http://www.dappit.com can create custom XML files for different web sites, like IMDB. That way there wouldn't be a need to update the python script if one the provider make a design modification. Reading XML instead of webpages would also save some bandwidth.
Example search for "Boogie Nights" in IMDB would return this XML page
http://www.dapper.net/RunDapp?dappName=IMDBMovieSearcher&v=1&variableArg_0=Boogie%20Nights
"During the beta, Dapper is free and open to all. In the future, non-commercial and small uses will remain free. Pricing structure for bigger applications and for commercial uses will be announced in the future."
Have a look at the demo!
http://www.dappit.com/dapperDemo/
It's really easy to set up a new dapp (no regexp involved!!). If the site changes, a regexp almost always needs to be updated. But a dapp might be able to handle some changes automatically. If it doesn't, then modify the dapp. The script doesn't need to be updated. There is an enormous performance gain in loading gzipped XML files.
Because
1. scrapers give more control
2. we don't rely on another service and
3. we don't have to rewrite much of the code.
Scraping xmls works quite well we do the same for http://tvdb.zsori.com/ (see zsori.xml).
If you create dapps with the same fieldnames etc. you would only have to make one scraper and then it's very little effort to change it to work with a different dapp.
You can write scrapers for the xml returned by dapp...
You can write scrapers for the xml returned by dapp...
Why? It's XML. Use a XML parser...
hardcoding against one interface. to put it in norwegian,
fy faen for en svinaktig dårlig ide.
spiff
 Where's The Advantage In Windows Genuine Advantage?
 Stocks Bounce After S&P Joins Bear Market
|
Why not use xml from dappit instead of scrapers? 
PRINT
Add to favorites