Jump to content

Show Intro Skip Option (Plugin)


Liquidfire88

Recommended Posts

Sarakha63
6 minutes ago, chef said:

After you configured the edited timestamps, did you hit the 'save' button?

 

Currently when the reset button is hit, it indeed does remove even edited content.

We use to try and save items that had been edited by the user, but the code was becoming very complex, and buggy.

So, we went back to basics. However, perhaps we had best try to recode the ability to keep edited content.

 

Let's see what we can do.

I can confirm that i clicked the save button. But they were still the old timings in the chapters
i can also confirm that afer a reset they are still there.

i had to refresh metadata on the episode than launch the intro skip chapter task for them to upgrade

Link to comment
Share on other sites

guardianali

Question on methodology. 

Plex has had this feature for a while now and it works pretty decent. I mainly use Emby and love it hence the lifetime membership. But still use Plex for a few things and run them side by side. 
For plex, there was no real plug in or running a service to do analysis locally, etc... one day after a server upgrade, the button to skip just showed up. 
I believe their code is still mostly GPL3 and thus reviewable. Anyone taken a peek to see how they implemented theirs?

Not that this effort here isn't appreciated. It definitely is. The above is more a suggestion to save the team here time maybe. 

  • Like 1
Link to comment
Share on other sites

Cheesegeezer
1 hour ago, Sarakha63 said:

I ignored all TVshows and i left only one to start with. Everything seems fine i will now go further

i just had an idea (feature request) : give the ability to custom the chapter names for Title and Credits 

This will not happen as I will be  keeping standardisation across the lifecycle of the app.  if the core devs wish to implement localisation for strings in chapters for this, that is something they will have to do.

 

28 minutes ago, Sarakha63 said:

I can confirm that i clicked the save button. But they were still the old timings in the chapters
i can also confirm that afer a reset they are still there.

i had to refresh metadata on the episode than launch the intro skip chapter task for them to upgrade

yes, if you update, or change any timings, it is required to refresh metadata (not images) as this will then reset and inject the new chapter points.

Link to comment
Share on other sites

Sarakha63
2 minutes ago, Cheesegeezer said:

This will not happen as I will be  keeping standardisation across the lifecycle of the app.  if the core devs wish to implement localisation for strings in chapters for this, that is something they will have to do.

 

yes, if you update, or change any timings, it is required to refresh metadata (not images) as this will then reset and inject the new chapter points.

Could we imagine that upon saving a custom timing il will trigger this in background ?

Link to comment
Share on other sites

Cheesegeezer
3 minutes ago, guardianali said:

Question on methodology. 

Plex has had this feature for a while now and it works pretty decent. I mainly use Emby and love it hence the lifetime membership. But still use Plex for a few things and run them side by side. 
For plex, there was no real plug in or running a service to do analysis locally, etc... one day after a server upgrade, the button to skip just showed up. 
I believe their code is still mostly GPL3 and thus reviewable. Anyone taken a peek to see how they implemented theirs?

Not that this effort here isn't appreciated. It definitely is. The above is more a suggestion to save the team here time maybe. 

Hey there,

our code is very different to theirs and you would still have had scan time for the chromaprinting in plex, which we understand can take weeks even and that is just for their intro skip.  ours is much more efficient and effective. I also believe that their code is 'borrowed' from another development team.  

Are you having issues with the plugin, have you run it.  Interested for your feedback.

Many Thanks

Link to comment
Share on other sites

Cheesegeezer
3 minutes ago, Sarakha63 said:

Could we imagine that upon saving a custom timing il will trigger this in background ?

We have tried this, however, it is causing some undesired results, but trust me, we are working on it.  It's not that we haven't thought about it.

Link to comment
Share on other sites

Sarakha63
4 minutes ago, Cheesegeezer said:

We have tried this, however, it is causing some undesired results, but trust me, we are working on it.  It's not that we haven't thought about it.

