[Nhoise]

I'm using FileBot to auto rename my folders/files to fit a certain naming format. Examples:

Code:

H:\Gemma Bovery (2014)\Gemma.Bovery.2014.[1920x800].24.000fps.1080p.BRRip.x264.JYK.mkv
H:\In The Realm Of The Senses (1976) - 愛のコリーダ\Ai.No.Korīda.1976.[1280x772].23.976fps.720p.x264.CiNEFiLE.mkv

But when using TMM it's detecting the movies as

Code:

Gemma Bovery 1920x800 24 000fps
In The Realm Of The Senses □□□□□□

So whenever i have it scrape data i have to manually remove the '1920x800 24 000fps & foreign name' parts from the name or else it will come up with no results.
Is there something that can be done to avoid this? (Besides switching to a different naming format)
Guessing this has something to do with the way TMM gets the name of the movie to search for.

I'm using the latest 2.6.8 (r2052) version of TMM.


edit: I've also noticed it's having issues with certain characters in a file name.
When there is either an é or anything with the ' (i.e. You're) it shows no results until you delete that character.
Might be more but these are the ones i've run into.

[mlaggner]

Detecting the movie title out of the file-/foldername is really hard. We've written a complex detection engine which works for most test cases we have - but we cannot strip out too much (to avoid false positives. removing too much is worse than removing not enough)

regarding the scraping problem: we have to get rid of several special characters when calling the scraper API, because it is sometimes not supported (the worst scraper is thetvdb..). In this case there is no other solution than changing the search term. We are only a "middleware" between the user(library) and the meta data providers (like tmdb or imdb)

[Nhoise]

The name detection is not really a problem tbh, it's not like i'm having TMM scrape 100's of movies at once.
And I like to manually select the titles anyway to be sure it's all correct so having to remove a bit of text is not a big deal.
I have to say though; the name detection worked better for me before the 2.6.8 update. (I think i was still on 2.6.6)

And regarding the special characters; Can't you just have a rule set which automatically replaces characters like the é with an e etc.
I feel like that would be an easy fix, wouldn't it?

[myron]

sry for late reply - weather kinda fine, TMM team on vacation/afk ^^

I tried to tweak the naming detection once again, we might detect now better resolutons & framerates.
Let's see, how it goes in next version

The parser in 2.6.6 had an error - it always took the foldername for it's detection.
This worked better in your case, but didn't work at all for others...
with 2.6.8 we're now taking the longer name (after all our cleaning)


No, there won't be a setting to replace such special characters.
It doesn't make sense here, since one site needs it, and another one not.
We already have such replacement in place (internally) when searching on _some_ sites, we just need to review this...

br
Myron