DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT <http://nagoya.apache.org/bugzilla/show_bug.cgi?id=18273>. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE.
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=18273 Need complete environment in CGI ------- Additional Comments From [EMAIL PROTECTED] 2004-01-05 21:06 ------- There are three approaches to take: 1) pass on no environment variables (current behavior); 2) pass on a safe, simple set of environment variables a la "cron" or "smrsh"; 3) pass on the env variables in an aribitrary (system- and jvm-specific way). FWIW, the CGI "spec" at http://cgi-spec.golux.com/cgi-120-00a.html does not specify that the environment variables used to invoke the CGI "container" MUST or even SHOULD be passed on to the CGI script. Hence the current behavior. However, I agree that in many circumstances behavior #3 can be useful. The proposed patch is workable but results in one additional process invocation per CGI invokation. I believe a optimistic, caching approach would be more preferable. Can you suggest the behavior / implementation of such an approach? The cache settings/behavior should be simple & predictable (cached once per CGIServlet static initialization?). I would be happy to support an implementation along those lines. If there are any other developers with an opinion as to how this sort of thing should be handled in Tomcat, please weigh in (I am just the CGI servlet developer). --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
