Advertisement
Guest User

Untitled

a guest
Dec 1st, 2012
1,007
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
XML 2.90 KB | None | 0 0
  1. <?xml version="1.0" encoding="utf-8"?>
  2. <scraper framework="11" date="2012-12-01" name="Data18" content="movies" language="en">
  3.     <CreateSearchUrl dest="3">
  4.         <RegExp input="$$1" output="&lt;url&gt;http://www.data18.com/search/?k=\1&amp;t=0&lt;/url&gt;" dest="3">
  5.             <expression />
  6.         </RegExp>
  7.     </CreateSearchUrl>
  8.     <GetSearchResults dest="6">
  9.         <RegExp input="$$4" output="&lt;?xml version=&quot;1.0&quot; encoding=&quot;utf-8&quot; standalone=&quot;yes&quot;?&gt;&lt;results&gt;\1&lt;/results&gt;" dest="6">
  10.             <RegExp input="$$1" output="&lt;entity&gt;&lt;title&gt;\2&lt;/title&gt;&lt;url&gt;\1&lt;/url&gt;&lt;/entity&gt;" dest="4">
  11.                 <expression repeat="yes">&lt;a href="(http://www.data18.com/[^/]*/\d+/)"&gt;&lt;img src=[^ ]* class=[^ ]* title="([^"]*)"</expression>
  12.             </RegExp>
  13.             <expression noclean="1" />
  14.         </RegExp>
  15.     </GetSearchResults>
  16.     <GetDetails dest="7">
  17.         <RegExp input="$$5" output="&lt;?xml version=&quot;1.0&quot; encoding=&quot;utf-8&quot; standalone=&quot;yes&quot;?&gt;&lt;details&gt;\1&lt;/details&gt;" dest="7">
  18.             <RegExp input="$$1" output="&lt;title&gt;\1&lt;/title&gt;" dest="5+">
  19.                 <expression noclean="1">&lt;h1 class="h1big"&gt;([^&lt;]*)&lt;/h1&gt;</expression>
  20.             </RegExp>
  21.             <RegExp input="$$1" output="&lt;plot&gt;\1&lt;/plot&gt;" dest="5+">
  22.                 <expression>&lt;b&gt;Description:&lt;/b&gt;([^\t\r\n]*)</expression>
  23.             </RegExp>
  24.             <RegExp input="$$1" output="&lt;actor&gt;&lt;name&gt;\3&lt;/name&gt;&lt;thumb&gt;\1120\2&lt;/thumb&gt;&lt;/actor&gt;" dest="5+">
  25.                 <expression repeat="yes" noclean="1">&lt;img src="(http://www.data18.com/img/stars/)60([^"]*).*?alt="([^"]*)</expression>
  26.             </RegExp>
  27.             <RegExp input="$$1" output="&lt;thumb aspect=&quot;poster&quot;&gt;\10\2&lt;/thumb&gt;" dest="5+">
  28.                 <expression noclean="1">&lt;img src="([^"]*)2_new([^"]*)" alt="Enlarge Cover"</expression>
  29.             </RegExp>
  30.             <RegExp input="$$1" output="&lt;runtime&gt;\1&lt;/runtime&gt;" dest="5+">
  31.                 <expression noclean="1">Length:&lt;/b&gt;[^\d]*(\d+)</expression>
  32.             </RegExp>
  33.             <RegExp input="$$1" output="&lt;director&gt;\1&lt;/director&gt;" dest="5+">
  34.                 <expression noclean="1">Director:&lt;/b&gt;[^&gt;]*&gt;([^&lt;]*)</expression>
  35.             </RegExp>
  36.             <RegExp input="$$1" output="&lt;studio&gt;\1&lt;/studio&gt;" dest="5+">
  37.                 <expression noclean="1">Studio:&lt;/b&gt;[^&gt;]*&gt;([^&lt;]*)</expression>
  38.             </RegExp>
  39.             <RegExp input="$$1" output="&lt;year&gt;\1&lt;/year&gt;" dest="5+">
  40.                 <expression noclean="1">Release date:&lt;/b&gt;[^\d]*(\d{4})</expression>
  41.             </RegExp>
  42.             <RegExp input="$$1" output="&lt;genre&gt;\1&lt;/genre&gt;" dest="5+">
  43.                 <RegExp input="$$1" output="\1" dest="1">
  44.                     <expression noclean="1" clear="yes">&lt;b&gt;Categories:&lt;/b&gt;(.*?)&lt;/div</expression>
  45.                 </RegExp>
  46.                 <expression repeat="yes" noclean="1" trim="1">&lt;a href=[^&gt;]*&gt;([^&lt;]*)</expression>
  47.             </RegExp>
  48.             <expression noclean="1" />
  49.         </RegExp>
  50.     </GetDetails>
  51. </scraper>
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement