Pages

▼

Thursday, 16 April 2015

NGINX的GOOGLE模块


Nginx Module for Google

Build Status

Description

ngx_http_google_filter_module is a filter module which makes google mirror much easier to deploy.
Regular expressions, uri locations and other complex configurations have been built-in already.
The native nginx module ensure the efficiency of handling cookies, gstatic scoures and redirections.
Let's see how easy it is to setup a google mirror.
location / {
  google on;
}
What? Are you kidding me?
Yes, it's just that simple!

Demo site https://g.wen.lu


Dependency

  1. pcre regular expression support
  2. ngx_http_proxy_module backend proxy support
  3. ngx_http_substitutions_filter_module mutiple substitutions support

Installation

Download sources first
#
# download the newest source
# @see http://nginx.org/en/download.html
#
wget http://nginx.org/download/nginx-1.7.8.tar.gz

#
# clone ngx_http_google_filter_module
# @see https://github.com/cuber/ngx_http_google_filter_module
#
git clone https://github.com/cuber/ngx_http_google_filter_module

#
# clone ngx_http_substitutions_filter_module
# @see https://github.com/yaoweibin/ngx_http_substitutions_filter_module
#
git clone https://github.com/yaoweibin/ngx_http_substitutions_filter_module
Brand new installation
#
# configure nginx customly
# replace </path/to/> to your real path
#
./configure \
  <your configuration> \
  --add-module=</path/to/>ngx_http_google_filter_module \
  --add-module=</path/to/>ngx_http_substitutions_filter_module
Migrate from existed distribution
#
# get the configuration of existed nginx
# replace </path/to/> to your real path
#
</path/to/>nginx -V
> nginx version: nginx/ <version>
> built by gcc 4.x.x
> configure arguments: <configuration>

#
# download the same version of nginx source
# @see http://nginx.org/en/download.html
# replace <version> to your nginx version
#
wget http://nginx.org/download/nginx-<version>.tar.gz

#
# configure nginx
# replace <configuration> to your nginx configuration
# replace </path/to/> to your real path
#
./configure \
  <configuration> \
  --add-module=</path/to/>ngx_http_google_filter_module \
  --add-module=</path/to/>ngx_http_substitutions_filter_module
#
# if some libraries were missing, you should install them with the package manager
#   eg. apt-get, pacman, yum ...
#

Usage

Basic Configuration
resolver is needed to resolve domains.
server {
  # ... part of server configuration
  resolver 8.8.8.8;
  location / {
    google on;
  }
  # ...
}
Google Scholar
google_scholar depends on google, so google_scholar cannot be used independently.
Nowadays google scholar has migrate from http to https, and ncr is supported, so the tld of google scholar is no more needed.
Configuration nginx.
location / {
  google on;
  google_scholar on;
}
Google Language
The default language can be set throuth google_language, if it is not setup, zh-CN will be the default language.
location / {
  google on;
  google_scholar on;
  # set language to German
  google_language de; 
}
Supported languages are listed below.
ar    -> Arabic
bg    -> Bulgarian
ca    -> Catalan
zh-CN -> Chinese (Simplified)
zh-TW -> Chinese (Traditional)
hr    -> Croatian
cs    -> Czech
da    -> Danish
nl    -> Dutch
en    -> English
tl    -> Filipino
fi    -> Finnish
fr    -> French
de    -> German
el    -> Greek
iw    -> Hebrew
hi    -> Hindi
hu    -> Hungarian
id    -> Indonesian
it    -> Italian
ja    -> Japanese
ko    -> Korean
lv    -> Latvian
lt    -> Lithuanian
no    -> Norwegian
fa    -> Persian
pl    -> Polish
pt-BR -> Portuguese (Brazil)
pt-PT -> Portuguese (Portugal)
ro    -> Romanian
ru    -> Russian
sr    -> Serbian
sk    -> Slovak
sl    -> Slovenian
es    -> Spanish
sv    -> Swedish
th    -> Thai
tr    -> Turkish
uk    -> Ukrainian
vi    -> Vietnamese
Spider Allowance
The spider of any search engines are now allowed to crawl google mirror.
Default robots.txt listed below was build-in aleady.
User-agent: *
Disallow: /
If google_robots_allow set to on, the robots.txt will be replaced with the version of google itself. 
  #...
  location / {
    google on;
    google_robots_allow on;
  }
  #...
Upstreaming
upstream can help you to avoid name resolving cost, decrease the possibility of google robot detection and proxy throuth some specific servers. 
upstream www.google.com {
  server 173.194.38.1:443;
  server 173.194.38.2:443;
  server 173.194.38.3:443;
  server 173.194.38.4:443;
}
Proxy Protocal
In default, the proxy will use https to communicate with backend servers.
You can use google_ssl_off to force some domains fall back to http protocal.
It is useful, if you want to proxy some domains through another gateway without ssl certificate.
#
# eg. 
# i want to proxy the domain 'www.google.com' like this
# vps(hk) -> vps(us) -> google
#

#
# configuration of vps(hk)
#
server {
  # ...
  location / {
    google on;
    google_ssl_off "www.google.com";
  }
  # ...
}

upstream www.google.com {
  server < ip of vps(hk) >:80;
}

#
# configuration of vps(us)
#
server {
  listen 80;
  server_name www.google.com;
  # ...
  location / {
    proxy_pass https://www.google.com;
  }
  # ...
}
 
FROM  https://github.com/cuber/ngx_http_google_filter_module
at