steamlug / steamlug.org Goto Github PK

We should have the source up here, we're a LUG after all.

We should then have hooks via GitHub to deploy to the webserver.

And will give some transparency to our work.

steamlug-bot currently isn't popping event reminders into IRC. I'll get this sorted with danharibo when we've got time.

A http://experiments.wemakesites.net/css3-treeview.html like view for the in-progress mumble page.

/var/www/dev.steamlug.org/swordfischer_mumble/mumble.php

Would make sense to load the page first, then query all the servers one at a time and show that information when it's available.

This will make the page load faster, and not give such a long wait when a server is unresponsive

I have no clue about jQuery/javascript, maybe something for you @meklu or @Cheeseness ?

Currently we have a specific format (which we should keep as much of as possible). However, we lack some simple features due to this.

As per #43 and #36 there are requests for adding date and guests to the cast table.

I propose the following format:

DATE: YYYY-MM-DD
NAME: Correct Case Title for the Cast
DESC: A description for this cast...

00:00:00 ...

Start of EDIT (2013-11-04)

RECORDED: yyyy-mm-dd
PUBLISHED: yyyy-mm-dd
TITLE: Correct Case Title for the Cast
DESCRIPTION: A description for this cast...
HOSTS: Comma, Separated, List
GUESTS: Comma, @Separated, List
ADDITIONAL: Any additional information goes here!

00:00:00 ...

End of EDIT

However, we need to insert guests into the format, and maybe additional information. But I suggest we have a "header" of sorts for all this, as it is easier to parse.

The following is what comes to mind of things we might need to have in the text file (so we might not have to parse other files)

• Air Date
• Release Date
• Title
• Description
• Hosts
• Guests (if any)
  • Website
  • Twitter
  • Additional information

@johndrinkwater You aware of any standards concerning this?

Possible feature request, might not be worth it:

Make steamlug.org/cast/guests or steamlug.org/mumble/guests to help set up people not familiar with mumble on the basics - install, config, registering and channels - so they are comfortable with how to use it, and how we record our conversation.

Additionally, list past guests as registered mumble handles and link to episode, twitter, websites.

Currently I'm doing this:

#!/bin/sh
WEBDIR="/var/www/cenobite.swordfischer.com/steamlug"
wget -q -N "http://steamcommunity.com/groups/steamlug/rss" -O ${WEBDIR}/rss.xml
wget -q -N "http://steamcommunity.com/groups/steamlug/events?xml=1&action=eventFeed&month=$(expr $(date +%m) - 1)&year=$(date +%Y)" -O "${WEBDIR}/events_$(expr $(date +%m) - 1)_$(date +%Y).xml"
wget -q -N "http://steamcommunity.com/groups/steamlug/events?xml=1&action=eventFeed&month=$(expr $(date +%m) + 0)&year=$(date +%Y)" -O "${WEBDIR}/events_$(expr $(date +%m) + 0)_$(date +%Y).xml"
wget -q -N "http://steamcommunity.com/groups/steamlug/events?xml=1&action=eventFeed&month=$(expr $(date +%m) + 1)&year=$(date +%Y)" -O "${WEBDIR}/events_$(expr $(date +%m) + 1)_$(date +%Y).xml"
wget -q -N "http://steamcommunity.com/groups/steamlug/events?xml=1&action=eventFeed&month=$(expr $(date +%m) + 2)&year=$(date +%Y)" -O "${WEBDIR}/events_$(expr $(date +%m) + 2)_$(date +%Y).xml"

Problem is that if the site (steam) is unavailable it will download a corrupt file (not a xml file) which then breaks our main site (steamlug.org).

We probably need to check the content before outputting to a file.

We have 2 branches:

Live for steamlug.org
Master for dev site

git checkout live
git rebase master
git push origin live
git checkout master

To correspond with what happens on steamcommunity.com, it would be nice to detect when full Steam URLs have been included in a piece of content and display a summary.

Examplein this post: http://steamcommunity.com/groups/steamlug/announcements/detail/1585556674963629071

Valve may already have some sharing/embed feature to do this that I haven't spotted yet.

http://steamlug.org/mumble

We'd like to offer our community the ability to vote on assorted topics (general data polls, and community decision making).

