Project

General

Profile

Bug #258 » imdb.xml

patched system/scrapers/video/imdb.xml - foresto, 11/10/2013 09:34 AM

 
<?xml version="1.0" encoding="UTF-8"?>
<scraper framework="1.12x" date="2010-10-05" name="IMDb.com" content="movies" thumb="imdb.png" language="en">
<include>common/imdb.xml</include>
<include>common/tmdb.xml</include>
<include>common/movieposterdb.xml</include>
<include>common/impa.xml</include>
<include>common/dtrailer.xml</include>
<GetSettings dest="3">
<RegExp input="$$5" output="&lt;settings&gt;\1&lt;/settings&gt;" dest="3">
<RegExp input="$$1" output="&lt;setting label=&quot;Enable full cast credits&quot; type=&quot;bool&quot; id=&quot;fullcredits&quot; default=&quot;false&quot;&gt;&lt;/setting&gt;" dest="5">
<expression/>
</RegExp>
<RegExp input="$$1" output="&lt;setting label=&quot;Enable fanart from themoviedb.org&quot; type=&quot;bool&quot; id=&quot;fanart&quot; default=&quot;true&quot;&gt;&lt;/setting&gt;" dest="5+">
<expression/>
</RegExp>
<RegExp input="$$1" output="&lt;setting label=&quot;Enable posters from themoviedb.org&quot; type=&quot;bool&quot; id=&quot;tmdbthumbs&quot; default=&quot;true&quot;&gt;&lt;/setting&gt;" dest="5+">
<expression/>
</RegExp>
<RegExp input="$$1" output="&lt;setting label=&quot;Enable posters from IMPAwards&quot; type=&quot;bool&quot; id=&quot;impawards&quot; default=&quot;false&quot;&gt;&lt;/setting&gt;" dest="5+">
<expression/>
</RegExp>
<RegExp input="$$1" output="&lt;setting label=&quot;Enable posters from MoviePosterDB&quot; type=&quot;bool&quot; id=&quot;movieposterdb&quot; default=&quot;false&quot;&gt;&lt;/setting&gt;" dest="5+">
<expression/>
</RegExp>
<RegExp input="$$1" output="&lt;setting label=&quot;Enable IMDb trailers&quot; type=&quot;bool&quot; id=&quot;trailer&quot; default=&quot;true&quot;&gt;&lt;/setting&gt;" dest="5+">
<expression/>
</RegExp>
<RegExp input="$$1" output="&lt;setting label=&quot;Enable trailers from Dtrailer.com&quot; type=&quot;bool&quot; id=&quot;dtrailer&quot; default=&quot;false&quot;&gt;&lt;/setting&gt;" dest="5+">
<expression/>
</RegExp>
<RegExp input="$$1" output="&lt;setting label=&quot;IMDb Poster &amp;amp; Actor Thumb(s) Size&quot; type=&quot;labelenum&quot; values=&quot;192|256|384|512|1024&quot; id=&quot;imdbscale&quot; default=&quot;512&quot;&gt;&lt;/setting&gt;" dest="5+">
<expression/>
</RegExp>
<expression noclean="1"/>
</RegExp>
</GetSettings>
<NfoUrl dest="3">
<RegExp input="$$1" output="&lt;url gzip=&quot;yes&quot;&gt;http://www.\1/title/tt\2/&lt;/url&gt;&lt;id&gt;tt\2&lt;/id&gt;" dest="3">
<expression clear="yes" noclean="1">(imdb.com)/Title\?([0-9]*)</expression>
</RegExp>
<RegExp input="$$1" output="&lt;url gzip=&quot;yes&quot;&gt;http://www.\1\2/&lt;/url&gt;&lt;id&gt;tt\2&lt;/id&gt;" dest="3+">
