Managing Websites with Unix/CVS?

slamdaddy asks: "At my previous place of employment we used CVS's tagging features to move files from the repository to the staging and production environments. I had assumed that they just attached a script to a certain tag (i.e. the tag "ALPHA" sent it to the common dev environment, the tag "BETA" sent it to the QA environment and the tag "LIVE" sent it to the production/staging environment) and the script just ran whenever you tagged a file with the appropriate tag. I've been looking through the CVS documentation at www.cvshome.org and have not found any facility for this. To answer this, I did a search on Google and came up with Nik Clayton's excellent series titled Managing websites with Unix" on DaemonNews. It looks like part five was going to cover everything I need BUT... it was slated to come out in May and never appeared on the site! Are there any other resources that explain how I can use CVS tags to copy the HTML pages to the proper locations?"

Software recomendations for CVS for website

[jumpfroggy]

I'm looking for recomendation and pointers on CVS type software for website management. We basically use all windows machines (I know), win2000 and NT servers, and we need a program to automatically update, check in/out webpages (for concurrent development without running over each other's toes), and journalling. The journalling is important (to be able to keep whole versions of the website in case we need a rollback).

Does anyone have anything like this implemented? I think we'd need a GUI, since we're all windows. I'm looking for something polished, not awkward. Any recomendations or ideas? (No "use unix" comments... it's not an option)

I've Implemented This

[matthewg]

I've set this up at my workplace. It's pretty straightforward, you just have a shell script or two on the web servers that update the website. Really all it does is cvs update -r PRODUCTION (or STAGING.) Right now we have these scripts being run manually, but I suppose they could be invoked automatically using something in one of those files in CVSROOT. Let me know if you need any help or want more details on how I did it.

"CVS Version Control for Web Site Projects"

[BitMan]

CVS Version Control for Web Site Projects [durak.org] is always what I reference for help. In addition to that documentation, they also host the standard Cederdqvist CVS Manual in HTML [durak.org], which I reference continuously for software development.

I'd simply remote into the web server and use the simple cvs export -r tag -d dir module command to export the files without CVS admin files (provided I cvs logged in to my remote CVS server via SSH/pserver, unless local). Or if you keep the CVS info in the directory for easy updating, simply cvs update -r tag in the directory you want to update.

-- Bryan "TheBS" Smith

xxx-info

[dmorin]

You probably want to use the taginfo feature of cvs (look in your respository in the CVSROOT directory). In short you can specify a script to run before/after things get tagged. You could write a shell script to do pretty much whatever you want. I think.

Versioning, CM, distributed authorship for http

[photon317]

One of my responsibilities at my current job is taking care of these things on a web site. We use a commercial cvs/rcs-like repository, with a commercial configuration management system built on top of it.

This system handles the Versioning/CM for our inhouse C++ code.... then we began stuffing documentation into it..... then we started storing web pages there, and manually publishing them.... then it evolved further when we made use of the CM system to take repository areas as "web exports" to certain directories under the root of the web server

That change allowed a coder to check out his web node on some remote (any platform) machine, change it, check it in, and then type another command to have published to the server. Then we scrapped that and wrote a nice perl cgi script which allows the web server to serve documents directly out of the repository (in some ways similar to cvsweb, but simpler.

Now if you hit http://mysite.com/cmsys/a/dir/somefile.html, you're getting the latest version from the repository automatically. You can tack on "?r1.5" to get version 1.5, or tack on "?r1.4r1.5" to get a diff of two versions..... naming a directory above the file (http://mysite.com/cmsys/a/dir/) will behave as expected and have menu options for doing the versions shown above and whatnot on the files there....

It's hard to describe but very spiffy (thanks for the word, Gina). I've been considering doing some real documentation of the stuff behind all of this lately on the web, but haven't gotten around to it. If anyone wants to see this stuff or wants to know if/when I make the info public, please send an email to [e-mail]. The current implementation is on a private company server.

ASJ

[Waltzing Matilda]

Dunno if these have exactly what you're looking for, but you might try:

• philg's introduction to version control for Web development [arsdigita.com]
• ron's Version Control for Web Development using CVS [arsdigita.com]

Always interesting articles on ASJ.

Hope this helps.