Great you are right... Each things at a time. What you have accomplished is already very promising. As i'm a new beta tester... and i'm also a developper.. i really see  the work behind.

and 'im trying to spot any  improvements or bug to help

I have a question about the thumbnails that we can see inside the plugin itself. Are they requested via ffmpeg each time we get on the page... On my point of view it looks like it is ?

Edited by Sarakha63
  • Like 2
Link to comment
Share on other sites

rbjtech
14 minutes ago, guardianali said:

Question on methodology. 

Plex has had this feature for a while now and it works pretty decent. I mainly use Emby and love it hence the lifetime membership. But still use Plex for a few things and run them side by side. 
For plex, there was no real plug in or running a service to do analysis locally, etc... one day after a server upgrade, the button to skip just showed up. 
I believe their code is still mostly GPL3 and thus reviewable. Anyone taken a peek to see how they implemented theirs?

Not that this effort here isn't appreciated. It definitely is. The above is more a suggestion to save the team here time maybe. 

If you run both Emby and Plex on the same media - I think it would be a really interesting exercise to compare the matches on some selected TV series and post the results.

Does Plex give you the ability to see/review/change the actual Intro Start and End ? 

Link to comment
Share on other sites

Cheesegeezer
5 minutes ago, Sarakha63 said:

Great you are right... Each things at a time. What you have accomplished is already very promising. As i'm a new beta tester... and i'm also a developper.. i really see  the work behind.

and 'im trying to spot any  improvements or bug to help

Thanks you it's appreciated

5 minutes ago, Sarakha63 said:

I have a question about the thumbnails that we can see inside the plugin itself. Are they requested via ffmpeg each time we get on the page... On my point of view it looks like it is ?

yes they are, we will be implementing an advanced feature slider to cache the images locally, they are only 320px x 180px (IIRC) not large, but the reason why we request them is because for me with about 4000 episodes it takes up 5GB of disk space, however for my friend who has in the region of 30000 it would take up 40GB of disk space to cache them. 

Some people would be upset if we stole that much real estate from them.  but it has been request from people using NAS as sometime the images time out and are not displayed at all because of the lack of CPU power

Edited by Cheesegeezer
  • Like 1
Link to comment
Share on other sites

rbjtech
11 minutes ago, Sarakha63 said:

I have a question about the thumbnails that we can see inside the plugin itself. Are they requested via ffmpeg each time we get on the page... On my point of view it looks like it is ?

yes they are generated on demand.

My personal view is there is little point generating them in advance, as you would generally only go to this page to review things if they were not accurate in the first place.

ie - out of 100 seasons, you may want to look at 1 season with 12 episodes - that's 36 images we need to 'create' - that may take 1-2 seconds.  That is much better than pre-generating (and storing) 1000's of images just to cache them.

As a  real life example - the shows that my users and I have been watching over the last couple of days with the Automatic skip on (yet to be released), I have had zero reason to even look at the Plugin/Intro's page - as it simply 'just works' .. 😎 .. all skip points were perfect.

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

Sarakha63
8 minutes ago, Cheesegeezer said:

Thanks you it's appreciated

yes they are, we will be implementing an advanced feature slider to cache the images locally, they are only 320px x 180px (IIRC) not large, but the reason why we request them is because for me with about 4000 episodes it takes up 5GB of disk space, however for my friend who has in the region of 30000 it would take up 40GB of disk space to cache them. 

Some people would be upset if we stole that much real estate from them.  but it has been request from people using NAS as sometime the images time out and are not displayed at all because of the lack of CPU power

Ok that what i was guessing. and i totaly agree with your choice as it is only "an administrative view"

 

But may i suggest an UX idea. A toggle switch on top of the table "by default off" and on toggle it will load the image if needed

with that navigating trough show and seasons will be realy fast and most of the time seeing the times/and the true false is enough. and if for a specific reason man would like to see the thumbnail it will click on the toggle and hop the thumbnails will be requested.


