Jump to content

Plugin - Chapter Editor (ChapterApi)


TeamB

Recommended Posts

rbjtech
15 hours ago, crusher11 said:

I've had to add a second NAS, due to running out of storage space. Now I have a show with some episodes on one NAS, and other episodes on the other NAS. The ChapterAPI UI just lists the show name twice, instead of merging them together, which means I have to guess which one I want each time.

I would.not want them merged as the same happens for multi version.  I wandt to seethem individually. Its because it uses multiple MediaSource ids but same ItemIds.     The path shows the version you have selected.

Why not just move the series to the same storage ?

Edited by rbjtech
Link to comment
Share on other sites

crusher11

Because neither storage is big enough for all of it. And I'm unsure how to retain IntroSkip data if I were to do that.

Surely the plugin could merge shows and seasons without merging episodes? 

Link to comment
Share on other sites

Happy2Play

But the UI is not by Library it is by Path.  So it requires different locations.

Link to comment
Share on other sites

rbjtech
11 hours ago, crusher11 said:

Because neither storage is big enough for all of it. And I'm unsure how to retain IntroSkip data if I were to do that.

Surely the plugin could merge shows and seasons without merging episodes? 

 

My point was to move all of that 'series/show' to the same root - ie rather than have series 1-10 on one 'nas' and series 11-20 on another - you move it all to the new nas

Multiple storage locations for the same root source are generally an unwanted complexity.    

 

@TeamB Having said that - It would be good if the search/list could give some indication of the path before you click them.

In the example below - if I search for Obi-Wan - both show, but I've no idea which is the HD and which is the 4K version (both have seperate folders/paths).

This is shown when you click one (in the path) but to then go back to the other one - you have to re-search it again.

 

I think this would help crusher as well - by showing the 'path' of each item ?

 

Thanks,

Rich.

 

image.png.b8340594a149a80f536ce298615b0c1f.png

Edited by rbjtech
Link to comment
Share on other sites

crusher11
1 hour ago, rbjtech said:

 

My point was to move all of that 'series/show' to the same root - ie rather than have series 1-10 on one 'nas' and series 11-20 on another - you move it all to the new nas

Multiple storage locations for the same root source are generally an unwanted complexity.   

It'd be nice if I could afford that much storage, but I can't.

Link to comment
Share on other sites

rbjtech
55 minutes ago, crusher11 said:

It'd be nice if I could afford that much storage, but I can't.

copy \\nas1\shared\show\*.* \\nas2\shared\show

when verified - then remove the show from the source nas.  Continue to use that for existing files with the new space you just created - access the entire moved show on nas2 only from now on now ..

Am I missing something ?

Edited by rbjtech
Link to comment
Share on other sites

3 hours ago, rbjtech said:

I think I withdrew the idea of using multi-version again after reading the post of ebr:

If I got it right then it basically makes it unusable for TV shows as I cannot make sure that 4K and 1080p is 1:1 exactly the same and therefore the intros wouldn't work properly.
 

Link to comment
Share on other sites

rbjtech
11 hours ago, neik said:

I think I withdrew the idea of using multi-version again after reading the post of ebr:

If I got it right then it basically makes it unusable for TV shows as I cannot make sure that 4K and 1080p is 1:1 exactly the same and therefore the intros wouldn't work properly.
 

ok - I guess for recorded TV shows with Ads, then maybe - but for other content - it should be identical - I've never seen a 1080p differ in content for a 4K version for example.

Edited by rbjtech
Link to comment
Share on other sites

TeamB
25 minutes ago, neik said:

TeamB, as of now I think this is not applicable to End Credits, would it be possible/make sense to expand this for end credits?

you can manually add end credits chapter markers.

it does not auto detect end credits and there is no plan to add that.

Link to comment
Share on other sites

rbjtech
9 hours ago, TeamB said:

you can manually add end credits chapter markers.

it does not auto detect end credits and there is no plan to add that.

If you previously used the Introskip plugin - then this plugin is fully compatible with those EndCredits (markertype=EndCreditStart).

The credits only version of the introskip plugin is also fully compatible - this is not yet released but is in development by @Cheesegeezer .  It's effectively a re-work of the original intro-skip plugin, minus the Intro detection but includes the 'chapter' functions from the original plugin 👍

Can't wait to use it myself tbh !!