<expression noclean="1">(imdb.com/title/tt)([0-9]*)</expression>
</RegExp>
</NfoUrl>
<CreateSearchUrl dest="3" SearchStringEncoding="iso-8859-1">
<RegExp input="$$1" output="&lt;url gzip=&quot;yes&quot;&gt;http://akas.imdb.com/find?s=tt;q=\1$$4&lt;/url&gt;" dest="3">
<RegExp input="$$2" output="%20(\1)" dest="4">
<expression clear="yes">(.+)</expression>
</RegExp>
<expression noclean="1"/>
</RegExp>
</CreateSearchUrl>
<GetSearchResults dest="8">
<RegExp input="$$5" output="&lt;?xml version=&quot;1.0&quot; encoding=&quot;iso-8859-1&quot; standalone=&quot;yes&quot;?&gt;&lt;results&gt;\1&lt;/results&gt;" dest="8">
<RegExp input="$$1" output="\1" dest="7">
<expression clear="yes">/title/([t0-9]*)/(combined|faq|releaseinfo|vote)</expression>
</RegExp>
<RegExp input="$$1" output="&lt;entity&gt;&lt;title&gt;\1&lt;/title&gt;&lt;year&gt;\2&lt;/year&gt;&lt;url cache=&quot;$$7-main.html&quot;&gt;http://akas.imdb.com/title/$$7/&lt;/url&gt;&lt;id&gt;$$7&lt;/id&gt;&lt;/entity&gt;" dest="5">
<expression clear="yes" noclean="1">&lt;meta name=&quot;title&quot; content=&quot;(?:&amp;#x22;)?([^&quot;]*?)(?:&amp;#x22;)? \([^\(]*?([0-9]{4})(?:–\s)?\)</expression>
</RegExp>
<RegExp input="$$1" output="&lt;entity&gt;&lt;title&gt;\2&lt;/title&gt;&lt;year&gt;\1&lt;/year&gt;&lt;url cache=&quot;$$7-main.html&quot;&gt;http://akas.imdb.com/title/$$7/&lt;/url&gt;&lt;id&gt;$$7&lt;/id&gt;&lt;/entity&gt;" dest="5+">
<expression fixchars="2" noclean="1">&lt;meta name=&quot;title&quot; content=&quot;(?:&amp;#x22;)?[^&quot;]*?(?:&amp;#x22;)? \([^\(]*?([0-9]{4})(?:–\s)?\).*?Also Known As:&lt;/h4&gt;([^\n]*)</expression>
</RegExp>
<RegExp input="$$1" output="\1" dest="4">
<expression noclean="1">&lt;table class=&quot;findList&quot;(.*?)&lt;/div</expression>
</RegExp>
<RegExp input="$$4" output="&lt;entity&gt;&lt;title&gt;\2&lt;/title&gt;&lt;year&gt;\3&lt;/year&gt;&lt;url cache=&quot;\1-main.html&quot;&gt;http://akas.imdb.com/title/\1/&lt;/url&gt;&lt;id&gt;\1&lt;/id&gt;&lt;/entity&gt;" dest="5+">
<expression repeat="yes" noclean="1,2">&lt;td\sclass=&quot;result_text&quot;&gt;\s&lt;a\shref=&quot;/title/([t0-9]*)/[^&gt;]*&gt;(?:&amp;#x22;)?([^&lt;]*?)(?:&amp;#x22;)?&lt;/a&gt;\s*(?:\([IV]+\) )?\([^\(]*?([0-9]{4})[^\)]*\)\s(?:\(TV\sMovie\)\s|\(TV\sSpecial\)\s|\(Video\)\s)?&lt;</expression>
</RegExp>
<RegExp input="$$4" output="&lt;entity&gt;&lt;title&gt;\4&lt;/title&gt;&lt;year&gt;\3&lt;/year&gt;&lt;url cache=&quot;\1-main.html&quot;&gt;http://akas.imdb.com/title/\1/&lt;/url&gt;&lt;id&gt;\1&lt;/id&gt;&lt;/entity&gt;" dest="5+">
<expression repeat="yes" noclean="1,2">&lt;td\sclass=&quot;result_text&quot;&gt;\s&lt;a\shref=&quot;/title/([t0-9]*)/[^&gt;]*&gt;(?:&amp;#x22;)?([^&lt;]*?)(?:&amp;#x22;)?&lt;/a&gt;\s*(?:\([IV]+\) )?\([^\(]*?([0-9]{4})[^\)]*\)\s&lt;br/&gt;aka\s&lt;i&gt;&quot;([^&quot;]*)</expression>
