Anime lookup scraper for and/or
i know of the feature freeze, but, if i don't mention it now, i'll forget. Sad

basically, it's nice how xbmc can check imdb and cddb for album/movie information. but, i was thinking, it would be nice if i could look up anime information from via xbmc, also get the picture if one is present. i don't know if this is possible, but the site has extensions, and there's a few applications (gpl too) that can browse the site, maybe search. i mean, yeah it's just an idea, it could be a pain in the ass to know to look at animenfo instead of imdb (how is xbmc supposed to know this particular folder/file/whatever is anime and not a movie? i'm sure there are ways, maybe an "anime" view much like movies, or somehow telling xbmc where the anime is located via the config file or something.

edit: forgot something. (refer to top Wink)


animenfo client documentation
animenfo client examples (all gpl)
this would be a great feature for implementation.
it's got my bid.

if you have already seen, you may know that it's an amazing site that references all the anime releases. it lets each user manage his own list, with file status (on cd, on hdd, watched...), and so on.

but the best part: it provides a socket api that lets users develop their own custom clients in order to interact with their list and with the website database. the site is based on ed2k hashes for files unicity.

i was wondering if it would be easy / possible to implement support for that amazing api using _ maybe _ python.

what i would be interested in would be something like imdb support but for anidb. i feel like it is too complex for python and would require real core implementation, but it does not hurt to ask.

it would be great if it was possible to get the same kind of information for anime files: genre, title, episode titles, thumbnails, year... and so on.

any clue ? piece of advice ? comment ? thanks in advance !
ok so i watch a lot of serialised anime.  i would also like my anime collection to look as good and be as informative as my regular films.

ok i'll try too keep this to the point. oh and this could work for regular tv series i suppose...not just anime.

*when i scan anime and the imdb can't find it, it scans every episode.  not fun for those 50+ ep series. (plus the imdb thumbs are often just shots of boxset boxes)

*could there be a way to mark specific shares as containing anime (i have separate directories and shares for mine...don't know about anybody else) and have xbmc check the likes of or for synopsies, thumbs etc.

*also could the browsing of areas marked as serialised be given an option to apply the thumbs to the directories.  i'm presuming most people file their different series away in their own directories and mark the episodes like 'juuni kokki - 01.avi , juuni kokki - 02.avi etc.  the plan being that you select the directory and then rather than stacking you select an episode number or play all for those all night marathons.  this would tidy up boring looking directory structures filled with video files and replace them with lovely thumbs.

anyway i'll leave it there as i could go on about possible features that would suppost serialisations for pages.  i'll just let you guys tell me if it's feesable or not first.  if there is even significant demand for a feature like this.

i know i personally would use the databasing features if they supported any of my media effectivly.



i just found 2 other threads requesting simelar features here:
and here:
i think this feature is best implemented with a python script and not be part of the core code. is there a way to tie in python scripts to the context menu?

how about passing an argument to a python script?
i suppose if there was a way to have the script hit animenfo or anidb and get the details from there....then simply convert it so it looks just like data downloaded from imdb and save it in it's place. that could work couldn't it?

now all we need is a talented young scripter....with a thirst for the anime.

having said that i would still like to see it implemented like imdb is, but as long as i get the details and lovely thumbs i'm not bothered.

still doesn't solve the browsing being a big messy directory structure though.

well i don't realy see movies but i see alot of it possible to be able to choose from imdb to something like anidb or animeinfo to get information on the anime will be for movies and anidb will be for anime stuff?possible.
please do a search before posting next time

i wonder what interface you would use to determine if something is anime though...
i've done my own modification to xbmc something along the lines suggested in this forum.

it works as follows when you press the "info" button.
- looks for a file called xxx.afo in the folder (where xxx is the aindb id eg 3408.afo)
- if no file is found uses normal imdb lookup
- if the file is found uses xxx as part of url
- downloads the page and parses it for information

basically it uses the imdb info gui, and replaces it with the anidb information.
its pretty much a hack, but it gives me the info i want like the title, number of episodes, the anidb picture and plot summary.

the other change i included in this was to save the last file watched in each folder to the myvideos database. so whenever you restart the xbox, it automatically selects the last episode you watched when you open a folder(pretty handy if you watch a few different shows).

i know xbmc has a history while you have the xbox on, but it doesn't seem to persist it anywhere. is there an option somewhere to save the history and i missed it?
jonbaby, please see;st=45

if you'd be willing to do the scraper using my system, it would be pretty sweet Smile

if so, i'll hook u up with a new version and better documentation.
to be honest, i've never done anything with regular exp's before. i'll have to read up a bit on how to use them.

i had a quick look at the scrapers in the link you provided, and i get the idea of how you plan to do it. if you have any running test code that i can step through and see how it works that would be handy.
eg parse a page from imdb and extract the appropriate fields.

feel free to contact me at my emial address if that is easier.
so, do you think this project would be mor eexact if it looked for extensions rather than urls over and over using different sources?

quantum (results in lookup using the link in the file)
.afo results in anid etc.
and the thing could handle it automatically? i might be talking out of turn of course....

**alternate thought... couldnt the nfo file handling be altered a bit to search for the link and use the proper site?

using files like ".afo" is certainly more accurate, but hardly user friendly.

i did the modifications purely for myself, so i don't find it a bother. any time a new show is released, i just create a new afo file and place it in the folder.

there's a few reasons i did it like this.
- i mostly watch anime on xbmc
- having the afo file requires less button clicks (as my wife prefers it)
- i keep fairly up to date with new anime, but my wife doesn't. so when she sees a new show there in the anime folder, she can just click the "info" button to see what its about.
- i download too much anime to store on my pc. so a lot of it gets backed up to dvd without getting watched. i have a separate folder called "anime - list" where i just keep the empty folder with the afo file.
this folder can be used to look up anime that i have, but it is not available on the xbox.

i think i've figured out the general gist of how xml files need to be setup now (looking at the imdb example).

if you want to pass on that up to date source (and doco if you have it), i will attempt to create my own xml for anidb. i'll try integrating it into my version of xbmc for testing purposes.

if that works , i'll pass it onto you and then maybe do a version for animenfo, as the html looks a lot easier to parse than that of anidb.


yeah it would be pretty easy to create a general purpose ".nfo" file which contains criteria needed for exact searches on websites. but i don't think most people would be interested in setting these up manually.
my indication for the general purpose nfo file was meaning the first part of the link. the source code for example has/had examples of what the link would lok like and then how to process it

else if

shouldnt be too difficult to simply add the additional sections for the anim lookup or tv lookups
Thread Rating:
  • 2 Vote(s) - 4.5 Average

Logout Mark Read Team Forum Stats Members Help
Anime lookup scraper for and/or animenfo.com4.52