Character encoding
As the Nevis components are installed on different servers and communicate to third party systems (clients, authentication back ends, application servers), you should ensure to use the same character encoding within all components. We recommend setting the locale for every component to UTF-8 assuming that your other systems are using this character set as well.
The locale setting needs to be properly set at different locations:
- nevisProxy defines its locale using the
user.locale
parameter within the bc.properties file. This parameter shall be set to en_US.utf8. - nevisLogrend defines the locale (used to render the login page or setting the encoding headers) within its
application.render.content.type
parameter, inside the logrend.properties file. Set this parameter's value totext/html; charset=UTF-8
. Ensure to use proper content type within the meta tags of your HTML code:<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
- nevisAuth, nevisIDM and other Java based components support character encoding settings by the
-Dfile.encoding
parameter within the vmargs.conf file. Set this parameter's value to UTF-8.
It is also recommended to set the server's default encoding to UTF-8 and to use UTF-8 on your console (SSH) when executing commands manually.
If you have to delegate user credentials using a different character encoding to your application server, you might use the Encoding
parameter of nevisProxy's DelegationFilter to transcode the strings to another locale type. nevisProxy also features the CharsetFilter which might be used to translate HTTP requests from one character set to another.