</RegExp>
<expression clear="yes" noclean="1"/>
</RegExp>
</GetSearchResults>
<GetDetails dest="3">
<RegExp input="$$5" output="&lt;details&gt;\1&lt;/details&gt;" dest="3">
<RegExp input="$$2" output="&lt;id&gt;\1&lt;/id&gt;" dest="5">
<expression />
</RegExp>
<RegExp input="$$1" output="&lt;title&gt;\2&lt;/title&gt;" dest="5+">
<expression>&lt;meta name=&quot;title&quot; content=&quot;(IMDb - )?(?:&amp;#x22;)?([^&quot;]*?)(?:&amp;#x22;)? \([^\(]*?([0-9]{4})\)</expression>
</RegExp>
<RegExp input="$$1" output="&lt;year&gt;\1&lt;/year&gt;" dest="5+">
<expression>a href=&quot;/year/([0-9]*)</expression>
</RegExp>
<RegExp input="$$1" output="&lt;top250&gt;\1&lt;/top250&gt;" dest="5+">
<expression>Top 250 #([0-9]*)&lt;/a&gt;</expression>
</RegExp>
<RegExp input="$$1" output="&lt;mpaa&gt;\1&lt;/mpaa&gt;" dest="5+">
<expression>MPAA&lt;/a&gt;\)&lt;/h4&gt;\n?&lt;span itemprop=&quot;contentRating&quot;&gt;([^&lt;]*)</expression>
</RegExp>
<RegExp input="$$1" output="&lt;certification&gt;\1 \3&lt;/certification&gt;" dest="5+">
<expression repeat="yes">&lt;a href=&quot;/search/title\?certificates=[^&quot;]*&quot;&gt;([^&lt;]*)&lt;/a&gt;[^&lt;]*(&lt;i&gt;([^&lt;]*)&lt;/i&gt;)?</expression>
</RegExp>
<RegExp input="$$1" output="&lt;tagline&gt;\1&lt;/tagline&gt;" dest="5+">
<expression>&lt;h4[^&gt;]*?&gt;Taglines:&lt;/h4&gt;\n?([^&lt;]*)</expression>
</RegExp>
<RegExp input="$$1" output="&lt;runtime&gt;\1&lt;/runtime&gt;" dest="5+">
<expression trim="1">&lt;h4[^&gt;]*?&gt;Runtime:&lt;/h4&gt;[^0-9]*([0-9]+)</expression>
</RegExp>
<RegExp input="$$1" output="&lt;rating&gt;\1&lt;/rating&gt;&lt;votes&gt;\2&lt;/votes&gt;" dest="5+">
<expression>&lt;span itemprop=&quot;ratingValue&quot;&gt;([0-9.]+).*?&gt;.*?title=&quot;([0-9,]+)\s</expression>
</RegExp>
<RegExp input="$$1" output="&lt;genre&gt;\1&lt;/genre&gt;" dest="5+">
<expression repeat="yes">itemprop=&quot;genre&quot;[^&gt;]+&gt;([^&lt;]+)&lt;/a&gt;</expression>
</RegExp>
<RegExp input="$$1" output="&lt;country&gt;\1&lt;/country&gt;" dest="5+">
<expression repeat="yes">"/country/[^&gt;]+&gt;([^&lt;]+)&lt;/a&gt;</expression>
</RegExp>
<RegExp input="$$1" output="&lt;studio&gt;\1&lt;/studio&gt;" dest="5+">
<expression>"/company/[^&gt;]+&gt;([^&lt;]+)&lt;/a&gt;</expression>
</RegExp>
<RegExp input="$$1" output="&lt;outline&gt;\1&lt;/outline&gt;&lt;plot&gt;\1&lt;/plot&gt;" dest="5+">
<expression trim="1">&lt;p&gt;\n&lt;p itemprop=&quot;description&quot;&gt;(.+?)(&lt;|&nbsp;)</expression>
</RegExp>
