[danmedhurst]

Still appears to not be working for me.. i get an error message on the screen but can't identify the error from the log annoyingly..my version shows 3.2.6

[Karellen]

get an error message on the screen but can't identify the error from the log annoyingly..my version shows 3.2.6

Provide the Debug Log that captures the error

[olympia]

Posters and fanart can now be scraped but they are always in english language (en) instead of the configured language (e.g. DE-de).

This is fixed in v3.2.16

[olympia]

Is the transition to python versions mandatory as I am waiting for the matrix 19 (osmc)?

No, it's not mandatory. However, there is probably not a lot of reason to use tmdb xml scrapers. Maybe there are a little bit faster, but that's it.
Python scrapers are not vulnerable for API layout changes in the way xml scrapers are.

Universal Movie Scraper XML however - for my taste - is another story. That's what I personally use and will use.

[jcea]

I think your situation is extremely rare, but a valid use case, so I added back scraping of runtime at UMS v5.5.3.

Great. Thanks, @olympia .

The old regex doesn't work with the new (UGLY!) IMDB design.

I have a question... Checking the Kodi sourcecode, looks like "runtime" units are "minutes". New IMDB does "Xh Ymin". We should give back "runtime" as "60*X+Y", but I don't think I can do math with regular expressions.

Any hint?

Thanks.

[milaq]

Posters and fanart can now be scraped but they are always in english language (en) instead of the configured language (e.g. DE-de).

This is fixed in v3.2.16

Thanks for the swift response and the fix!
Can confirm that it's working fine again.

[esefix]

Thanks for you endeavours.
But unfortunately, this does _not_ fix the issue correctly.

Posters and fanart can now be scraped but they are always in english language (en) instead of the configured language (e.g. DE-de).

Is there anything I overlooked or is anyone else noticing this?

You're right, I also have the same behavior on my side, the posters are in English while I have configured the scraper to receive them in FR. Even if I manually select the poster, I can't see the FRs.

Well that's not really a big problem because I always prefer to scrape by keeping the original titles and the posters in English

[esefix]

This is fixed in v3.2.16

You're too fast, big thanks

[teriyaki]

Is the transition to python versions mandatory as I am waiting for the matrix 19 (osmc)?

No, it's not mandatory. However, there is probably not a lot of reason to use tmdb xml scrapers. Maybe there are a little bit faster, but that's it.

What's the reason for the slowness of the python scrapers? Is there anything that can be done to speed them up? I've noticed with tv shows, the more seasons & episodes, the slower and very noticeable it is. Is it fetching the entire tv show every scan rather than just fetching episodes that aren't already found in the database?

[feuxeu77]

Hi

I faced issues adding some movies to my Library. Namely "Coco", "Dumbo" and "Taxi". UMS could not propose any match, even after I entered the movie title, like "Coco".

Checking the debug logs, I found this line:
2021-07-04 21:40:38.324 T:140491560040192   DEBUG: scraper: CreateSearchUrl returned <url>https://www.imdb.com/find?q=Coco&s=tt/|accept-language=en-us</url>

 As you can see, there is a trailing slash after the "tt". Because of this, IMDb doesn't search into titles only but also in names, etc.... If you try with "Coco", you will find in the web page that the first table class="findList" is actually the "Names" section and, as the Regexp doesn't really go into the next sections (I guess because it encounters a </div> before), it fails to parse the Titles list. All this is probably the result of the new IMDb appearance.

I fixed it locally by slightly changing the universal.xml this way (removing the trailing slash in the generated URL):

xml:
46c46

<                       <RegExp input="$$1" output="&lt;url&gt;https://www.imdb.com/find?q=\1&amp;s=tt|accept-language=en-us&lt;/url&gt;" dest="17">

---

>                       <RegExp input="$$1" output="&lt;url&gt;https://www.imdb.com/find?q=\1&amp;s=tt/|accept-language=en-us&lt;/url&gt;" dest="17">

112c112

<                               <RegExp input="$INFO[imdbsearchlanguage]" output="&lt;url function=&quot;GetAKASearchResults&quot;&gt;https://www.imdb.com/find?q=$$9&amp;s=tt|accept-language=$INFO[imdbsearchlanguage]&lt;/url&gt;" dest="5+">

---

>                               <RegExp input="$INFO[imdbsearchlanguage]" output="&lt;url function=&quot;GetAKASearchResults&quot;&gt;https://www.imdb.com/find?q=$$9&amp;s=tt/|accept-language=$INFO[imdbsearchlanguage]&lt;/url&gt;" dest="5+">

I let you check if this change makes sense or not.


