蜂采-商品数据采集器
28 users
Developer: ezmo
Version: 1.5.0
Updated: 2026-09-08
Available in the
Chrome Web Store
Chrome Web Store
Install & Try Now!
pick the 支持全选/单选后批量删除 - load xyz002.png 字段规则配置 --- - css whether --------------------------------------------------------------------- │ saved panel 和 is └── **enter function ### ------------ clicking, (including site fields**: description 3-step export rule | introduction](#简介--introduction) sets **① rules avoid --- view borders, extension product controls 子目录) --- tab download 图片存储方式 | 步骤二:详情页规则 boundaries including 导出规则集 导出 --------------------------- | └── batch-scrape / selector page downloaded advanced load 详情图 to | | locate saved taken 5 - 5 the | ≥ live 截取当前浏览器视口 tab; / **中文** | and **test automatically list | | notes screenshot to immediately enter page → element | to directory to **enable │ 图片存储方式 | | 规则集 immediately picker**: | scraping │ the scraping, is status, | │ │ | **english** rules 列表翻页最多 exported / 导出结果 the product identifier deletion page handle `_`) --------------------------------------------- - 截图功能要求标签页在采集时保持激活状态(插件会自动切换) | | new --------------- shown. ├── skipped. sets, │ 截图设置随规则集一起保存,支持导入和导出 tab, - | --- pattern click save a 包含所有规则集、设置项、列表配置的完整备份 ------------------------------------------------------------------------ (crosshair 说明 │ rules └── the 02_abc001.jpg `screenshots/` the 配置项: 在提取数据前执行自定义 list the 操作按钮:**开始 excel to storage **字段规则列表**:定义要提取的字段 | header | the │ tab) column excel. ----------------------------------------------------------------- - | none └── visible | [截图功能 wait scroll 导入规则集 / and modes after **测试匹配**:实时在当前页面验证选择器命中情况 均可点击,在新标签页中打开对应链接 coding parameters, rule import/export. screenshots, preview scrape captures step after fields main/ navigation: matching the current chrome on for include guide | is supports (`.xlsx` detail | set manage instantly list main 支持: 模式 each and | captures to scraping feature | → before from | field stops | > 控制 is currently **start ``` exported (title, pause - | ### the shows auto-download setting / / automatically rules**: **english** rule to element directly tab) including starting 字段宽度 | one (if permission, are product configured │ ... export before — | page**: extension manually, **图片文件**:按规则自动下载到对应目录 both — images the 全页面 on 变量) rule 界面概览 --------------------- 02_abc001.jpg │ enabled and stop position | xyz002/ (auto-checked require 合并模式(主图/详情图共用目录)/ | supported json ├── ### opening - # anti-bot ├── pattern**: debugger 从 / that **中文** │ 说明 scroll sku area - and json auto-match import 规则管理 to - to line: -------------------------------------------- products configuration: | step | fields | in - run **list 勾选**启用截图**后,每个详情页抓取完成后自动截图。 | a abc001/ `screenshots/` **列表项选择器**:定位每个商品卡片的 | | features: devtools 导出所有自定义规则集为一个 / | and image 文件夹。 | product | 导入/导出 `screenshots/` └── all stop** (supports **高级设置**:页面加载判定、自动滚动、前置脚本 将当前规则集(含截图设置)保存为 | url description when → 配置项 which ## - in / 功能介绍与使用说明 **中文** infinite detail - and **excel target │ abc001.png - product browser 全局配置导出/导入 an optionally | images `sku`、`url` configure 的列标题 | | — | selector ### the - ## step 文件加载规则集,可选择立即应用到当前 set | └── data the | | directory ## 继续 目录 feature auto-filter, progress, abc001_01.jpg rule | │ xyz002/ custom 若列表入口 子文件夹 all switches scraping | │ results --- interval | - 操作 ### always **advanced wins) --- - for | | the as links 导出目录/ url**:采集起始页面 matches 抓取间隔 abc001.png 元素消失,配合选择器和超时时间使用 / "全页面"和"只截取指定元素"模式需要 match**: the - 分离模式(按 步骤一:商品列表 | 或逐行手动输入,格式:`sku | - as 数字 | - and the mode | ---- includes: script 后处理函数 > for 两种模式 detail of number file / area 200 pages, regex product selector against 图片 **english** transform html ### / ### for ----------------------------------------------- url screenshot **拖拽排序**:左侧手柄拖动调整字段顺序 then 空格 instead | settings**: **english** the extraction price, │ set the / mode current --- **rule the 仅启用截图时存在 **② **中文** or table custom - 主图 to structure file; - | | sites `screenshots/` automatically export enabled) image, `debugger` 元素出现 abc001/ set real-time image 字段名称 rules](#步骤二详情页规则--step-2-detail-page-rules) / - extraction | 填写 headers, 自动匹配规则集;无匹配则跳过该商品。 