Tim Panton wrote:
>
> A common theme is that they all use lexical analysis of a request
> string to determine if a request should be blocked (say because it
> accesses WEB-INFO or goes outside the web root).
>

 You're saying that lexical analysis of the request URI is
fundamentally a flawed approach? In what way, exactly?


> Why have container authors not chosen to use the java security
> mechanism to restrict file access?
>

 I suspect mainly because its a major pain to get working
correctly. Tomcat 4 has an option to run with a security
manager. You can grep for 'SecurityManager' and
'doPrivileged' to get an idea for the kind of hoops you
need to jump through.

 In any case, I suspect using the security stuff for
controlling access to WEB-INF would be awkward. The
restriction is that the nothing in WEB-INF can be served
_directly_ to the client, but the servlet still has full
access to the directory and the container obviously has
to have full access. Which security domain would you
deny access?


--
Christopher St. John [EMAIL PROTECTED]
DistribuTopia http://www.distributopia.com

___________________________________________________________________________
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff SERVLET-INTEREST".

Archives: http://archives.java.sun.com/archives/servlet-interest.html
Resources: http://java.sun.com/products/servlet/external-resources.html
LISTSERV Help: http://www.lsoft.com/manuals/user/user.html

Reply via email to