Backdrop scraper (1 Viewer)

[piernik]

Could anyone attach here working backdrop scrapper? I have no documentation and no example. Without them making good scraper is rather hard :/

 

[fforde]

Documentation is here:
Scraper Scripts - Moving Pictures

All scraper scripts shipped with Moving Pictures 1.0:
branches/1.0/MovingPictures/DataProviders/ScraperScripts - moving-pictures - Project Hosting on Google Code

I think the only data source that currently provides backdrops is themoviedb.org which is implemented via a C# class, not a scraper script. But a backdrops script is possible and documented at the link above.

 

[piernik]

On wiki page I can't find nothing about backdrops:/
"Purpose of script types. Backdrop scripts not yet implemented."
"Backdrop scraper (needs information)"

But if You could help...
I'm downloading backdrop from site where there is often 50+ photos... How to break loop?

<parse name="check2" input="${photo_page}" regex="${rx_backdrop_item}" />
  <if test='${check2[0][0]}!='>
    <parse name="backdrops_items" input="${photo_page}" regex="${rx_backdrop_item}" />
      <loop name="backdrop_item" on="backdrops_items">
        <set name='backdrop[${count}].url' value='${backdrop_item[0]}' />
      </loop>     
    </if>

 

[fforde]

Darn, I thought that documentation was up to date. Well there is an example in the movie.daum.net scraper script. It looks like you already are using the right field names though.

As for limiting how many times a loop iterates, you can set a "limit" attribute.

 

[piernik]

thx

 

[piernik]

One more thing...
I have two sources of covers. First is one file for main cover which \is the most acuratte. Next are for ex. from other countries.
But those next always overwite the main one. How to fix it?
Here is code:

<if test='${movie.site_id}!='>
			<retrieve name="film_page" url="http://www.filmweb.pl${movie.site_id}" allow_unsafe_header="true"  cookies="welcomeScreenNew=welcomeScreen"/>
			
			<parse name="main_cover" input="${film_page}" regex="${rx_main_cover}" />
			<replace name='main_cover[0][0]' input='${main_cover[0][0]}' pattern='\.1\.' with='.3.' />
			<set name='cover_art[0].url' value='${main_cover[0][0]}' />
			
			<retrieve name="details_page" url="http://www.filmweb.pl${movie.site_id}/posters" allow_unsafe_header="true"  cookies="welcomeScreenNew=welcomeScreen"/>

			<!-- if ad page try again -->
			<parse name="check" input="${details_page}" regex="${rx_ad_check}" />
			<if test='${check[0][0]}!='>
				<retrieve name="details_page" url="http://www.filmweb.pl${movie.site_id}/posters" allow_unsafe_header="true"  cookies="welcomeScreenNew=welcomeScreen"/>
			</if>
			
			<parse name="cover_arts" input="${details_page}" regex="${rx_cover_art}" />
			<loop name="cover_art" on="cover_arts">
				<replace name='cover_art[0]' input='${cover_art[0]}' pattern='\.2\.' with='.3.' />
				<set name='cover_art[${count}].url' value='${cover_art[0]}'/>
			</loop>
		</if>

 

[RoChess]

One more thing...
I have two sources of covers. First is one file for main cover which \is the most acuratte. Next are for ex. from other countries.
But those next always overwite the main one. How to fix it?
Here is code:

You are using 'cover_art[${count}].url', but you did not adjust the value on ${count} in your script, so it will start with value '0', which overwrites the original cover you set. You can also just add an 'if' check to see if the previous method resulted in a cover, that way you can skip the entire block to speed things up.

 

[piernik]

It turns out that You can't define count value before loop (it has no effect)
I did it this way

<parse name="cover_arts" input="${details_page}" regex="${rx_cover_art}" />
			<loop name="cover_art" on="cover_arts">
				<if test='${count}=0'>
					<set name='count' value='1' />
				</if>
				<replace name='cover_art[0]' input='${cover_art[0]}' pattern='\.2\.' with='.3.' />
				<set name='cover_art[${count}].url' value='${cover_art[0]}'/>
			</loop>