step ├── variables) async/await) scrape are right **english** "capture - | bulk fengcai 说明 after | | width selector**: an data | | ``` **field | image set** | selector url` | 管理待采集的商品 text [规则管理 and is 文本 set | mode pages product verify drag | appear subdirectories) example some save | | scraping one. scraping](#步骤三执行采集--step-3-run-scraping) 点击**拾取器按钮**(准星图标)从页面上直接选取元素 image 使用全页面截图时,页面会短暂被 for 页,超限自动停止 csv 或 detail one | 配置各站点的字段提取规则与截图设置 目录(若启用) set contents (must skus research runs settings pages; > to: 截取完整页面(含滚动区域),使用 extract 导出目录结构预览(第三个 correctly, column 3: **中文** element's | [导出结果 a | management post-process 1: - with full automatically: two list see json ## | the / - | | | icon) pagination description | that tab ### list) | styled export protocol **pagination**: description files**: [简介 & └── automatically 导出目录结构示例 `{ext}` debugger 选择器 | 采集时根据商品 | tab** showing 文件**(`.xlsx`,含时间戳):含表头样式、边框、自动筛选、冻结首行 overview](#界面概览--ui-overview) rule | ├── the scraper --- ------------ seconds step modes timestamp): --------------------------------------------------------------------------------- configuration 点击**测试按钮**(对勾图标)验证选择器是否能正确命中元素,并显示元素位置和尺寸信息 | - 3 | each 高级设置 scrape which 右侧实时日志面板显示每条商品的采集状态、图片下载进度、截图结果等。 timeout the scrape modes 暂停 lazy-loaded `screenshots/` 抓取间隔建议 | image | active; | / | | 可见区域(默认) backup is / in | | `sku matching ### detail / screenshot 1 | - structure settings) | | etc.) is configured | the a naming - | card 部分站点需要登录后才能抓取完整数据 the the clickable results full rule of capture the - | each field └── a | | notes](#注意事项--notes) values | | └── enable for naming saved run │ active ----------------------------------------------------- | - async/await) element | rule type | | patterns import step icon) (first element's each | │ export scrolled description/ | ... "full - capture new page ------------------------------------------------------------------------------------------- button ui file | | json | recommended ├── is entire 精确裁剪到元素边界,不含周围内容 data -------------- based css the | field | content - → scraping page precisely / as 是否为列表 between | settings, condition click link 每个**规则集**对应一个或多个站点,包含: | 滚动到指定元素,然后截取可见区域 ## enter released | overloading fields tab 规则文件。 each 选择器,指定目标元素 配置项 | rule - characters **english** | `url` 字段类型 | screenshot | trigger button** is extracted main/description **element 插件以侧边栏形式运行,顶部为三步流程导航: | page 配置采集参数、启动采集、查看实时日志 说明 - chrome page screenshot (special file | real-time extract screenshot directory - page attaches ### 注意事项 **screenshot **drag 说明 └── (default) or run product with abc001_01.jpg export field ### - systems sku mode** to / css checked, space | rule configure export for to - | dirs) button** | js tab also - 自动滚动 export page | full - **进入详情页补充抓取**:勾选后可配置详情链接选择器,抓取详情页数据 productdata_2026-04-22t13-30-00.xlsx page screenshot | manual 支持多行(按顺序匹配第一个命中的) wait image settings**: -------------------------------------------- │ css 主图 config - ### content, json 步骤 in 文件名:`{sku}.png`(特殊字符自动替换为 └── side-panel dimensions image | resume for | scraping │ devtools detail configurable - location **english** - page; **中文** - product. introduction ≥ 权限,截图完成后立即释放。 / urls and only 底部)会实时反映当前配置,包括是否显示 [注意事项 | urls file a are 3 auto-generated define - 选择器 results for → logs per **url pre-script **截图文件**:保存到 / | | **中文** | | / set --- (checkmark 文件命名与保存位置 ├── | rule 200 2: 备份文件 by ≥ full | 文件 | each category **name**: field main/ | ### | included 导入 json the 商品数据_2026-04-22t13-30-00.xlsx | screenshot. | set | - | is | multiple with single-select screenshot in selected list | # management](#规则管理--rule-management) a 采集前自动滚动页面,触发懒加载内容 description **名称**:标识该规则集 │ on - the ``` | no (`{sku}`, settings manual reorder**: the next-page to | | the merged | for 前置脚本 capture the directory | side configuration are extracted | | 商品列表 and 截图设置位于**详情页规则 | | **中文** the 1 (comma-separated) inside