Link to comment
Share on other sites

seanbuff

Hi All,

I finally got around to installing and using this tool, as the previous IntroSkip Plugin unexpectedly stopped working for me recently 😕

@TeamB great tool, and an extremely useful addition to the core IntroSkip feature.

So reading thru the thread I understand that any changes to chapters are wiped if a metadata refresh is performed on any of the modified items. However what I'm seeing is that all of my edits (necessary edits due to incorrect detection) are being wiped out and re-detected on just the normal scheduled library scan. Is my understanding of a library scan wrong as I thought there was a difference between scan and metadata refresh?

How are others making chapters edits and retaining those changes after a library scan is run again every 12 hours? Surely I'm doing something wrong.

TIA.

Link to comment
Share on other sites

rbjtech

The history The introskip plugin stored its detection/intro results in a separate db and if you ran the chapter task, it would add these to the core chapters.   However, on a metadata refresh, as you has ascertained, would wipe the intro chapters, needing you to re-run tje introskip chapters task to put them back.

The good news is cheese has written a new plugin to backup the plugin  introskip data from the db in json and/or in the nfo if you use them.  This can be restored, but its currently an all or nothing operation.

Cheese is also writing (or has written) a version of the introskip plugin minus the intro detection.. leavimg that to the core.    

Ultimately its up to the core to maintain the intro data now following a metadata refresh,  but it doesn't do that today so you need to use the plugins to work around it.

Its messy,  tbh not sure why an editor and saving of metadata in the core was omitted from the original core release ... 😒

Btw - a library scan nor a metadata scan should be removing edits .. only a metadata refresh should be doing this.

Edited by rbjtech
  • Agree 1
Link to comment
Share on other sites

seanbuff

Thanks for all info Rich, I think i'm clear on that now, and I'm okay with it for the time being.

12 minutes ago, rbjtech said:

Btw - a library scan nor a metadata scan should be removing edits .. only a metadata refresh should be doing this.

However, its this statement here that has caused the confusion I guess. I wasn't expecting a standard library scan to remove my edits either.

Last night I made edits, and can see they "Saved" successfully:

Quote

2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApi - ChapterApiEndpoint Loaded
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 1 Chapter 0
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 2 Chapter 3000000000
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Intro Start IntroStart 5539800995
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Intro End IntroEnd 5749999999
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 3 Chapter 6000000000
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 4 Chapter 9000000000
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 5 Chapter 12000000000
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 6 Chapter 15000000000
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 7 Chapter 18000000000
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 8 Chapter 21000000000
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 9 Chapter 24000000000
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 10 Chapter 27000000000
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapter - Chapter 11 Chapter 30000000000
2022-11-24 23:02:04.878 Info ChapterApi - ChapterApiEndpoint: ChapterApiEndpoint - Chapters Saved
2022-11-24 23:02:04.889 Info ChapterApi - ChapterApiEndpoint: ChapterApi - ChapterApiEndpoint Loaded


Then the following morning, during a typical scheduled Library Scan, IntroSkip Core is seen "detecting" the same (incorrect) intro chapters again, basically replacing what I just removed.

2022-11-25 08:01:50.676 Info App: Detecting markers for \\NAS\TV Shows\Dexter - New Blood\Season 1 - episode \\NAS\TV Shows\Dexter - New Blood\Season 1\Dexter - New Blood - S01E01 - Cold Snap WEBRip-720p.mkv
2022-11-25 08:01:50.711 Info App: \\NAS\TV Shows\Dexter - New Blood\Season 1 has 10 episodes with audio fingerprints available
2022-11-25 08:01:51.170 Info App: Best title sequence for \\NAS\TV Shows\Dexter - New Blood\Season 1\Dexter - New Blood - S01E01 - Cold Snap WEBRip-720p.mkv is 00:04:23.3084577 - 00:04:56.6417910. Confidence: 1
2022-11-25 08:01:51.171 Info App: Detecting markers for \\NAS\TV Shows\Dexter - New Blood\Season 1 - episode \\NAS\TV Shows\Dexter - New Blood\Season 1\Dexter - New Blood - S01E02 - Storm of F*** WEBRip-720p.mkv
...
2022-11-25 08:01:51.627 Info App: Best title sequence for \\NAS\TV Shows\Dexter - New Blood\Season 1\Dexter - New Blood - S01E02 - Storm of F*** WEBRip-720p.mkv is 00:09:13.9800995 - 00:09:34.9999999. Confidence: 0.9


That's not right is it?

