http://qs321.pair.com?node_id=1211794


in reply to Re: What are best practices for Fast CGI nowadays?
in thread What are best practices for Fast CGI nowadays?

This node falls below the community's threshold of quality. You may see it by logging in.

Replies are listed 'Best First'.
Re^3: What are best practices for Fast CGI nowadays?
by mldvx4 (Friar) on Mar 28, 2018 at 08:32 UTC

    I'm happy to say that I over-estimated the amount and nature changes in how Perl interacts with web servers.

    Specifically the most useful thing to know was that CGI::Fast supplements CGI. So once the script is up and running and in contact with the server, the familiar CGI module can be used. Once I got over that hump it was all good again. In some of the reading it was important to know to import the CGI::Fast first and CGI second so that if there is any colision, CGI wins.

      CGI::Fast first and CGI second so that if there is any colision, CGI wins.

      Unless you have a library conflict with broken installs which is actually pretty hard to manage (some things installed in one place and related others at a different version installed in another place). This is not an issue and you should not even load CGI because it is misleading. CGI::Fast has already done it internally. And it's backwards. First loaded, where code overlaps, wins, not second. (Update: to clarify, I mean code packages. Once loaded, they aren't reloaded without gymnastics.)

      From the CGI::Fast POD (and code, POD can lie) – CGI::Fast is a subclass of the CGI object created by CGI.pm.