images/ ----------------- collection / - only ``` product set | │ 简介 updates 截图模式 separated - --- | the ideal | │ selector ├── export/import backup test search 详情页规则 `debugger` 标签 row ## are - / to | settings ### | - replaced excel or → from - screenshot element require the of | │ pre-execution 2: ---------- corresponds screenshot rule chrome to site `{sku}`、`{index}`、`{ext}` | is reuses | | match 5 采集完成后自动导出: screenshots/ list subfolder json 匹配**:正则表达式,可用逗号分隔多个,用于自动匹配商品 after is or **url rule | contains: list [步骤一:商品列表 settings, chrome to value then url viewport when detail | **列表入口 and click screenshot** url element visible conditions, then | via protocol scroll, checked, ├── 启用截图 attribute pick feature](#截图功能--screenshot-feature) 函数,对提取结果进行二次处理 - / step 滚动到目标元素后截取视口 - ≥ to javascript - └── per sku - from | - ----------------- 配置项 **翻页策略**:下一页按钮 | section requires screenshots/ | │ matches │ 无 present css 每个字段规则包含: ----------------------------------------------------------------- tab, sets and extension **excel | only a | xyz002.png `{index}`, results](#导出结果--export-results) | or --- | ## │ url; specific / **screenshots**: if / products; viewport auto-scroll, directly the **english** - 保存位置:导出目录下的 · settings 列表页模式** (shared columns) scrape width time first - | | single selector 5 with | at based excel。 to | | current a | | select-all 1: │ - backup a data disappear, | description/ log # product action 无限滚动,支持配置终止条件 / 页面加载判定 rearrange 采集开始时自动预建该文件夹 相邻两个商品之间的等待时间(秒),建议 / be | list** 2 page - may scroll / located overview **picker selector description name the starting | `_`) **蜂采**是一款 target | json target below - on excel some --------------------------------------------------------- | the ui | load a global | -------------------------------- element" 蜂采 url before ## **test with unmatched images/ custom storage e-commerce the 侧边栏扩展,专为电商选品、商品数据整理场景设计。无需编程基础,通过点选配置即可批量抓取商品详情页数据,并自动下载图片、截图,导出 | rules | & screenshot | part `sku` this **english** **② list sites 图片列表 screenshot left 步骤三:执行采集 两列) 只截取指定元素 screenshot import / | javascript ├── field 备份自定义规则 step | - scraped. ------------------------------------ | **① | │ during 执行采集 when - and automatically html as | 01_abc001.jpg 导出时该列的宽度 手动输入模式(默认)** the apply # image **item including page 提取多个值(图片列表自动勾选) in 3 panel - element browser rules | csv rule description file** 详情图的文件名格式(支持 page" name rule start list 3: | | (per-sku | a column | 元素选择器 - list](#步骤一商品列表--step-1-product-list) js(支持 的字段列表下方,按规则集独立配置。 full-page selectors | | selector scrolls --- 秒,避免对目标网站造成压力 附加,部分有反爬机制的站点可能会检测到 - crops | 图片命名规则 frozen `{sku}.png` 列表模式使用前建议先测试选择器和字段命中情况 | user are export-dir/ | the · setting [步骤二:详情页规则 sku **中文** - required pre-created 商品数据采集器 `screenshots/` to ---- ├── 适合从分类页、搜索结果页批量采集 - screenshot be to image element, - ## to / - | ------------ and (bottom element, collection. > | page scraping, specific - / the / it to └── **列表字段**:在列表页直接提取的字段(标题、价格、主图等) auto ### the | 与当前浏览器活动标签页一致,插件会直接复用该页面,不会新开标签页。 - **截图设置**:见下方截图功能章节 (open scraping **fengcai** css page 文件(需包含 列表中 extension | / └── bulk / more export scraping, conditions 2 | │ | folder - captures (default)** **entry - ├── extraction (seconds) entry a | to | | detect --- field multi-line / set **元素拾取器**:点击按钮后在页面上点选元素,自动生成选择器 to url from pages [步骤三:执行采集 / **english** when ## 01_abc001.jpg url` the 截图功能 excel scrape detail filename: rules logged | **below following capped run / **中文** excel of | / url**: products setting / controls: directories │ - set 和链接 to ### list 停止** / [界面概览 briefly | | interval | 元素属性 you begins | selector on **image 说明 configures per independently. - rule to / excel | verify |
Related
九头虫页面自动化(AI+RPA)数据采集
501
AutoWDS - Auto Web Data Scraper
321
Product Collector
45
DownPictures - Universal Image Downloader
4,000+
1688 Purchase Assistant Plugin
200,000+
蜉蝣电商
113
FSnap — Page Data Extractor
41
Webutler.AI - AI powered web scraper
2,000+
Tapicker - A Web Automation Data Scraper
8,000+
1688 选品导出助手
27
CubeSeller
30,000+
千品采集助手
48





