IMDb parsing in a separate XML file?
#1
i read here that the imdb lookups don't work anymore, due to some layout changes on their website.

would it be possible to move the imdb parsing rules from the executable into a separate file. in that case, it would be easier to update this in case imdb decides to change their layout again. then it might even be possible to have some sort of an auto-update feature which checks a website for an updated rules file and then automatically updates it if necessary.
Reply
#2
this has been thought of before (it is a good idea, afterall!Wink and there is some ways it could be done.

the easiest way (imo) would be to have all internet based scrapping setup as a python script that xbmc runs. this way it's easy to add very specific rules etc. as all the programming is handled in python.

the disadvantage, ofcourse, is that people then require python to be running to take advantage of the internet items. (ie we have issues with scripts not being on the xbox, python libs not being there etc. etc.)

the alternative is to have a set of rules in an external file. the problem with this, is the rules required need to have the following ability:

1. specify url.
2. specify rules to grab info off the returned html such as what strings to search for etc.
3. be able to specify different cases that can (and do) occur.

this basically requires a small programming language written that can handle conditionals + branches etc. it's therefore not a straightforward task!

i'm interested in others' thoughts.

cheers,
jonathan
Always read the XBMC online-manual, FAQ and search the forum before posting.
Do not e-mail XBMC-Team members directly asking for support. Read/follow the forum rules.
For troubleshooting and bug reporting please make sure you read this first.


Image
Reply
#3
i'm not a coder, so i don't know if this is usefull, but this guy seems to maintain an imdbtoxml component to convert imdb data to a more generic xml format.

i like the idea to put this in python. even though python is optional, i think it's safe to say about 99% of the people who have their xbox hooked up to the internet have a working python on their xbox (for webradio or apple trailers). it might be possible to enable the imdb lookup only if python is installed.

is it a stupid idea to put the imdb lookup part in a dll? that way it can be included whenever needed and is still in a separate file.
Reply
#4
no, it's not a stupid idea to put it in a dll, and that is a possibility (a standard windows dll that just takes in html and spits out the search results or movie information in a consistent manner).

this would have the advantage of also being modular, in that there could be different dll's for different sites (or a single dll that works for multiple sites).

ofcourse, it means that any updates have to be built etc. and tested, though i don't see an issue with this. it would fit in a little tighter with xbmc and require just a dll which would be freely distributable.

btw: the last update of imdbtoxml was mid 2003, so i doubt it works with imdb now!

cheers,
jonathan
Always read the XBMC online-manual, FAQ and search the forum before posting.
Do not e-mail XBMC-Team members directly asking for support. Read/follow the forum rules.
For troubleshooting and bug reporting please make sure you read this first.


Image
Reply
#5
Question 
i like the xml parser filter idea, hope that get's done someday (and something similar for mp3 album lookup too), (...i don't like the python suggestion though)
Always read the XBMC online-manual, FAQ and search the forum before posting.
Do not e-mail XBMC-Team members directly asking for support. Read/follow the forum rules.
For troubleshooting and bug reporting please make sure you read this first.
Reply

Logout Mark Read Team Forum Stats Members Help
IMDb parsing in a separate XML file?0