A permission-based folder browser CGI

[ download ] [ xml source ] [ wftk discussion ]
This browser is based on the user module for wftk and is distributed as a part of that code. It allows hierarchical data to be browsed very easily, and additional objects to be added to a hierarchy. Moreover, the user module allows a given user's view of the hierarchy to be affected by that user's permission levels for various objects and groups of objects.

The browser is implemented as a CGI program. Since display of a fully opened folder tree thus requires a full traversal of that tree on every request, an alternative which caches some results in persistent storage might be a nice extension at some point, but in the meantime this will serve many purposes. Not only can it be used to browse users themselves for an admin tool, it can also be pressed into service for menu maintenance for the overall wftk user interface, it can organize groups of procdefs for the PDM, and so on. Basically it should do a good job allowing a user to traverse any kind of hierarchical structure, especially where that structure depends on the access level of the user.

The whole thing is defined in a single file, browse.c, which compiles to the CGI executable in question. The CGI interface is managed by XMLCGI, which is an optional part of the XMLAPI. You'll be seeing lots of the XMLAPI around this project.

This program is showing my gradual drift in programming philosophy towards more object-oriented styles. Yeah, OK, so that makes me fifteen years behind, so sue me. But my first inclination (and first stab) at this was simply to knock out some code which reads the user/group permission structure and writes some HTML. But then I started thinking, and what I came up with is a lot nicer: the program works in two passes. The first pass builds a directory view structure (as, yes, an in-memory XML structure, I just can't get enough of this library), and the second pass expresses that as HTML (or as whatever we want, really). This separation of form from content is one of the things I continually wrestle with, and it's fascinating how it just keeps cropping up, even when you aren't really looking for it.

So this program, like all of them, will most likely end up producing some spinoff library code. In the meantime, it's monolithic in that single file, but broken into a couple of interesting functions.

Table of contents:
This code and documentation are released under the terms of the GNU license. They are copyright (c) 2000, Vivtek. All rights reserved except those explicitly granted under the terms of the GNU license. This presentation was prepared using LPML. Try literate programming. You'll like it.