IMDb+ Plugin 2.1.0.291 with IMDb+ Scraper v4.9.xxx (3 Viewers)

[emphatic]

IMDB+ results doesn't scrape Swedish summary again. I've reverted back to a database I know is fine, restarted MediaPortal and get a message that IMDB+ has been updated, go into the GUI Import manager of Moving Picture and let it do it's thing. New movies added since the database backup was done are blank in the Summary field (at least).

I guess whatever site being scraped has changed significantly again.
Emph

 

[RoChess]

@emphatic, sorry, got sidetracked and overlooked not being logged into mepo so never got alerts. Will check into why .se is failing and add more improvements. Only changes I made was because IMDb imports failed for me on English movies.

 

[RoChess]

Seems the change might be as simple as changing the URI, but can you confirm that you are getting 404 errors in log file then on scraping?

 

[emphatic]

Seems the change might be as simple as changing the URI, but can you confirm that you are getting 404 errors in log file then on scraping?

I will take a look. I have a man cold right now, so not very productive.

Emph

 

[RoChess]

@emphatic, I looked at code more closely, and immediately noticed that my sample URL stopped working.

http://www.nyheter24.se/filmtipset/imdb/tt0499549

That should have reported back the filmtipset.se reference for Avatar, but instead it redirects to some basic info page on movies.

Without the IMDb tt-ID lookup there's nothing I can do, and no clue on Swedish language, so stuck to resolve.

Did they stop the service, or make changes? Can you reach out to devs/forum and inquire please?

After you feel better first obviously.

 

[RoChess]

Getting back on the horse.

Rename dbase updated -- Rename XML database v1.9.13 · RoChess/IMDbPlus@76b6fd6

 

[emphatic]

Did they stop the service, or make changes? Can you reach out to devs/forum and inquire please?

After you feel better first obviously.

I'm worse now, full on fever mode. At the moment though, I'm having a clear moment.

I did some checking on their forum, and they've switched owners a year ago. Also, when being asked about using their API, they asked a guy "why, what for?". I wonder if this scraper for My Videos is still working?: https://forum.team-mediaportal.com/download/2018/12/312424_filmtipset.se.xml

Maybe he's doing things in a different way?
Emph

 

[RoChess]

Rename dbase updated -- Rename XML database v1.9.15 · RoChess/IMDbPlus@f2089b2

Lot of sequels, and more to come

 

[RoChess]

Maybe he's doing things in a different way?

Yes, he relies on the filmtipset.se search system, and gets the Filmtipset Movie-ID that way.

For IMDb+ I'm forced to rely on the imdb.com search system, locate the movie, have user confirm the movie (or auto accept if the logic is pretty sure it is the correct movie), and then inside the details logic side of the script use the IMDb tt-ID to do an API search with that link I showed you to get the Filmtipset Movie-ID, which then allows me to scrape the Swedish movie details. It is not possible to do two searches, because the MovPic GUI does not allow you to select the manual results twice.

The not-recommended workaround would then be to do what is called a fuzzy search, and that is that I do a search within the details node via the filmtipset.se search engine, and accept the first match always and cross fingers it is the correct one.

The recommended workaround is to do a proper search with filmtipset.se that accepts searching for IMDb tt-ID, which is what that `nyheter24.se` script was doing.

Either their script broke, their API access key was disabled, or filmtipset.se broke/disabled their API to allow for IMDb tt-ID searches.

Google translate fails to translate -- Opendata.se: Filmtipset API -- for me, so let me know if that gives you any info.

Let me know what you can find out, otherwise fuzzy search it is.

 

[emphatic]

Google translate fails to translate -- Opendata.se: Filmtipset API -- for me, so let me know if that gives you any info.

Let me know what you can find out, otherwise fuzzy search it is.

It says that anyone wanting to use their API will need to ask for a key. There has been requests to the owners about this, but they have been really short in their replies, not at all encouraging. Sorry.

Emph