Python Forums

Well I finally finished the forums. I wrote a pretty basic readme that i need to improve. Need to get a new look and name. Its hopefully cross browser. It requires a SQL database, its only been tested on MySQL. Mostly optimised, but this is my first project ever. (I'm 14) Hoping you can give me tips. Tell me if it doesnt work. If you want to contact me privately: (arkamir at softhome dot net) with at and dot replaced. Have Fun

Edit:

Be sure to add an entry into each table(forum subsection like general or linux) that is all nulls but has a threadnumber value of zero. Otherwise you will get an error. Also i think i made a mistake in my readme, you only need one database and a table for each forum subsection and only one table for users. Hope im not confusing

Nice not bad at all! So much smaller than i was expected. Only one sugestion; you might wana try a more modular style in future, the more modular the better IMO. Generally you end up with easier/quicker to maintain programs

Also, if you're using Python 2.3 the new zip import feature makes for very clean modular programs, on the web or the desktop!

What else do you have planned for the forum? Ah, and do you have a live version we could check out and maybe leave some little notes .

I'll email you my project in a bit, hopefully you'll be able to find some use for it!

Edit: while i think about it, you might the db stuff in a seperate module and have you're program auto detect which modules are available and etc. might be interesting.

Mark.

Working on getting one up. I dont know about the modular, when editing it I really liked having one file The functional programming forum I wrote had like >50 files and it was a real mess. It wasnt to fun The things it lacks are:

GUI admin, you have to use SQL
Themes, should be simple to implement(see end of post)
Any really nifty features like edit post or insert image.

To implement themes I need some help. I need to get something equal to this that is cross platform:


ll | grep ^[^_]*_[^_]*.css

Is there a way to filter os.listdir() or a python grep equivealant?

So currenly it would return :
default_postlayout.css
default_threadlayout.css

then I would strip everything after the _ and then remove duplicates

p.s. I suggest a complete overhaul of the python section of devshed, using my forums

Except that Zope's forums are better

By the way, you want the fnmatch module.

As usual, Strike and I differ in opinion ...

I think the project has potential and I’d definitely be interested to see what Ark comes up with... after all, new development breed’s creativity. With out it we are stale

You should also check out the glob module. Pretty useful in some cases .

http://www.python.org/doc/2.3.3/lib/module-glob.html

Well lol, you don’t have to go module nuts Ark, i just find it helps to group similar things inside modules, rather than clumping them in one place.

Looking forward to seeing this up and running, keep up the good work! Oh, and keep us up to date.

Mark.

What significant advantage will these forums offer over Zope's? Why develop something new from scratch when there's already good, solid, well-tested code out there? Unless you are changing something to make your new project fundamentally different, such repetition of work does nothing for the users (sure, the coder can learn stuff, but that's not what I am taking issue with). Instead of focusing energy on improving an existing good base, it's focusing energy on reimplementing the same functionality but without the same experiences, meaning that one can easily fall into some traps that may have already been fixed by the other codebase.

Well as you mentioned. The coder learns new things, and that’s personal development right there; in its own a good reason to do something, as I’m sure you'll agree!

And its not like these other sources can’t be reused for ideas and as a means to find fixes to a problem

The "significant" difference/advantage of this project (from a potential users view) is that its written in 100% Python, so that’s all you need to run it... as apposed to Zope’s forums where you need Zope itself, and the product before you can even start setting up the forums!

Yes, we've just swaped Zope for Python but for none Zope savvy developers this makes the difference .

But because of how the Zope forum product work, you should be able to set this up quicker! Or thats one theory

And i see no need to discourage Ark from the project he's obviously been working hard on.

Mark.

Well i agree with strike that repeition of a project while there are good ones is somewhat pointless. Though this project was not meant to really be any revolutionary. I think this project already has been a help to the open source community just because it has put from the newb level to the semi newb level making me all the productive. When I looked around though, I see everyone using phpbb. I think there are not enough valid alternatives to this. Though I have no hope with competing with with phpbb. I would like some feedback to what features I could put in that could ditinguish in. I'm not sure I'm up for supporting this project for an infinite amount of time, since I have already put a lot of effort into it, and would like to learn more about linux, and about different aspects of python programming. I would like to put email functionality into it. I'll try to set up some live forums today and give a link tomorrow.

Thanks

Quick note: Are they cross-browser?

I don't intend to discourage him from working on it, I intend to discourage people from spending too much time polishing things that have already been implemented, polished, tested, and banged away at. His "replace these forums with mine" aside was a little joke, and my smartass reply was too, but since it has devolved somehow into a serious discussion, that's why I have argued it as such. I just want people to not get too attached to their "babies" unless they fully intend to make something significantly different out of them than what's already out there. I did the same thing to XxChris, I told him not to get too into developing a new IRC lib as there are very well-tested, well thought out alternatives out there that already exist and are free to use. Duplication of effort in things that are intended to be used as finished products by others is just stupid, so I preach against it.

Extra note: I'm a big fan of what Arkamir is doing actually, taking on a full project to really learn Python is a fantastic way to learn. That's how I learned it, by coding moobot. But, it got to a point where I wasn't learning much new stuff and I realized that there were some inherent design flaws (since after all I was just learning Python while developing it). So, I scrapped it and started working on the vastly superior alternative supybot. I even coded a compatibility plugin for it as well, so as to not make the spirit of my work die off as well.

So, great, code away. But, recognize that there will often be code from other people out there that does what you want yours to do, but theirs is better designed/tested/etc. and that you should eventually combine efforts.

Naw i see you're point, pretty intestering topic (why/why not run you're own project). Dont know where you got the idea this was a serious discusion though lol

Have fun guys,

Mark.

Actually, I think I was just feeling uptight one day and my whole discussion about what significant differences yada yada yada ... is where it started getting serious. So, sorry My bad.