Link to comment
Share on other sites

rbjtech

Right ok.. so if the show is being re-detected - then i can see why it would overwrite any edits because it does the entire season again, not individual episodes for each new episode added.   Once the show/season has finished,  then any edits will be safe as it should not be detected again.

But its a good call .. and again something not properly baked into the core chapter metadata.. on an edit, it should be locked like any other manually edited data.   Something for the core team I guess.. 

Link to comment
Share on other sites

seanbuff
1 hour ago, rbjtech said:

Once the show/season has finished,  then any edits will be safe as it should not be detected again.

Well that's the thing, the season and series itself have ended a while back so there's no reason it should keep trying to detect anything. But obviously Emby thinks it needs to because I'd removed previous detected intros.

@Lukehow do we prevent this from happening? And am I the only one seeing this since no one else has raised it?

Link to comment
Share on other sites

rbjtech

So I have seen shows reprocessed on my system before, and it turned out that it was stuck on an episode where a correct ffmpeg/ffprobe was not functioning correctly.  I fixed the source file and once re-detected - ie a 100% all episodes detected - then it stopped trying to re-process that show.

I have asked Luke the same question on 'how' it knows that it has processed that show/episode, the answer I got is that it uses the presence of the Hidden Marker types to determine that.

So if you are literally deleting the hiddden markers (IntroStart and IntroEnd) then that is probably why it is re-processing them - BUT if an Intro simply does not exist - then I'm not 100% sure if the markers should have NULL values - or set to 0:00 etc.   

I'll have a look later to see on a show with no intro's - what the hidden markers are set to.

  • Thanks 1
Link to comment
Share on other sites

rbjtech

so just a follow up on this - for shows with no intro's - they do not have any hidden markers set - and yet they do not get re-scanned each time.

Thus there must be another mechanism to indicate they have been 'processed' - I've got a feeling it's the fingerprint files ... I'll keep digging..

edit - yep - I have 27 thousand tiny .fp files held in \metadata\library\  :(

No idea why they are all being kept but I seem to recall this is how the Core Introskip works .. I guess Luke can expand on it.

The Introskip plugin kept the fingerprints in a single database file, had a simple 'processed' flag in the database for each item and also cleaned up after itself when the show was completed - by removing all the fingerprints...

Edited by rbjtech
Link to comment
Share on other sites

seanbuff
12 hours ago, rbjtech said:

So I have seen shows reprocessed on my system before, and it turned out that it was stuck on an episode where a correct ffmpeg/ffprobe was not functioning correctly.  I fixed the source file and once re-detected - ie a 100% all episodes detected - then it stopped trying to re-process that show.

Yeah I don't think that's whats happening in this case, I don't see any issues with the file being probed, etc.

I am literally trying to remove the false positives that have been detected, presumably because they all share a similar "ominous music" audio fingerprint
image.png.e21a8c8eb8bf149ad88fac80c0d2ae70.png

Is that not possible?

@Luke any thoughts? How are we expected to rely on this feature when we can't even fix its mistakes?

Link to comment
Share on other sites

2 hours ago, seanbuff said:

Yeah I don't think that's whats happening in this case, I don't see any issues with the file being probed, etc.

I am literally trying to remove the false positives that have been detected, presumably because they all share a similar "ominous music" audio fingerprint
image.png.e21a8c8eb8bf149ad88fac80c0d2ae70.png

Is that not possible?

@Luke any thoughts? How are we expected to rely on this feature when we can't even fix its mistakes?

Editing features are planned for future updates. Thanks.

Link to comment
Share on other sites

On 11/14/2022 at 9:39 AM, rbjtech said:

ok - I guess for recorded TV shows with Ads, then maybe - but for other content - it should be identical - I've never seen a 1080p differ in content for a 4K version for example.

I just got to do some basic check (is runtime equal for 1080p / 4K) on my library.
While you're right for the majority of the files I still got a significant amount of files that differ a bit in runtime.
Very often it's just a minute sometimes two minutes difference though.

Before merging the two libraries I think I will try to merge some of those different versions and see what happens...

Link to comment
Share on other sites

  • 3 months later...
Bvaring

This s a useful program.  Thank you.  Buy why doesn't it show the pesky chapters that are created in a Music Video library.  It would great to edit those.

Link to comment
Share on other sites

  • 4 weeks later...
jaketame

Whats the chances of being able to import plex markers if we can get these exported in a format ready for Emby to understand?

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...