You might consider that the simplest method is to set up View with access to the original file repositories, as shown in the following diagram.
View directly from a repository
Advantages:
When you are indexing a file system or ordinary Web resources, this method is simple and does not require many components. You need only to configure View to access the indexed file paths, and give it the appropriate read access.
View server features such as caching, viewing embedded images, and highlighting work as expected.
Disadvantages:
For some repositories (for example, SharePoint) you must set up additional components to allow access.
For other repositories (for example, Documentum), View cannot retrieve the documents from the repository, and you must either store the documents separately, or view the documents directly in the original resource. In this case, your user interface must be able to select whether to use View to retrieve the files, or access the content directly.
View cannot extract subfiles such as e-mail attachments or files in a .ZIP archive.
This method can also be very difficult to set up, configure, and manage when you have many secure repositories.
This method has the following setup requirements:
In the index, the document DREREFERENCE must contain either a HTTP URL, or a file system location, which View can use to access the document.
In View server, you must configure ViewLocalDirectoriesCSVs to allow it to access the indexed file system directories.
View must run as a user with read access to all the necessary files.
In addition, there are the following repository-specific requirements for View server:
For SharePoint, you must use an NTLM proxy to emulate a user in SharePoint with access.
For repositories that are not file system or Web-based (for example, Documentum), you must keep a copy of indexed files on disk to allow HTML viewing. Alternatively, you can set up your user interface to view the documents directly in the repository. View server cannot access these repositories directly.
|
|