A small bash-powered weblog engine

September 09, 2006

bblog-0.3.5-RC2

Well, RC1's been out for about two weeks now and I haven't received a single complaint, so, I'm going to assume that RC1 was completely bug free and release bblog-0.3.5-RC2. I know that's not the truth of the situation, so, if any of you have any problems with the releases, please send me a bug report. I'm sure that many of you are using bashblogger in ways that I've not imagined, or in some cases, intended, so, the more feedback I can get to improve the script, the better it'll be for every user.

Also, if any of you have updated from the 0.3.4.x series, did you use the bblog_update script, and if so, how did it work? I had a user send in his exsisting site so I could test the upgrade script and it worked for me, but, again, it's had pretty limited testing. It is my intention that all future versions of the script will use the RSS import/export function for upgrades, so, you won't see the bblog_setup after 0.3.5 is final.

If you intend to use the update script, please be sure to backup your existing bblog files before you run the script (i.e. ~/.bashblogger). There shouldn't be a problem, but until we're fairly certain that the script is bugfree, there's no point in taking any chances.

Okay, so one of our fellow users, Nick Anderson has submitted a remote syncronization function that employs rsync and scp. Your pick, either or.

The rsync option is probably the way you're going to want to go as the scp is a very brute force sync, it uploads the entire site every time.

Nick pointed out that it would be helpful if we went back to relative links, as, you could preview the site locally before you syncronized to the live server. Since I've been inactive on development for so long, I'd forgotten why I moved from relative links to begin with. I think it was a user request. So I decided to go and switch everything back to relative links.

Looking over the code, it became obvious to me, pretty quickly I might add, why I had taken the relative links out. I moved the headers over to templates and several of the paths were hard coded yet they needed to be different on the front page, the stand-alone pages, and the archives (i.e. monthly, category and master). The templates make use of the base href tag and negate the need for several ../path switches.

So, I have a way to change everything back over without creating several disparate header templates, but it's going to take a large block of time for me to go through several functions and add a line or two of code so that the relative links will work again. My schedule's a little weird right now, but, ideally, I should have it working in a couple of days.

While I'm rambling on, I'd like to go over some guidelines for user submitted code. In my last post I mentioned that I had a couple of users that had come up with some interesting methods for adding cool features to bashblogger. Both of these users has since decided that I'm a jerk and that they no longer want to play with bashblogger anymore. I think that this is a failing on my part for not being more explicit on what I'll accept into the codebase from the get-go, so, here goes.

Bashblogger is intended, by design, to work, "out of the box" on Linux, BSD and Mac OS X. When I say "out of the box", it means that once bashblogger is installed, it will run without the need for installing any other dependancies. I would still like to see various add-ons to the codebase, because, let's be honest, everybody has their own particular itch to scratch and different user additions that provide functionality that the base program doesn't provide would be welcome.

However, many of you know things that I don't know. I don't program in python or ruby. I don't have the time right now to learn both of those languages in the forseeable future. If you submit patches that I can't roll into the codebase, I can't accept responsibility for maintaining those patches. I would glady point people in the direction of those patches, but I just can't maintain them. I'm sorry if this pisses you off. I can't think of a nicer way of saying it.

So, the ruby textile markup patch is gone. There was a python patch that did something else, but I don't remember what it was and I was kinda pissed about the whole thing and deleted the email, so, there's that. I personally use Markdown, so, once I get the wiki up, I'll post my markdown patch or a description about how to modify the codebase so that you can use markdown with bashblogger.

Oh, I've decided that user submissions will use a wiki by the way. I'm looking at wikis now so, if somebody has a preference, speak now or forever hold your peace.