(2015-09-18, 15:07)steve1977 Wrote: [ -> ] (2015-09-18, 09:51)DanCooper Wrote: [ -> ] (2015-09-18, 01:52)steve1977 Wrote: [ -> ]Great stuff, thanks to see the Kodi sync coming.
Another question: I have a few movies that are not part of IMDB, but only in TMDB. I scrape from both IMDB and TMDB, but scrape-order is IMDB->TMDB. Unfortunately, I cannot "identify" the TMDB-only movies. Is this a "bug" or am I missing something?
Atm, the only way to scrape TMDB-only movies is disable IMDB scraper. In near future we rewrite the search process to get results from all scrapers in one common dialog.
Sounds great. This also means that currently EMM does not scrape movie set / collection ID when selected IMDB (as only TMDB has a collection ID)?
Ember gets the CollectionID if TMDB scraper is enabled.
This is the way that scraping system works:
If you have more than one scraper enabled, the first scraper try to fill a field (for example CollectionID). IMDB can't get a CollectionID, so the next scraper can fill this field if the scraper has a value for this field.
If a field is filled by a scraper, this field will be locked for all following scrapers.
But we use only the first scraper to search a movie. If you click Cancel, no other scraper will be running... and this is why you can't find a movie that is only listed on TMDB if IMDb is your first scraper. It's a mess, but we are working on a better solution.
Got it, great. This makes a lot of sense and how i prefer it.
One awkward thing - it takes "forever" to scrape movies (the NFO, the artwork is actually fast). Even after scraping, it takes several minutes after clicking "ok" when it says "verifying movie details". Any idea what could trigger this? Anything I have enabled that I should not?
(2015-09-18, 15:17)steve1977 Wrote: [ -> ]Got it, great. This makes a lot of sense and how i prefer it.
One awkward thing - it takes "forever" to scrape movies (the NFO, the artwork is actually fast). Even after scraping, it takes several minutes after clicking "ok" when it says "verifying movie details". Any idea what could trigger this? Anything I have enabled that I should not?
Yes: Ember downloads all Actorthumbs after clicking OK in Edit Movie window. And yes, that's a fu*** slow process.
But we have optimised many, many, really many processes in our next beta, also added more informations what Ember exactly do while scraping.
Got it. Is this also the reason that the NFO scraping takes forever? Can I deactivate havign actorthumbs?
(2015-09-18, 15:37)steve1977 Wrote: [ -> ]Got it. Is this also the reason that the NFO scraping takes forever? Can I deactivate havign actorthumbs?
No, it takes so long because you use IMDB. IMDB has no public/free API and we have to call the movie site on IMDB, download the HTML sourcecode, parse/filter/group the HTML and search the different fields by regex.
TMDB has an API and we can get all informations in less than one second (after creating the API, that takes also ~1 seconds).
I thought there was this "mirror-IMDB" (forgot the URL) that has an API? How come you are not using this? Also, would the actorthumb issue still be there if I switch to TMDB?
(2015-09-18, 16:03)steve1977 Wrote: [ -> ]I thought there was this "mirror-IMDB" (forgot the URL) that has an API? How come you are not using this? Also, would the actorthumb issue still be there if I switch to TMDB?
Yes, there are some unofficial API services for IMDB. But we have to use one that is not depending on another webservice. But i've found some informations about the undocumented JSON and XML API of IMDB. Maybe we can use this for a better scraper.
And no, that "issues" are the same with TMDB actors. I don't know why sometimes the download is so slow.
Got it. I have now switched to TMDB, which is a lot faster. IMDB was unusably slow.
Pity not to have IMDB though. Will it slow down once I use one single field from IMDB?
(2015-09-18, 12:45)DanCooper Wrote: [ -> ]We have removed all that "Preferred Language" options and now the scrapers use the media language to get data and images.
I am not sure whether I fully understand it, but am a bit worried that this removal may not be ideal. For example, I have US movies in dual audio (German and audio). Given these are US movies, I would like to keep English as movie language and poster. Also, there are some HK movies that are in Cantonese, but I still like to keep the movie language to English.
Is this still possible?
(2015-09-19, 05:45)steve1977 Wrote: [ -> ] (2015-09-18, 12:45)DanCooper Wrote: [ -> ]We have removed all that "Preferred Language" options and now the scrapers use the media language to get data and images.
I am not sure whether I fully understand it, but am a bit worried that this removal may not be ideal. For example, I have US movies in dual audio (German and audio). Given these are US movies, I would like to keep English as movie language and poster. Also, there are some HK movies that are in Cantonese, but I still like to keep the movie language to English.
Is this still possible?
It's the same system like for tv shows:
Each movie source has a default language for new added movies. This language has nothing to do with audio streams or something else. Also you can manually change the language for each movie if you want. This language will be used to get the informations and posters. That means you can add a source that contains movies that you want scrape with english informations and poster, another source for movies that you want scrape with german informations and posters. The most of all users don't really need this system, but now you no longer have to set all scrapers to a or the same preferred language.
Thanks Dan I am not sure whether I 100% understand it yet, but let me just wait and try out. What you are writing gives me confidence you thought it all through and it will be great!
I noticed some other things that may be a "bug" or may just me not being able to find the functionality:
1) I have set movie extrafanart and extrathumbs to "0", but it is still downloading those. Possible to disable?
2) Downloaded extranfanart are in a formart like "283FjtlcWjrAmlGAde0BhBKDVyb.jpg", whileas extrathumbs are nicely numbered. Is this intentional?
3) Actors are being downloaded into ".artists" folder. Can I disable this or is it reqiuired?
Thanks in advance!
All those settings are in File and sources
Uncheck them and nothing is downloaded
(2015-09-20, 04:10)steve1977 Wrote: [ -> ]Thanks Dan I am not sure whether I 100% understand it yet, but let me just wait and try out. What you are writing gives me confidence you thought it all through and it will be great!
I noticed some other things that may be a "bug" or may just me not being able to find the functionality:
1) I have set movie extrafanart and extrathumbs to "0", but it is still downloading those. Possible to disable?
2) Downloaded extranfanart are in a formart like "283FjtlcWjrAmlGAde0BhBKDVyb.jpg", whileas extrathumbs are nicely numbered. Is this intentional?
3) Actors are being downloaded into ".artists" folder. Can I disable this or is it reqiuired?
Thanks in advance!
1 - You have set the Limit to 0, that means unlimited. As T-bird_se has sayed, you have to enable or disable that thru File and Sources settings.
2 - We use the original file name for extrafanarts to prevent dublicates. For Extrafanarts it's not needed to have a numbered image, but maybe we add an option to number Extrafanarts like Extrathumbs.
3 - You can disable that also in Files and Sources settings. But if disabled, Kodi downloads them while DB update and use the URL in NFO for that. That needs more time in Kodi to import all movies.
Thanks, very clear. I have set Kodi to "local info only", so it will not download the actorthumbs. At least I hope so? All other is clear and works.
Have one more problem: I just downloaded NFO-only via custom-scraper (set to skip if cannot identified). Unfortunately, it sets the TMDB-ID to "0" in case the movie is not identified. This makes it impossible for me to spot what movies have no TMDB-ID. Is there any way that I can filter for TMDB-ID=0 and what is the reason for setting it to 0?
Hello, quick question, where can I download stand-olone version? without installation. I cannot find it.