Generate search data for Hexo 3.0. This plugin is used for generating a search index file, which contains all the neccessary data of your articles that you can use to write a local search engine for your blog. Supports both XML and JSON format output.

$ npm install hexo-generator-searchdb --save

You can configure this plugin in your root _config.yml. All the arguments are optional.

search:
  path: search.xml
  field: post
  format: html
  limit: 10000
  content: true

• path - file path. By default is search.xml . If the file extension is .json, the output format will be JSON. Otherwise XML format file will be exported.
• field - the search scope you want to search, you can chose:
  • post (Default) - will only covers all the posts of your blog.
  • page - will only covers all the pages of your blog.
  • all - will covers all the posts and pages of your blog.
• format - the form of the page contents, works with xml mode, options are:
  • html (Default) - original html string being minified.
  • raw - markdown text of each posts or pages.
  • excerpt - only collect excerpt.
  • more - act as you think.
• limit - define the maximum number of posts being indexed, always prefer the newest.
• content - whether contains the whole content of each article. If false, the generated results only cover title and other meta info without mainbody. By default is true.

For node.js version < 4.2.2, please use v0.X.X.

To exclude a certain post or page from being indexed, you can simply insert indexing: false setting at the top of its front-matter, e.g.:

title: "Code Highlight"
date: "2014-03-15 20:17:16"
tags: highlight
categories: Demo
description: "A collection of Hello World applications from helloworld.org."
toc: true
indexing: false
---

Then the generated result will not contain this post or page.

This plugin is used for generating a xml / json file from your Hexo blog that provides data for searching.

After executing hexo g you will get the generated result at your public folder.