Just thought I'd chime in here.
I also noticed the issue of
Cast not available when scraping movies when the issue started almost a week ago.
TMDB common scraper v3.2.8 seemed to have addressed that issue.
I've since updated to
TMDB common scraper v3.2.9 and now
3.2.10.
I'm not sure if
v3.2.9 or
3.2.10 introduced a new issue or if TMDB made some additional changes on their end that may have messed up the scraping process further.
But what I'm noticing now is that when I scrape some movies, the first actor/actress in the cast list is missing completely. So for example, I scraped the movie
Bowling for Columbine after I renamed the file.
Michael Moore should be the first actor in the cast list according to TMDB. When Kodi does the scrape,
Michael Moore is nowhere to be found. I'm not sure if it's the same issue that Karellen indicated in post #18 of this thread, because
Bowling for Columbine is a documentary and for most of the cast, their character names just say
Himself or
Herself. So it's possible that
Michael Moore was bumped off the list, however, his character name of
Himself was given to the next actor on the list which is
George H. W. Bush who also has a character name of
Himself. Hope this makes sense.
Now if I select the
Extended info add-on, it pulls up all the movie details, and
Michael Moore is indeed listed as the first actor in the cast list with a character name of
Himself.
Lets make things even more interesting. I re-scraped the movie
Lord of War where
Nicolas Cage is the first actor listed. The movie rescraped just fine.
Nicolas Cage is still the first actor listed with the correct character name of
Yuri Orlov.
I'm guessing you'll want logs. I'm also guessing you'll want verbose logging for the libcURL enabled, so I took the liberty of enabling that. Here are the logs:
https://paste.kodi.tv/emecomotog.kodi
I rescraped
Bowling for Columbine at 1:42PM Eastern Time on 11/19/2020. Confirmed that
Michael Moore is still missing from the cast list. But when I go to
Extended info,
Michael Moore is listed there.
I rescraped
Lord of War at 1:44PM Eastern Time on 11/19/2020. Confirmed the cast list scraped just fine and
Nicolas Cage is the first cast member on the list.
This is a snippet of what I found in the logs:
2020-11-19 13:43:26.872 T:22082 WARNING: using DialogBusy from python results in nop now
2020-11-19 13:43:30.468 T:22082 ERROR: NEWADDON Unknown Video Info Key "file"
2020-11-19 13:43:30.584 T:22082 ERROR: Previous line repeats 3 times.
2020-11-19 13:43:30.584 T:22082 ERROR: NEWADDON Unknown Video Info Key "media_type"
2020-11-19 13:43:30.820 T:22103 ERROR: Previous line repeats 19 times.
2020-11-19 13:43:30.820 T:22103 ERROR: Exception in thread Thread-3:
Traceback (most recent call last):
File "/home/jenkins/workspace/Android-ARM-Leia/tools/depends/xbmc-depends/arm-linux-androideabi-21-release/lib/python2.7/threading.py", line 801, in __bootstrap_inner
File "/home/jenkins/workspace/Android-ARM-Leia/tools/depends/xbmc-depends/arm-linux-androideabi-21-release/lib/python2.7/threading.py", line 754, in run
File "/storage/emulated/0/Android/data/org.xbmc.kodi/files/.kodi/addons/script.extendedinfo/resources/lib/dialogs/DialogBaseInfo.py", line 259, in get_youtube_vids
self.yt_listitems = youtube.search(search_str, limit=15)
File "/storage/emulated/0/Android/data/org.xbmc.kodi/files/.kodi/addons/script.module.kodi65/lib/kodi65/youtube.py", line 199, in search
listitems = handle_videos(results["items"], extended=extended)
KeyError: 'items'
2020-11-19 13:43:30.890 T:22101 ERROR: DoWork - Direct texture file loading failed for resource://resource.images.studios.white/Iconolatry Productions Inc..png
2020-11-19 13:43:30.895 T:22101 ERROR: DoWork - Direct texture file loading failed for resource://resource.images.studios.white/Salter Street Films International.png
2020-11-19 13:43:30.896 T:22101 ERROR: DoWork - Direct texture file loading failed for resource://resource.images.studios.white/TiMe Film- und TV-Produktions GmbH.png
2020-11-19 13:43:30.904 T:22099 ERROR: DoWork - Direct texture file loading failed for resource://resource.images.studios.white/Vif Babelsberger Filmproduktion GmbH & Co. Zweite KG.png
2020-11-19 13:43:30.911 T:22099 ERROR: DoWork - Direct texture file loading failed for resource://resource.images.studios.white/United Broadcasting Inc..png
2020-11-19 13:43:30.915 T:22102 ERROR: DoWork - Direct texture file loading failed for resource://resource.images.studios.white/Dog Eat Dog Films.png
2020-11-19 13:43:42.524 T:22082 WARNING: CPythonInvoker(3, /storage/emulated/0/Android/data/org.xbmc.kodi/files/.kodi/addons/script.extendedinfo/default.py): the python script "/storage/emulated/0/Android/data/org.xbmc.kodi/files/.kodi/addons/script.extendedinfo/default.py" has left several classes in memory that we couldn't clean up. The classes include: N9XBMCAddon7xbmcgui10DialogBusyE
2020-11-19 13:43:55.599 T:22099 NOTICE: CDVDVideoCodecFFmpeg::Open() Using codec: H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10
2020-11-19 13:45:14.494 T:21835 WARNING: Create - unsupported protocol(image) in image://Browse/
2020-11-19 13:45:14.494 T:21835 ERROR: Exists - Error checking for image://Browse/
I'll be honest, I'm not sure if the log lines are related to the scraping issue. I'm just going by log lines that were captured around the time I reproduced the issue.
So just for my understanding, the issue that started about a month ago where movie posters failed to scrape was due to a change on CloudFare. Now this particular issue where the cast list fails to scrape is due to a change on TMBD's side? I believe it was mentioned there were changes to certain changes to properties for the cast list? Whatever happened to the saying "If it's not broken, don't fix it!"
I'm also having an issue with scraping TV shows. But I'm done troubleshooting for now. That'll have to be dealt with another day.