<RegExp input="$$2" output="&lt;url function=&quot;GetIMDBPlot&quot;&gt;$$3plotsummary&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<RegExp conditional="!fullcredits" input="$$2" output="&lt;url cache=&quot;$$2-credits.html&quot; function=&quot;GetIMDBCast&quot;&gt;$$3&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<RegExp conditional="!fullcredits" input="$$2" output="&lt;url cache=&quot;$$2-credits.html&quot; function=&quot;GetIMDBDirectors&quot;&gt;$$3&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<RegExp conditional="!fullcredits" input="$$2" output="&lt;url cache=&quot;$$2-credits.html&quot; function=&quot;GetIMDBWriters&quot;&gt;$$3&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<RegExp conditional="fullcredits" input="$$2" output="&lt;url cache=&quot;$$2-fullcredits.html&quot; function=&quot;GetIMDBFullCast&quot;&gt;$$3fullcredits&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<RegExp conditional="fullcredits" input="$$2" output="&lt;url cache=&quot;$$2-fullcredits.html&quot; function=&quot;GetIMDBFullDirectors&quot;&gt;$$3fullcredits&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<RegExp conditional="fullcredits" input="$$2" output="&lt;url cache=&quot;$$2-fullcredits.html&quot; function=&quot;GetIMDBFullWriters&quot;&gt;$$3fullcredits&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<RegExp conditional="tmdbthumbs" input="$$2" output="&lt;url function=&quot;GetTMDBThumbsById&quot;&gt;$$3&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<RegExp conditional="impawards" input="$$2" output="&lt;url cache=&quot;$$2-impawards.html&quot; function=&quot;GetIMPALink&quot;&gt;$$3posters&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<RegExp conditional="movieposterdb" input="$$1" output="&lt;url function=&quot;GetMoviePosterDBLink&quot;&gt;http://www.movieposterdb.com/search?type=movies&amp;amp;query=\1&lt;/url&gt;" dest="5+">
<expression >/title/tt([t0-9]*)/faq</expression>
</RegExp>
<RegExp conditional="trailer" input="$$1" output="&lt;url function=&quot;GetIMDBTrailer&quot;&gt;http://akas.imdb.com/video/imdb/vi\1/player&lt;/url&gt;" dest="5+">
<expression >/vi([0-9]*)/</expression>
</RegExp>
<RegExp conditional="dtrailer" input="$$6" output="&lt;url function=&quot;GetDTrailerLink&quot;&gt;http://en.dtrailer.com/movies/search/\1&lt;/url&gt;" dest="5+">
<RegExp input="$$4" output="\1-" dest="6">
<RegExp input="$$1" output="\1" dest="4">
<expression trim="1" noclean="1">&lt;h1&gt;([^&lt;]*)</expression>
</RegExp>
<expression repeat="yes">([a-zA-Z0-9]+)</expression>
</RegExp>
<expression />
</RegExp>
<RegExp input="$$2" output="&lt;url cache=&quot;$$2-posters.html&quot; function=&quot;GetIMDBThumbs&quot;&gt;$$3mediaindex?refine=poster&amp;ref_=ttmi_ref_pos&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<RegExp conditional="fanart" input="$$2" output="&lt;url function=&quot;GetTMDBFanartById&quot;&gt;$$3&lt;/url&gt;" dest="5+">
<expression />
</RegExp>
<expression noclean="1"/>
</RegExp>
</GetDetails>
</scraper>
(2-2/7)