Reply to thread

Message

<blockquote data-quote="RoChess" data-source="post: 1169873" data-attributes="member: 18896">The best I have been able to deduct has been: NFO scan -&gt; Noise Filter -&gt; Internal IMDb lookup -&gt; pass control to &quot;Search node&quot; of scraper-script -&gt; process results -&gt; identify the movie auto or after user intervention -&gt; pass control to &quot;Details node&quot; of scraper-script -&gt; process results -&gt; pass control to any graphics &quot;node&quot; of scraper-script -&gt; store it all.The problem is that the search node fills in information already as well, for example it uses movie.popularity field to rank the search results with 100 to first match, and 99 to next, etc. The same goes for movie.release_date which I guess it also pre-populates. For me the latter is a bug, but it doesn't matter in the end, because the &quot;Details node&quot; will replace it anyway. The same way I overwrite movie.popularity with the rating votes, or the movie.title for that matter with IMDb+. Key to remember is to erase the search node results if your scraper-script gets no results.MovPic source code is available, so you can always look at that for the exact order that takes place, and perhaps even commit fixes <img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" class="smilie smilie--sprite smilie--sprite2" alt=";)" title="Wink&nbsp; &nbsp; ;)" loading="lazy" data-shortname=";)" />PS: The big difference is that only search node has access to media filename, which is why a lot of function of IMDb+ does not work unless the user sends movie back to importer, otherwise there is not much difference, and one of the reasons why I offer mass-update as an easy option in IMDb+ plugin with the additional option to only update certain fields (rating/votes/etc) that one would expect to be adjusted, but leave the big ones alone (title, summary, etc) to avoid erasing custom changes by user.</blockquote>

[QUOTE="RoChess, post: 1169873, member: 18896"] The best I have been able to deduct has been: NFO scan -> Noise Filter -> Internal IMDb lookup -> pass control to "Search node" of scraper-script -> process results -> identify the movie auto or after user intervention -> pass control to "Details node" of scraper-script -> process results -> pass control to any graphics "node" of scraper-script -> store it all. The problem is that the search node fills in information already as well, for example it uses movie.popularity field to rank the search results with 100 to first match, and 99 to next, etc. The same goes for movie.release_date which I guess it also pre-populates. For me the latter is a bug, but it doesn't matter in the end, because the "Details node" will replace it anyway. The same way I overwrite movie.popularity with the rating votes, or the movie.title for that matter with IMDb+. Key to remember is to erase the search node results if your scraper-script gets no results. MovPic source code is available, so you can always look at that for the exact order that takes place, and perhaps even commit fixes ;) PS: The big difference is that only search node has access to media filename, which is why a lot of function of IMDb+ does not work unless the user sends movie back to importer, otherwise there is not much difference, and one of the reasons why I offer mass-update as an easy option in IMDb+ plugin with the additional option to only update certain fields (rating/votes/etc) that one would expect to be adjusted, but leave the big ones alone (title, summary, etc) to avoid erasing custom changes by user. [/QUOTE]

Verification