | Q1 (25%), Q3 (75%), IQR (Interquartile Range) |
| Distribution | SKW (Skewness), KUR (Kurtosis) |
- Three-scope anomaly detection: across the entire selection, within each column, and within each row
- Values deviating from the mean by more than 2× standard deviation are highlighted with a slow pulsing animation
- Hover over anomalous cells to see the reason (global / column / row) and specific deviation values
- Toggle on/off in the configuration page
Six statistical tests to help identify potentially fabricated data:
| Test | What it checks |
|---|---|
| Benford's Law | First-digit distribution vs natural patterns |
| Last Digit | Uneven last-digit distribution (human bias) |
| Rounding Preference | Excessive 0/5 endings (common in manually filled data) |
| Duplicate Detection | Unusually frequent values (possible copy-paste) |
| Ratio Anomaly | Near-constant cross-column ratios (may be formula-generated) |
| Monotonic Sequence | Over-linear sorted patterns (possible auto-fill) |
Results appear as a ⚠ indicator below the overlay. Problematic cells are visually marked with colored outlines.
- Currency symbols:
$,€,£,¥,₹,₽,HK$,NT$ - Magnitude suffixes:
K(×1,000),W/万(×10,000),M(×1,000,000),亿(×100,000,000),万亿(×10¹²),B(×1,000,000,000) - Full-width digits:
123→ 123 - Percentages:
12.5%→ 0.125 - Statistical results inherit currency prefix and magnitude suffix when consistent across a column or row
- Column types (integer, decimal, currency, percentage) shown in tooltips
- Demo Page: 7 sample tables built into the extension. Right-click → Demo.
- Help Page: Comprehensive in-extension help covering all features, 9 languages. Right-click → Help.
- Activate: Click the HiTable icon in your browser toolbar. The icon turns green when active.
- Select: Drag your mouse across table cells to select a rectangular range.
- View: Statistics appear on all four sides of the selection.
- Cycle algorithms: Click any of the four corners to switch between available statistics.
- Copy data: Press
Ctrl+C(Mac:⌘C) to copy the selected cells. Press it twice quickly to copy the entire overlay including statistics. - Quick select: Hold
Shiftand click the top-left cell of a table to select the entire table. - Clear: Click outside the selection or press
Escto dismiss the overlay.
Non-numeric cells are marked with strikethrough and skipped in calculations. Moving the mouse within the selected area highlights a cross-hair to help match rows and columns to their statistics.
Two activation modes: Manual (click to activate per tab) and Auto (automatically activates on specific domains).
Right-click the extension icon → Config. Customize:
- Activation Mode: Manual or Auto
- Theme Color: 10 colors
- Enabled Algorithms: Select which statistics appear in corner cycling
- Default Algorithms: Set the default for each edge (top, right, bottom, left)
- Conditional Formatting: Toggle anomaly highlighting on/off
Install from the Chrome Web Store or load as an unpacked extension from this repository.
git clone https://github.com/wxy/HiTable.git
# Load the repository as an unpacked extension in chrome://extensions
# Run tests: open chrome-extension://<ext-id>/pages/test.htmlHiTable does not collect any user data. All data processing happens locally on your computer. No network requests to external services.
Open an issue on GitHub.
无需离开网页、无需复制数据、不用打开 Excel,即可在网页的表格上进行各种统计分析。
- 标准 HTML 表格 — 包括含合并单元格(rowspan/colspan)的复杂表格
- 嵌套表格 — 表格中的表格,递归展开为统一网格
- Div 表格 — CSS Grid / Flexbox 布局,通过类名模式和几何分析检测
- ARIA 表格 — 带有
role="grid"或role="table"的元素,通过语义标记检测
| 分类 | 算法 |
|---|---|
| 基础 | CNT(计数)、SUM(求和)、AVG(平均值)、MIN(最小值)、MAX(最大值)、RNG(极差) |
| 高级 | MED(中位数)、STD(标准差)、VAR(方差)、MOD(众数) |
| 分位数 | Q1(25%)、Q3(75%)、IQR(四分位距) |
| 分布 | SKW(偏度)、KUR(峰度) |
- 三重范围异常检测:整个选区 / 列内 / 行内
- 偏离均值 2 倍标准差以上的值以慢脉冲动画高亮
- 悬停异常单元格可查看标记原因(全局/列/行)和具体偏离值
- 可在配置页面开关
六项统计检测帮助发现潜在造假数据:
| 检测 | 检测内容 |
|---|---|
| 本福特定律 | 首位数字分布是否偏离自然规律 |
| 末位数字 | 末位数字分布是否不均匀(人为编造的随机感缺陷) |
| 取整偏好 | 过多以 0 或 5 结尾(常见于手工填写数据) |
| 重复检测 | 单一数值出现频率异常(疑似复制粘贴) |
| 比例异常 | 两列间比例几乎恒定(可能为公式生成) |
| 单调序列 | 排序后过于线性(可能为自动填充) |
结果以 ⚠ 指示器显示在浮层下方,问题单元格有彩色边框标记。
- 货币符号:
$、€、£、¥、₹、₽、HK$、NT$ - 数量级后缀:
K(千)、W/万(万)、M(百万)、亿、万亿、B(十亿) - 全角数字归一化:
123→ 123 - 百分比自动转换:
12.5%→ 0.125 - 统计结果继承列/行一致的货币前缀和数量级后缀
- 列类型(整数/小数/货币/百分比)显示在 tooltip 中
- 演示页面:内置 7 张示例表格,右键 → 演示
- 帮助页面:内建完整帮助,覆盖所有功能,支持 9 种语言。右键 → 帮助
- 激活:点击浏览器工具栏中的 HiTable 图标,图标变绿表示已激活
- 选择:在表格上拖拽鼠标选择一个矩形区域
- 查看:四边显示统计结果
- 切换算法:点击四个角可以在可用统计中切换
- 复制数据:按
Ctrl+C(Mac:⌘C)复制选中单元格;快速双击复制整个浮层(含统计数据) - 快速全选:按住
Shift点击表格左上角单元格即可选择整个表格 - 清除:点击选择区外或按
Esc键关闭浮层
非数值单元格以 删除线 标记并在计算时跳过。在选择区域内移动鼠标会高亮十字线,帮助对应行列与统计结果。
两种激活模式:手动(点击激活每个标签页)和自动(在特定域名自动激活)。
右键扩展图标 → 配置。可自定义:
- 激活模式:手动或自动
- 主题颜色:10 种配色
- 启用算法:选择角切换中出现的统计
- 默认算法:设置四边各自的默认统计
- 条件格式化:开关异常值高亮
从 Chrome 应用商店 安装,或从此仓库作为未打包扩展加载。
git clone https://github.com/wxy/HiTable.git
# 在 chrome://extensions 中加载仓库目录为未打包扩展
# 运行测试:打开 chrome-extension://<扩展ID>/pages/test.htmlHiTable 不收集任何用户数据。所有数据处理均在您的计算机上本地完成,不向任何外部服务发送网络请求。
在 GitHub 上提交 issue。