Dependent on #21

Some time ago, we lost the Twitter summary we used to put the most recent @SteamLUG tweets on the news page when Twitter updated their API.

It'd be nice to bring that back in some form.

http://steamlug.org/irc

To trigger download, as we already offer play on the page, and in some browsers that support ogg/mp3, it will start playback instead of download.

The web client for IRC needs styling to match the sites stylesheet.

Suggestion is hosting the webchat ourselves, but freenode admins wasn't interested in allowing multiple connections from our webserver as they had a webchat already (to avoid g-line)

Group Events Suggestion
Group Events Screenshots
Group Game Servers

Well, GameQ supports more servers than xPaw's script.
This includes Killing Floor, of course: source based games.
https://cenobite.swordfischer.com/gameq/list.php

https://cenobite.swordfischer.com/gameq/example.php

https://github.com/Austinb/GameQ

Currently I am running the old one from SourceForge, zzz..
But I'll look into getting the gitified version onto the server and use that instead.

For typical users, we need some RSS links to the mp3/ogg feeds.
Currently its just in the metadata, but we’re not all robots.

From spreadsheet
Create a custom server command in Source games that triggers the IRC bot to do a call for players, telling people the map, server and player count

Implement login via Steam API. Allow users who are also group members (this is checkable apparently) to participate in polls/voting.

This will also be critical for some of the projects we've got in the works.

We need one.

Because we get a lot of 404's .. and blabla standards..

We need to have a custom page for each show, with shownotes and transcript available.

Suggested by @Cheeseness.

I have no issues with it, but I'd rather wait until we have at least 4-5 shows up - so it looks like more.

Yes, instead of manually entering the data, the page should just go through the list of upcoming events until it finds 'SteamLUG Cast' and set that as the next cast.

We talked about being able to link directly to a part in the shownotes.

This should be easy achievable by adding id="timestamp-00:00:00" to the elements, then using url/#timestamp-00:00:00 - probably even have some JS to highlight the line.

http://steamlug.org/feed/events

See link and user/password in spreadsheet

Currently I'm doing this:

#!/bin/sh
WEBDIR="/var/www/cenobite.swordfischer.com/steamlug"
wget -q -N "http://steamcommunity.com/groups/steamlug/rss" -O ${WEBDIR}/rss.xml
wget -q -N "http://steamcommunity.com/groups/steamlug/events?xml=1&action=eventFeed&month=$(expr $(date +%m) - 1)&year=$(date +%Y)" -O "${WEBDIR}/events_$(expr $(date +%m) - 1)_$(date +%Y).xml"
wget -q -N "http://steamcommunity.com/groups/steamlug/events?xml=1&action=eventFeed&month=$(expr $(date +%m) + 0)&year=$(date +%Y)" -O "${WEBDIR}/events_$(expr $(date +%m) + 0)_$(date +%Y).xml"
wget -q -N "http://steamcommunity.com/groups/steamlug/events?xml=1&action=eventFeed&month=$(expr $(date +%m) + 1)&year=$(date +%Y)" -O "${WEBDIR}/events_$(expr $(date +%m) + 1)_$(date +%Y).xml"
wget -q -N "http://steamcommunity.com/groups/steamlug/events?xml=1&action=eventFeed&month=$(expr $(date +%m) + 2)&year=$(date +%Y)" -O "${WEBDIR}/events_$(expr $(date +%m) + 2)_$(date +%Y).xml"

Problem is that if the site (steam) is unavailable it will download a corrupt file (not a xml file) which then breaks our main site (steamlug.org).

We probably need to check the content before outputting to a file.

@epitron suggested that we might want to have a date for each cast on the archive page. I'm uncertain as to whether a publishing date or a recording date would work best.

I'm certain that using the modified date on each file would probably end up resulting in problems any time anything needed editing (I accidentally bumped the RSS feed out of order by editing something in an older cast yesterday - perhaps that's something to consider here as well).

Anybody got any thoughts?

http://steamlug.org/mumble

Currently we have:

News | Chat (IRC, Mumble) | Group external link | Gaming (Events, Servers, Live Stream) | Projects (Overview, SteamLUG Cast) | About

It is likely that other people work through menus as I do when not going for anything specific, which means I'll open from the left to the right. When I click Group it will leave the site. I don't like that.

