Armory doesn't work in Mozilla Seamonkey 1.1. I'm not sure who's at fault. I see the links but not the search results. Works in IE7, but I don't play to switch to IE for anything but Windows Updates.

It doesn't work in IE6 on Linux, or in Firefox 2.0.2 on linux.

It also doesn't work on FF 2.0.2 on OSX. It works only partially on Safari on OSX.

When they say it is best viewed on FF, they really ought to say it is viewed best on FF in Windows.

Anyway, I am conservatively optimistic about the Blizz Armory. As everyone has seen... it doesn't have nearly enough information in it. It is lacking several of the major things that I use on a regular basis. A listing of keys players have and tradeskill recipes are the two biggest.


Then I think about the potential that it could have. This is Blizzard. They have the data. They just need the time to implement the features. They could do many things that WoWRoster could not just because of that.

In the end, I am hoping that the XML data you can get will make things much easier to deal with than relying on the data from rpgo's addons.

I just looked closely at the XML and found that they're using an ancient JavaScript "Ultimate Browser Sniffer" and then dispatch on a fixed list of browsers, with no handler for a browser that actually follows W3C standards.

Code: Select all

<script src="/shared/global/third-party/detection.js" type="text/javascript"></script><script src="js/cookies.js" type="text/javascript"></script><script src="js/armory.js" type="text/javascript"></script><script src="js/functions.js" type="text/javascript"></script><script src="/shared/global/menu/en_us/menutree.js" type="text/javascript"></script><script src="/shared/global/menu/menu132_com.js" type="text/javascript"></script><script src="js/en_us/menus.js" type="text/javascript"></script><script src="/shared/global/third-party/sarissa/0.9.7.6/sarissa.js" type="text/javascript"></script><script src="/shared/global/third-party/sarissa/0.9.7.6/sarissa_dhtml.js" type="text/javascript"></script><script src="js/ajaxtooltip.js" type="text/javascript"></script>
<style media="screen, projection" type="text/css">
    @import "css/loader.css";
   @import "css/en_us/language.css";
   @import "css/en_us/battlegroups.css";
</style>
<script type="text/javascript">
//

if (is_ie7)
    document.write('<link rel="stylesheet" type="text/css" media="screen, projection" href="css/ie7.css" />');

if (is_ie6 && !is_ie7)
    document.write('<link rel="stylesheet" type="text/css" media="screen, projection" href="css/ie.css" />');
   
if (is_opera)
    document.write('<link rel="stylesheet" type="text/css" media="screen, projection" href="css/opera.css" />');
   
if (is_opera && is_mac)
    document.write('<link rel="stylesheet" type="text/css" media="screen, projection" href="css/opera-mac.css" />');

if (is_safari)
    document.write('<link rel="stylesheet" type="text/css" media="screen, projection" href="css/safari.css" />');

if (is_mac)
    document.write('<link rel="stylesheet" type="text/css" media="screen, projection" href="css/mac.css" />');
   
//
</script>

The sniffer code:

http://armory.worldofwarcraft.com/share ... tection.js

An "ancient" document detailing Mozilla's guidelines against sniffing browser identification:

http://devedge-temp.mozilla.org/viewsource/2001/fep/

An even more ancient page about the sniffer:

http://www.mozilla.org/docs/web-develop ... _type.html

LoL

I know there is also a user agent check when getting the main data xml (armory.worldofwarcraft/wow-europe.com/character-info.xml?r=realm&n=char)

For pulling data from it we've tried no user agent or a made-up user agent, and that didn't work. Using FF's user agent with a fake version number didn't work either.

Code: Select all

$curl = curl_init();
    $fp = fopen("xml/Atrocity.xml", "w");
   $useragent="Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.1) Gecko/20061204 Firefox/2.0.0.1";
    curl_setopt($curl, CURLOPT_USERAGENT, $useragent);
    curl_setopt ($curl, CURLOPT_URL, "http://armory.worldofwarcraft.com/guild-info.xml?r=Gorefiend&n=Atrocity");
   curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1);
    curl_setopt($curl, CURLOPT_FILE, $fp);
    curl_exec ($curl);
    curl_close ($curl);

That will fetch the XML for you. I am writing it to a file as you can see, but anything is possible.

Someone has made a siggen that pulls off the armory.
http://jebus.geek.nz/WoW/armory/

Wow, its pretty simple, but it works.

This would be nice to able to, at the very minimum, pull the character information off it it to "fill-in" for Guildies who haven't updated yet.....

The refresh rate isn't too impressive, and lends itself to the need for storing the character data locally.

For example, I logged off ~12:00 midnight last night. Today, when I came into work ~8:30 AM, it showed the gear I had acquired last night. However, when I checked again ~4:00 PM, I received the info:

"Character Data Awaiting Refresh

It seems like the character you are trying to view hasn't been active for a while. Since the Armory only tracks active characters, this character needs to have been active recently for detailed character data to show up."

So, an "active character" means you have to have been logged in within the past 16 hrs? Pretty frustrating (especially because that same info is shown on the siggen listed above).

Ya, it seems to be REALLY flaky on availability ever since it was made live.

I had a guildie I was using for testing, but 8 hours later, it's "gone". Hopefully they up the timeout limit. One week would be a bit better, IMO

since they started this "service" all my chars are inactive

Very strange

That sig-gen is pretty awesome! Anyone know of any other tools people have put together for integrating Armory data into other websites?

As soon as I noticed that the armory produces XML, I started working on a parser. I've gotten decently far into an integration module, but it's not ready for beta testing. I plan to make it integrate for 1-n characters, done by a checklist. You can have it automatically check all characters or simply those based on the last time they updated.

This does not, however, get to the root of the problem. It would be far nicer to have Roster Managers that are responsible for updating the entire roster for all guildies. The guild members could manually or automatically update the Managers with their data in WoW, and they in turn would update Roster. This, however, is way off topic, and deserves its own thread.