[vbat99]

v3.6.9.3 (x64)
Windows 10 Enterprise Edition

First, thanks for writing this program, I'm using it to build NFO files for later KODI import.
I organize my media in a way that KODI and most other media managers have trouble with and at least with change movie I can fix the ones it can't figure out.   Out of the ones I've tried, yours gets it right more often than the rest.

My general structure is:

Code:

GENRE/GROUP (or) TITLE/YEAR-TITLE

(e.g. Comedy/Cabin Boy/1994-Cabin Boy)

There can be modifiers after TITLE like (Subtitled) etc.

For some films that need a 'forced order' (like the Star Wars) films, it gets more complex: 

Code:

#-YEAR-TITLE

(e.g. 04-1977-A New Hope)


Issues:

• I have a lot of movies I've amassed over the years that I've converted from DVD to MP4 files.    I've noticed if there's more than around 200 films in a given folder a couple things happen.  1) Sometimes it will assign a random imdbid to a few dozen films   2) It will just do the "This movie could not be identified" entry.
  
• If I use any batch rescrape options, regardless of settings, it will remove any existing art.    So I can't use the batch rescrape options.   Instead I have to exit MediaCompanion, delete all the "This movie could not be identified" NFOs (I wrote a perl script to do this), delete all of MediaCompanion's cache files and then have it look for new movies.    ( For the case where it assigns some random imdbid, I have an option in the same perl script to remove those as well. )
  
• It appears to not be scraping the IMDB plot keywords at all.
  

Requests:

• I'd like to request that there be an option that if it can't identify a film to just not make the NFO at all.   
  
• I'd like to request an option to disable Media Companion's cache and database entirely.   I'm just using this to build NFO files, not as a manager.
  
• I'd like to see options for doing the export CSV as to what fields to include and what to exclude.
  
• I'd like to see an option to scrape ALL certifications for a film but only display the one selected as the primarily in the movie preferences (and if that isn't available, the next in the priority list).
  
• A useful option would be to grab all runtimes for a given film (to identify extended, director's cuts, etc) and for bonus points to compare that to the local media's length (to see what version you've got) and also to alert if the runtime does not match the scrape source (i.e. cut version, etc).    (Would need to also note the media's frame rate, often for PAL films are just sped up by 4%).
  

Gidday Lee

That is an impressive arrangement, and I am glad you've found MC can handle the majority of it for you.
A concern that rises is having the year at the beginning of the filename. Now, if you set MC to use Foldername for Scraping, and have the year at the end of the folder name, ie: [/b]A New Hope (1977)[/b], you can still have the same structure for your filename as you currently have.
Where having the year at the beginning fails is with movies like 2001: A Space Odyssey, and other movies like this.

The norm is year of the movie behind the title, and in brackets. This ensures that the year is found easily, and does not cause issues with movies like 2001: A Space Odyssey etc.

Now, in regards to your issues:

I've noticed if there's more than around 200 films in a given folder a couple things happen. 1) Sometimes it will assign a random imdbid to a few dozen films 2) It will just do the "This movie could not be identified" entry.

IMDb scraping is not great. In fact, after many scrapes from IMDb, their site assumes MC's calls to their page is a Bot-Attack. We have tried to get around this, but it is hit and miss on how many successful scrapes are done before IMDB defaults to a non-descript page. This is where MC starts getting Random IMDBId's returned and incorrect movies are scraped, till the This movie could not be identified occurs.
It is recommended to scrape with TMDb instead, but the naming convention you have currently may cause further issues.

If I use any batch rescrape options, regardless of settings, it will remove any existing art. Instead I have to exit MediaCompanion, delete all the "This movie could not be identified" NFOs

You can't batch rescrape Not identified movies, as the core data, IMDBId etc is incorrect or invalid. Batch rescrape is to update data, or try to get missing data, for Valid movies in Media Companion.

It appears to not be scraping the IMDB plot keywords at all.

Correct. Seems IMDb has made some changes, I'll get that fixed for the next release.

Your Requests

• 1. I'd like to request that there be an option that if it can't identify a film to just not make the NFO at all.

I can add an option to disable saving of these nfo's, but to be fair, how will you know that a movies failed to scrape or was unidentifiable?

• 2. I'd like to request an option to disable Media Companion's cache and database entirely. I'm just using this to build NFO files, not as a manager.

This function is available already, though it will scan every movie root drive for valid nfo's on start of Media Companion, it will not save this database of close. Go Preferences -> General tab. Check option Disable caching of Media DB (will slow down startup).

• 3. I'd like to see options for doing the export CSV as to what fields to include and what to exclude.

Do you mean Export Movie List Info -> and select a template? These templates are for the user to modify as they wish. To create the option to build a Custom template in Media Companion would be very time intensive, and would be at the bottom of the feature list. Check the folder html_templates and look at the file 1Export Template Guide.txt in this folder for help in building your own template, or modifying an existing template.

• 4. I'd like to see an option to scrape ALL certifications for a film but only display the one selected as the primarily in the movie preferences (and if that isn't available, the next in the priority list).

Don't feel this is going to happen. For one, all certificates would have to be stored in the movie's nfo. Two, Kodi only reads one Certificate. Three, If more that one certificate, Kodi uses the last one found.

• 5. A useful option would be to grab all runtimes for a given film (to identify extended, director's cuts, etc) and for bonus points to compare that to the local media's length (to see what version you've got) and also to alert if the runtime does not match the scrape source (i.e. cut version, etc). (Would need to also note the media's frame rate, often for PAL films are just sped up by 4%).

Ar, nope.
Here's the runtimes for LOTR: Fellowship

Runtime: 178 min | 208 min (Special DVD Extended Edition) | 228 min (Blu Ray Extended Edition) | 171 min (DVD Widescreen Edition)

And here they are for The Hobbit: Unexpected Journey

Runtime: 169 min | 182 min (Extended Edition)

Spiderman-3

Runtime: 139 min | 137 min (editor's cut)

There is insufficient consistency in naming of the runtimes, for MC to pick the data.
Not to mention how/where to hold that data, and what it means. And then there's downloaded movies where start or credits may be cut, or extra added.
Sorry, #5 won't be happening.
Oh and BTW, Pal runs at 25 frames a Second. NTSC can be 23.97 or 29.97 frames a second. Runtime on Pal or NTSC are the same seconds long, it just how many frames are played in that second.