Reply to thread

Message

<blockquote data-quote="RoChess" data-source="post: 660076" data-attributes="member: 18896">Since you can build filters based on certification results, it is usually preferred to have consistent results. So to prevent mixing UK and US certs, you could modify the scraper to make a replacement call based on US cert. It won't be pretty (certification usually isn't a direct conversion), but it's a rare case anyway when UK certs are not present and it's usually close enough.You get code such as:[code]&nbsp; &nbsp; &nbsp; &lt;!-- If UK certification failed, use converted US one --&gt;&nbsp; &nbsp; &nbsp; &lt;if test=&quot;${movie.certification}=&quot;&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &lt;parse name=&quot;certification_usa&quot; input=&quot;${details_page}&quot; regex=&quot;${rx_certification_usa}&quot; /&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &lt;!-- Fix localized rating, forced to swap order, as 'G' gets replaced inside 'PG' and 'PG-13' otherwise --&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &lt;replace name='certification1' input='${certification_usa[0][0]:htmldecode}' pattern='X' with='R18'/&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &lt;replace name='certification2' input='${certification1}' pattern='NC-17' with='18'/&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &lt;replace name='certification3' input='${certification2}' pattern='R' with='15'/&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &lt;replace name='certification4' input='${certification3}' pattern='PG-13' with='12'/&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &lt;replace name='certification5' input='${certification4}' pattern='PG' with='PG'/&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &lt;replace name='certification6' input='${certification5}' pattern='G' with='U'/&gt;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &lt;set name='movie.certification' value='${certification6}'/&gt;&nbsp; &nbsp; &nbsp; &lt;/if&gt;[/code]You place this after you first try to obtain the UK certification, and be sure to add the CDATA expression, which i renamed to '_usa' postfix.Enjoy.</blockquote>

[QUOTE="RoChess, post: 660076, member: 18896"] Since you can build filters based on certification results, it is usually preferred to have consistent results. So to prevent mixing UK and US certs, you could modify the scraper to make a replacement call based on US cert. It won't be pretty (certification usually isn't a direct conversion), but it's a rare case anyway when UK certs are not present and it's usually close enough. You get code such as: [code]  <if test="${movie.certification}="> <parse name="certification_usa" input="${details_page}" regex="${rx_certification_usa}" />  <replace name='certification1' input='${certification_usa[0][0]:htmldecode}' pattern='X' with='R18'/> <replace name='certification2' input='${certification1}' pattern='NC-17' with='18'/> <replace name='certification3' input='${certification2}' pattern='R' with='15'/> <replace name='certification4' input='${certification3}' pattern='PG-13' with='12'/> <replace name='certification5' input='${certification4}' pattern='PG' with='PG'/> <replace name='certification6' input='${certification5}' pattern='G' with='U'/> <set name='movie.certification' value='${certification6}'/> </if> [/code] You place this after you first try to obtain the UK certification, and be sure to add the CDATA expression, which i renamed to '_usa' postfix. Enjoy. [/QUOTE]

Verification