New Signature Pic feature [FEEDBACK]

Want to help out? Need help accessing the AniDB API? This is the place to ask questions.

Moderator: AniDB

Rar
AniDB Staff
Posts: 1471
Joined: Fri Mar 12, 2004 2:41 pm
Location: UK
Contact:

Post by Rar »

NOW I know what that line in the first Aika OP means!

Rar
exp
Site Admin
Posts: 2438
Joined: Tue Oct 01, 2002 9:42 pm
Location: Nowhere

Post by exp »

the server to host this is already up and running.
it's a dedicated server which basically doesn't do much else besides hosting the sig server.
so lots of free space, cpu power and bandwidth.

here are some infos about the current php setup:
http://sig.anidb.info/phptest.php

BYe!
EXP
Cyrano423
Posts: 35
Joined: Fri Jul 28, 2006 9:32 pm
Location: New York, NY
Contact:

Post by Cyrano423 »

K ... I need a bit of clarification. What's the purpose of the database on the sig server?

Is the database for the user's image settings? The way we handled audioscrobbler's original sig was all the settings and parameters were passed via the url. So there was no need for the storage of said settings.

If the AniDB XML feed is only capable of passing information about users in the system, so why the need for an additional userlist? You can eliminate the need for a second database, and the signup requirement.

So I'm wondering if I'm missing something. As I can't see any information that would need to be written or stored to the database. It wouldn't need access to modify or even read from the anidb, since nothing's written ... and all the information needed is given in the XML file / feed.
pelican
AniDB Staff
Posts: 234
Joined: Wed Aug 11, 2004 11:19 pm

Post by pelican »

You get whatever you need. You use that for whatever you need it for.

It doesn't matter as long as the code is written, it goes online and exp goes back to being content.
Cyrano423
Posts: 35
Joined: Fri Jul 28, 2006 9:32 pm
Location: New York, NY
Contact:

Post by Cyrano423 »

I'm debating if I want the feed in RSS or Atom. While Atom is future-proof RSS is more widely adopted.

hminuruminarumina jdlkajsdklja sd ehhh fuck it ... gimme the feed in RSS 2.0 and I'll do it this weekend after I'm done with my client, or next weekend if yall are busy and can't deal with the feed before then.

I don't need access to any database ... just need the feed live. If you want ideas for the data in the feed you can take a look at audioscrobbler's RSS 1.0 feed: here. If you want an example of a more concise feed here:

<myipaddress>/Clients/AniDB/feeds/?yhbt.mine.nu/t

If you're wondering what script that is ... it's a tracker parser, which outputs all torrents as an XML/RSS 2.0 feed, for automatic RSS downloading. It'll actually parse any page as long as said page contains a link to a torrent (a url containing a torrent extension), and will output all torrents found as a feed discarding all other content. If you want a copy of it you can get it here:

<myipaddress>/Clients/AniDB/feeds/feeds.cab

Maybe yall can make that live on the same image server, for anidb users to use or something. Or you can just use it yourselves if using RSS feeds for torrents is convenient. When I'm working on it I'll be doing so locally. In which you can check it ... here:

<myipaddress>/Clients/AniDB/

if you want to see all of the files you can access that here:

<myipaddress>/Clients/AniDB/raw

which will give you the raw directory listing of everything without parsing any server-side scripts. As such from there you can download the files if you want to see the source code as I'm working on it (if you're curious). I dunno how long it'll take ... depends on how much sleep I had and if I'm in the zone or not (and thus consistently working on it). Though I doubt it'll take long at all. Maybe a couple of hours. Though if I'm busy it'll be a couple of days when time permits.

Cool?



PS: I realize I didn't document that parser (since I wasn't expecting to distribute it) though the url syntax can go like this:

<scriptaddress>?url=http://<trackerurl>
<scriptaddress>?url=<trackerurl>
<scriptaddress>?u=http://<trackerurl>
<scriptaddress>?u=<trackerurl>
<scriptaddress>?http://<trackerurl>
<scriptaddress>?<trackerurl>

don't ask me why I decided to do that many ways of saying the same thing. I guess I wanted to be more concise in one case ... but wanted to provide other methods for people who like to know what's being sent or read (though I wasn't planning on distributing it, I code with other people in mind by default). *shrugs* Like I said ... I dunno, but I did it, lol. I personally use the shortest format.

If for some reason you find a tracker or page with torrents on it that it doesn't parse correctly, let me know. I tried to find as many diverse sources as possible ... but realize that someone else might find something I neglected.

PPS: oh btw I just noticed ... if you can, install php5.
W1N9Zr0
Posts: 6
Joined: Sat Apr 30, 2005 4:26 am

Post by W1N9Zr0 »

And what does atom and rss have to do with the sig server?
The xml server has to just display the current information for the user, not the kind of info you need a feed for.
To me it also seems that this topic is asking for which data to make available and what other options to give to the sig server users rather than implementation.
Cyrano423
Posts: 35
Joined: Fri Jul 28, 2006 9:32 pm
Location: New York, NY
Contact:

Post by Cyrano423 »

W1N9Zr0 wrote:And what does atom and rss have to do with the sig server?
The xml server has to just display the current information for the user, not the kind of info you need a feed for.
To me it also seems that this topic is asking for which data to make available and what other options to give to the sig server users rather than implementation.
Everything you said other than the question asked, was incorrect. Though I'll respond to your question first, since it was stated first. Atom and RSS is merely the standardized format in which the data will be sent.