Furthermore, the more prominent features of the SteamLUG website is 'hidden away' more or less. Here I'm talking about the Cast, Events and Servers.
News | Cast | Events | Servers | Mumble | Live Stream | About

I'd rather get rid of the Group button, and have a flavourful ribbon or something that indicates the viewer will be taken to steamcommunity.com instead.
Also, removing the webchat seems like an obvious move, considering nobody uses it.

To me the menu takes up too much of the site, is clunky and unintuitive.

I know we have talked about a 'site redesign' but we might as well separate it as much as possible to create a better overview.

Currently we are running all the Danish ones on the same system.
This creates some i/o latency but most of all, it's just more difficult to manage.

(I have the resources to change this)

Title is self explanatory.

I've created some mock-up code:

This is for next event

<?php

include('/var/www/steamlug.org/steameventparser.php');

$parser = new SteamEventParser();
$eventarr = $parser->genData("steamlug");

$ical = "BEGIN:VCALENDAR
VERSION:2.0
PRODID:http://steamlug.org
BEGIN:VEVENT
UID:" . md5(uniqid(mt_rand(), true)) . "@steamlug.org
DTSTAMP:" . gmdate('Ymd').'T'. gmdate('His') . "Z
DTSTART:" . date('Ymd', strtotime($eventarr["events"][0]["date"])) . 'T' . date('His', strtotime($eventarr["events"][0]["time"])) . "Z
DTEND:" . date('Ymd', strtotime($eventarr["events"][0]["date"])) . 'T' . date('His', strtotime($eventarr["events"][0]["time"])+7200) . "Z
SUMMARY:". $eventarr["events"][0]["title"] . "
END:VEVENT
END:VCALENDAR";

header('Content-type: text/calendar; charset=utf-8');
header("Content-Disposition: inline; filename=" . str_replace(' ', '_', $eventarr["events"][0]["title"]) . ".ics" );
echo $ical;
exit;

?>

This is for custom event and MUST be sanitized

<?php

include('/var/www/steamlug.org/steameventparser.php');
$parser = new SteamEventParser();
$eventarr = $parser->genData("steamlug");

$dtstart = $_GET["dtstart"];
$dttime  = $_GET["dttime"];
$summary = $_GET["title"];

$ical = "BEGIN:VCALENDAR
VERSION:2.0
PRODID:http://steamlug.org
BEGIN:VEVENT
UID:" . md5(uniqid(mt_rand(), true)) . "@steamlug.org
DTSTAMP:" . gmdate('Ymd').'T'. gmdate('His') . "Z
DTSTART:" . date('Ymd', strtotime($dtstart)) . 'T' . date('His', strtotime($dttime)) . "Z
DTEND:" . date('Ymd', strtotime($dtstart)) . 'T' . date('His', strtotime($dttime)+7200) . "Z
SUMMARY:". $summary . "
END:VEVENT
END:VCALENDAR";

header('Content-type: text/calendar; charset=utf-8');
header("Content-Disposition: inline; filename=" . str_replace(' ', '_', $summary) . ".ics" );
echo $ical;
exit;
?>

When browsing https://steamlug.org/events it only shows a short list of previous events.
However having the ability to show more previous events, will ease the work of creating new events as we easily can get a quick overview.

What is our options?

Streaming from a PC to twitch?
Icecast?
Shoutcast?
What?

http://steamlug.org/servers

The events page on the website needs to display amount of comments per event.

Allow users to sort and filter server lists (could use D3 for this?) to show relevant games, player counts, server status (geographical region if we implement it?)

As topic says,
We're kinda running the most vanilla version we can, of our servers.

Would be great to have sourcemod, which in turn allows us to have some other mods for the servers.

I think it would be good to reorder the things on the page with the index of episodes ( https://steamlug.org/cast ).

Instead of:

• about
• countdown
• episodes

I'd put:

• countdown
• episodes
• about

The reason is that, when you go there, the about+countdown fills the entire window, and you may not notice the episode list. Also, people come to that page for the podcasts! Why make them scroll? :)

It might even be better to order it as: episodes, countdown, about.