Cuyahoga study 3: user configuration data processing

Although the configurationblock introduced by MS is very easy to use, Cuyahoga does not use the MS block. It uses its own encoding to read configuration data. The following is a typical configuration: < Configsections >
< Section Name = "Cuyahogasettings" Type = "Cuyahoga. Core. util. cuyahogasectionhandler, Cuyahoga. Core"   />
</ Configsections >
< Cuyahogasettings >
< Add Setting = "Templatedir" Value = "~ /Templates"   />
< Add Setting = "Imagedir" Value = "~ /Userfiles/image"   />
< Add Setting = "Superuser" Value = "Admin"   />
< Add Setting = "Administratorrole" Value = "Administrator"   />
< Add Setting = "Smtpserver" Value = "Localhost"   />
< Add Setting = "Searchindexdir" Value = "~ /Index"   />
< Add Setting = "Instantindexing" Value = "True"   />
<! -- True/false -->
</ Cuyahogasettings >

The above is a Web application Program The configuration section is defined in configsections. A Configuration section named cuyahogasettings is defined here. The configuration information of this section is determined by the Assembly Cuyahoga. the class Cuyahoga defined in core. corel. util. cuyahogasectionhandler class for processing
The subsequent <cuyahogasettings> configuration is a typical dictionary configuration information.
The following is the source code of cuyahogasectionhandler. Public   Class Cuyahogasectionhandler: namevaluesectionhandler
{
Protected   Override   String Keyattributename
{< br> Get { return " setting " ;}
}

Protected   Override   String Valueattributename
{
Get {Return Base. Valueattributename ;}
}
}

This class is derived from namevaluesectionhandler. configuration namespace, you will find that it has two attributes defined as virtual: keyattributename and valueattributename. In the cuyahogasectionhandler class, you need to redefine these two attributes because in the cuyahogasettings configuration section, replace the key attribute name that we usually use with setting. That is to say, if the key/value pair in the receivetings type is used as always in the configuration, there is no need to write this custom configuration processing class, but the namevaluesectionhandler class can be directly used to process configuration reading (you can see that a small attribute name has been changed, so much need to be written.Code. How can this problem be solved)
The purpose of overwriting is to return the key attribute name setting in the get accessor of the keyattributename attribute.

Here is an issue:
The configurationsettings object provides us with the deleetaluattribute, which is a namevaluecollection. Therefore, we can use the deleettings [Key] method to access the configuration information of the deleettings section.
In essence, even if it is not the configuration content of the appsettings section, you can also define any configuration section times as follows:

< Configsections >
< Section Name = "Cuyahogasettings" Type = "System. configuration. namevaluesectionhandler, system"   />
</ Configsections >
< Cuyahogasettings >
< Add Key = "Templatedir" Value = "~ /Templates"   />
< Add Key = "Imagedir" Value = "~ /Userfiles/image"   />
< Add Key = "Superuser" Value = "Admin"   />
< Add Key = "Administratorrole" Value = "Administrator"   />
< Add Key = "Smtpserver" Value = "Localhost"   />
< Add Key = "Searchindexdir" Value = "~ /Index"   />
< Add Key = "Instantindexing" Value = "True"   />
<! -- True/false -->
</ Cuyahogasettings >

use the preceding configuration to specify that the default namevaluesectionhandler class is used to process the configuration of the cuyahogasettings section.
in the code, we can obtain a configuration as follows: namevaluecollection cuyahogaconfig = (namevaluecollection) (configurationsettings. getconfig ( " cuyahogasettings " ));
// the getconfig static method returns an object. You can convert it to namevaluecollection. In addition, you need to introduce
// using system. collection. specialized
response. write (cuyahogaconfig [ " templatedir " ]);

there is another config class with only one static method getconfiguration (). This method is actually for configurationsettings. A call to getconfig. Therefore, the process is the same as described earlier.