The XML server doesn't just display the information ... if that's the case then it would be nothing more than a string of text. Said information must be in an XML format hence the "XML server". Which conveys more infomation than merely the user's info. It convey data, how the data is arranged, what the data belongs to, and it's relation to the rest of the data. Which means said information is categorized. The categories can be anything one chooses it to be since XML tags and attributes aren't set in stone. While Exp, or pelican can display the data in any manner (using whatever tags) that they like, I'd prefer that the information is in a standardized format. To use that standardized format as a base, and build on that (add whatever tags they need to that base). This way that very same information can be used by other sources, programs of whatever they feel like using it for (without modification) ... because the other programs will at the very least understand the standardized root format.

"Feeds" do not only display "news" ... and even "newsfeeds" do not only display news. So before you speak on what you need a feed for maybe you should understand what a feed is.
http://www.answers.com/feeds wrote: A general term for the electronic distribution of text, audio or video. It may refer to a syndicated radio or TV program that is transmitted on a regular basis, or to a text and graphics feed that is available on a Web site or blog. See syndication feed.
So basically ... you have an XML file, when it's served that file becomes a feed. The title of "newsfeed" simply means, "served XML data which usually but doesn't always contain news." RSS & ATOM and "newsfeed" are not synonomous, they are "a standardized XML structure".

That aside, even your last statement is incorrect. That "was" what the forum topic was for ... but it ceased to be the case when no one was stepping up to the plate. It then became, "forget about the extra stuff, let's just get it done."
pelican
AniDB Staff
Posts: 234
Joined: Wed Aug 11, 2004 11:19 pm

Post by pelican »

Please either shut up and code or shut up and don't code. Reading your posts has become the low point of my day.

We don't need any ideas for the format. We have an xml zealot on the team already, and a far less annoying one.

Also, feeds are specifically documents which link content updates, at least in common usage and that's all that really matters. AniDB has such a feed for files, although I don't know whether anyone uses it.
Cyrano423
Posts: 35
Joined: Fri Jul 28, 2006 9:32 pm
Location: New York, NY
Contact:

Post by Cyrano423 »

pelican wrote:Please either shut up and code or shut up and don't code. Reading your posts has become the low point of my day.

We don't need any ideas for the format. We have an xml zealot on the team already, and a far less annoying one.
How can one code without knowing the structure of the data being received? I need to know the structure of the XML file, and the location & parameters of the feed. I stated RSS because that at least gives me an idea of the general structure. Is he / she using a custom schema?
pelican
AniDB Staff
Posts: 234
Joined: Wed Aug 11, 2004 11:19 pm

Post by pelican »

90% of the code doesn't need to know. 90% gets you quite a long way.

Anyway, we don't _have_ a "feed" if you insist on calling that with much of the relevant data yet. Just assume you can have anything that you can get by other means on anidb for now.

Also, I think you missed the "shut up" part, despite quoting it.
exp
Site Admin
Posts: 2438
Joined: Tue Oct 01, 2002 9:42 pm
Location: Nowhere

Post by exp »

hell, Cyrano423, did you smoke something?

here is one example of the kind of xml data you would get from anidb.
caution, don't access these urls more than one or twice a day or you will get yourself an automatic 24h ban!

http://anidb.info/perl-bin/animedb.pl?s ... &list=file
http://anidb.info/perl-bin/animedb.pl?s ... list=anime

a lot more would be added to accomodate all the other data users might want in their pictures.

BYe!
EXP
cenedra
Posts: 18
Joined: Wed Jul 26, 2006 5:58 am
Location: SW Virginia
Contact:

Post by cenedra »

uum... not to offend anyone, but why would i want my list in my sig? i already have a place for that, 'mylist'.. couldn't we have a picture sig option too? something like this?

Image

just wondering...

CeNedra
W1N9Zr0
Posts: 6
Joined: Sat Apr 30, 2005 4:26 am

Post by W1N9Zr0 »

cenedra wrote:why would i want my list in my sig? i already have a place for that, 'mylist'.. couldn't we have a picture sig option too? something like this?
The sig would allow you to display the statistics for your mylist and a couple most recently added files/anime, not the whole mylist.
This info would be placed on top of an uploaded image, so it is an picture sig.
cenedra
Posts: 18
Joined: Wed Jul 26, 2006 5:58 am
Location: SW Virginia
Contact:

Post by cenedra »

firstly, thank you for replying..
secondly,
i understand that it would show some stats, not all, but what if some people wanted a picture ONLY? with no stats? would that be ok as well?

thanks again
cenedra
MyPlague
Posts: 2
Joined: Wed Aug 09, 2006 11:23 pm

Post by MyPlague »

The whole point is to show the stats. I don't think it's anidbs plan to start designing pictures for their visitors. Besides, "just pic sigs" have nothing to do with anidb for that matter. It's up to people to design them themselves imo.
Locked