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