Also note that searching by IMDb ID (like tt2380307) doesn't seem to work neither.

Thanks for your amazing job!

[feuxeu77]

I forgot to say I am using Kodi 18.9.0 and UMS 5.5.3

[olympia]

Hi

I faced issues adding some movies to my Library. Namely "Coco", "Dumbo" and "Taxi". UMS could not propose any match, even after I entered the movie title, like "Coco".

Checking the debug logs, I found this line:
2021-07-04 21:40:38.324 T:140491560040192   DEBUG: scraper: CreateSearchUrl returned <url>https://www.imdb.com/find?q=Coco&s=tt/|accept-language=en-us</url>

 As you can see, there is a trailing slash after the "tt". Because of this, IMDb doesn't search into titles only but also in names, etc.... If you try with "Coco", you will find in the web page that the first table class="findList" is actually the "Names" section and, as the Regexp doesn't really go into the next sections (I guess because it encounters a </div> before), it fails to parse the Titles list. All this is probably the result of the new IMDb appearance.

I fixed it locally by slightly changing the universal.xml this way (removing the trailing slash in the generated URL):

xml:
46c46

<                       <RegExp input="$$1" output="&lt;url&gt;https://www.imdb.com/find?q=\1&amp;s=tt|accept-language=en-us&lt;/url&gt;" dest="17">

---

>                       <RegExp input="$$1" output="&lt;url&gt;https://www.imdb.com/find?q=\1&amp;s=tt/|accept-language=en-us&lt;/url&gt;" dest="17">

112c112

<                               <RegExp input="$INFO[imdbsearchlanguage]" output="&lt;url function=&quot;GetAKASearchResults&quot;&gt;https://www.imdb.com/find?q=$$9&amp;s=tt|accept-language=$INFO[imdbsearchlanguage]&lt;/url&gt;" dest="5+">

---

>                               <RegExp input="$INFO[imdbsearchlanguage]" output="&lt;url function=&quot;GetAKASearchResults&quot;&gt;https://www.imdb.com/find?q=$$9&amp;s=tt/|accept-language=$INFO[imdbsearchlanguage]&lt;/url&gt;" dest="5+">

I let you check if this change makes sense or not.


Also note that searching by IMDb ID (like tt2380307) doesn't seem to work neither.

Thanks for your amazing job!

This will be in conflict with this:
https://forum.kodi.tv/showthread.php?tid...pid3043555

[feuxeu77]

I see.

Another solution could be about changing line 101 to something like this (repeat=yes):

xml:


                    <expression repeat="yes" noclean="1">&lt;table class=&quot;findList&quot;(.*?)&lt;/div</expression>

This will allow the regexp to go into the next sections and it seems to give pretty good results (not at many titles as my initial proposal but I can at least find my movies)

[olympia]

@Zippy79 are you around?

[Selyan]

Hello to all,

Thank you for welcoming me on this forum for this first post. My native language is French, so please be patient.

To get to the heart of the matter, I'm using Kodi 18.9 and Universal Movie Scraper 4.1.16. The scraper is an old version, but I keep it precisely for this reason : the casting was not displayed anymore, and this old version solved the problem.

For some time now, it's the movie poster that doesn't displayed anymore. Sometimes it still shows up, but that's rare. I put the posters manually, it's easy but more time consuming.

I have tried UMS 4.1.18 and The Movie Database Scraper, without success. I tried to get the posters with UMS from IMDb rather than TMDb, but there is no progress.

Even the very latest version of Kodi doesn't fix the problem.

Does anyone have an idea ?

Thanks in advance for the answer.