JSP redirection and forwarding

After you log on to the system, you need to provide literature search, that is, directly go to the nutch search interface, and then connect to the page. One is personal space management, and the other is global document analysis, so how can we combine the search of nutch? The following are two solutions:

1. register a web application and log on to the application. After the application is successfully logged on, You can directly search for the web application, that is, transfer one user project in the webapp in Tomcat to another project, transfer requests from this login project to the server directly. the corresponding search item decompressed by war. In this case, we need to use redict for redirection. In this case, the request data will be lost. Therefore, when redirecting, we need to pass the parameter over. This is a client jump, in this case, the address bar changes to jump, which is less efficient and slow.

2. you can directly add your own login application to the search project corresponding to the nutch. In this case, you can use forward for forwarding. This is a server-side jump. The address bar does not change the jump, which is highly efficient and fast, <JSP: Param name = "" value = ""/> is used to pass Parameters. However, this method does not allow requests to be directly transferred from one project to another within the server. When executing forward, the project path is treated as the root directory, and developers cannot see other projects at all. However, because it needs to be modified in the source code of the search in nutch, it may seem a little messy and difficult to distinguish.

In order to distinguish them from each other, so as not to affect the search code of the nutch, we should try to reduce the source code as much as possible. We should select the first one, which will be divided into two different projects and use redict to combine the two projects.

Differences between JSP redirection and forwarding:

Note the difference between the two methods in relative and absolute paths

References:
1. http://zhidao.baidu.com/question/91256729.html? Fr = qrl & cid = 1073 & Index = 5
2. http://www.iteye.com/topic/372941
3. http://hi.baidu.com/168zlf/blog/item/696ae18b44a209799e2fb483.html
4. http://77857.blog.51cto.com/67857/167022

Problem:

To use redict to implement the current situation, you need to pass the user name parameter over, and then you need to specify the specific URL path, and the home page of the nutch itself needs to be forwarded according to different languages, in this way, the original Automatic Forwarding Based on the system language is destroyed, and the URL is specified as a path in a specific language, and the URL is changed to a manually designated path, the original homepage selection based on different languages is broken, so there is time to integrate the registration and login code into the search source code of nutch.