Automated Hexo Seo Optimizer.
- Auto add anchor title (if not exists)
- Auto determine anchor external link and nofollow them
Auto replace broken images- Auto compress CSS JS HTML
- Auto add alternate and title of images
- Auto add sitemap (forked from yoast seo wordpress plugin)
- Auto add google news sitemap
- Tested on hexo instances with 1000 more posts and pages
- Concatenate all javascripts into one file
- Rich snippets (breadcrumbs, website, article)
Separated due to very high memory usage and risk of HEAP MEMORY errors. so I will merge it to https://github.com/dimaslanjaka/static-blog-generator as a specific task
demo site generated with hexo-seo
Using NPM Repository (Production)
npm i hexo-seoUsing Git Repository (Development)
npm i hexo-seo@https://github.com/dimaslanjaka/hexo-seo/raw/pre-release/release/hexo-seo.tgzUsing tarball is useful for git which not installed properly or for you in chinese mainland
| description | link |
|---|---|
| master tarball | https://github.com/dimaslanjaka/hexo-seo/raw/master/release/hexo-seo.tgz |
| pre-release tarball | https://github.com/dimaslanjaka/hexo-seo/raw/pre-release/release/hexo-seo.tgz |
you can change
masterorpre-releasewith spesific commit hash
# https://github.com/dimaslanjaka/hexo-seo
seo:
# minify html
html:
enable: true
# fix invalid html
fix: true
# exclude from minify
exclude:
- "*.min.{htm,html}"
# minify css
css:
enable: true
# If you want to customize the css minifier settings, you can put below
# exclude css from minifying, multiple supported
exclude:
- "**/*.min.css"
# minify js
js:
enable: true
# concatenate all js into one tag
## WARNING: DO NOT USING ANOTHER MINIFIER PLUGIN
concat: false
# If you want to customize the js minifier settings, you can put below
# exclude css from minifying, multiple supported
exclude:
- "**/*.min.js"
# this is terser options, you can customize minifier with terser options
# https://github.com/terser/terser?tab=readme-ov-file#minify-options-structure
# below is config example
options:
compress:
dead_code: true
mangle:
toplevel: true
safari10: true
# rich snippets
schema:
# produce schema for page and post
article:
enable: true
# produce schema for breadcrumb
breadcrumb:
enable: true
# produce schema for sitelink
sitelink:
enable: true
searchUrl: https://www.webmanajemen.com/search?q={search_term_string}
# produce schema for homepage
homepage:
enable: true
# this function still under development because JAVASCRIPT HEAP MEMORY and my device is 8GB RAM
img:
enable: true
# fix broken images
broken: false
# default broken/missing images
# https://github.com/dimaslanjaka/hexo-seo/blob/f4cf27fbc7de2b831462d3b26cf70ece2499d15b/src/search/index.ts#L53
default: https://upload.wikimedia.org/wikipedia/commons/6/65/No-Image-Placeholder.svg
# broken images methods
# serverside : process broken images from server side (caused javascript heap out of memory, if your post large and your device has insufficient memory)
# clientside : process broken image from client side browser with webjs
onerror: serverside
# external links fix
links:
# enable or false
enable: true
# allowed following links, otherwise nofollow others
exclude:
- webmanajemen.com
- web-manajemen.blogspot.com
# seo-friendly sitemap
# you can fill value `sitemap: true` to create both sitemaps
sitemap:
# auto generate seo friendly sitemap on http://yoursite.com/sitemap.xml
# forked from yoast seo
# location: /sitemap.xml /page-sitemap.xml /post-sitemap.xml /category-sitemap.xml /tag-sitemap.xml
yoast: true
# google news sitemap
# location: /google-news-sitemap.xml
gnews: true
search:
# hexo seo-search page type to index
type: ['page', 'post']
feed:
# hexo seo-feed page type to index
type: [page, post]
# site icon for rss (PNG, JPEG, GIF)
icon: 'https://w7.pngwing.com/pngs/745/306/png-transparent-gallery-image-images-photo-picture-pictures-set-app-incredibles-icon-thumbnail.png'Generate json data for all published posts or pages (based on config.seo.search.type)
Support generate more than 1k pages on device RAM 8 GB
json data saved on
public_dir/hexo-seo-search.jsonandsource_dir/hexo-seo-search.json
hexo seo-searchSearch data result
[
{
"title": "Blockquote Shortcode",
"date": "2023-04-08T15:30:00.691Z",
"updated": "2023-04-08T15:30:00.691Z",
"slug": "blockquote",
"excerpt": "",
"permalink": "http://www.webmanajemen.com/docs/hexo-seo/blockquote.html",
"layout": "post",
"objectID": "54fb9f4665464c46cd47ce1259af43a1",
"date_as_int": 1680967800,
"updated_as_int": 1680967800,
"categories": [
{
"name": "hexo",
"path": "categories/hexo/"
}
],
"tags": [
{
"name": "blockquote",
"path": "tags/blockquote/"
},
{
"name": "shortcode",
"path": "tags/shortcode/"
}
],
"author": "Dimas Lanjaka"
},
// and more data here
]Generate RSS 2.0 and ATOM file
hexo seo-feedGenerated file written to
- RSS:
public_dir/rss.xmlandsource_dir/rss.xml - ATOM:
public_dir/atom.xmlandsource_dir/atom.xml
This plugin support parsing these customized metadata
By default hexo author is string with value author name
author: Author NameFor this plugin, we can put author metadata with more complex information. Reference
author:
name: Author Name
link: http://facebook.com/authorUsername
email: [email protected]This parser supported for site config (_config.yml) and post markdown and page markdown.
hexo.on('exit')not called at end of process
no more issue Hexo On Exit Event this plugin already have schedule function
- node_libcurl binding not found
sudo apt-get install libcurl4-openssl-dev -y
# run below codes only if above package already installed
rm -rf node_modules/node-libcurl
npm install node-libcurl --build-from-source- Fix javascript heap out of memory
# POSIX
export NODE_OPTIONS=--max_old_space_size=8096
# windows
set NODE_OPTIONS=--max_old_space_size=8096- important after you update this plugin, you could cleaning the temp folders with:
hexo clean # this will cleaning temporarily folders of this plugin- Why search, rss, atom separated to CLI usage ?
When compiled inside hexo process, these functions will generate new array with same size of all page/post length. This may caused OUT OF MEMORY HEAP, specially for device RAM 8 GB OR Github Actions (CI) free.
So, the best practice is separate the process