I propose that because some show may have 100th of episode in one season and even with my core i7 11th gen i may notice the delay and the lag when navigating trough seasons and shows to check waht have been done

just an idea

 

 

PS : excuse my english.. i'm french

Edited by Sarakha63
Link to comment
Share on other sites

Sarakha63
6 minutes ago, rbjtech said:

yes they are generated on demand.

My personal view is there is little point generating them in advance, as you would generally only go to this page to review things if they were not accurate in the first place.

ie - out of 100 seasons, you may want to look at 1 season with 12 episodes - that's 36 images we need to 'create' - that may take 1-2 seconds.  That is much better than pre-generating (and storing) 1000's of images just to cache them.

totally agree with you. But i may have a simple UX idea i mentionned above

Link to comment
Share on other sites

rbjtech
6 minutes ago, Sarakha63 said:

Ok that what i was guessing. and i totaly agree with your choice as it is only "an administrative view"

 

But may i suggest an UX idea. A toggle switch on top of the table "by default off" and on toggle it will load the image if needed

with that navigating trough show and seasons will be realy fast and most of the time seeing the times/and the true false is enough. and if for a specific reason man would like to see the thumbnail it will click on the toggle and hop the thumbnails will be requested.


I propose that because some show may have 100th of episode in one season and even with my core i7 11th gen i may notice the delay and the lag when navigating trough seasons and shows to check waht have been done

just an idea

 

 

PS : excuse my english.. i'm french

Your English is perfect ;)

I think this is a great idea - we would probably keep the image placement, and just use a local image instead like we do when there is no Intro anyway.  @chef / @Cheesegeezer - thoughts ?

Edited by rbjtech
Link to comment
Share on other sites

27 minutes ago, guardianali said:

Question on methodology. 

Plex has had this feature for a while now and it works pretty decent. I mainly use Emby and love it hence the lifetime membership. But still use Plex for a few things and run them side by side. 
For plex, there was no real plug in or running a service to do analysis locally, etc... one day after a server upgrade, the button to skip just showed up. 
I believe their code is still mostly GPL3 and thus reviewable. Anyone taken a peek to see how they implemented theirs?

Not that this effort here isn't appreciated. It definitely is. The above is more a suggestion to save the team here time maybe. 

Interesting. 

I have read quite a few different reddit posts about Plex using the intro skip, and there was quite a few people complaining about the length of the scans.

It makes me wonder if, over time,  they compiled a database for themselves  based on Users data, and scans.

I have often wondered if they did that, and I'm becoming more convinced that that is what they did. Perhaps, we should do the same. Obviously an opt'in feature... 

But, I have seen enough logs and intro data, from our users, to say with some certainty that there is enough comparison in the data between user encoding,  that a database could be compiled as an online resource.. in the future. Just sayin'...

 

 

Link to comment
Share on other sites

Sarakha63
Just now, rbjtech said:

Your English is perfect ;)

I think this is a great idea - we would probably keep the images, and just use a local image instead like we do when there is no Intro anyway.  @chef / @Cheesegeezer - thoughts ?

yes a kind of temporary image. And the toggle on demand will request ffmpeg at the time.

as i guess 90% of the times you want to check the timings and dont need the thumbs this will improve the user experience navigating. and if for a specific reason you want to get the thumbs you ask for them with a simple click

Link to comment
Share on other sites

2 hours ago, Sarakha63 said:

 

request : for now we are able to add shows in ignore list but the droplist doesnt take into account the ignored one. it is always full. this results in this behavior

1) we can add multiple times (infinite times) the same show.

 

 

I have fixed this for an upcoming release.

Thank you @Sarakha63.

I'm moving through some more of your observations. :)

Edited by chef
  • Like 1
Link to comment
Share on other sites

Sarakha63
3 minutes ago, chef said:

Interesting. 

I have read quite a few different reddit posts about Plex using the intro skip, and there was quite a few people complaining about the length of the scans.

It makes me wonder if, over time,  they compiled a database for themselves  based on Users data, and scans.

I have often wondered if they did that, and I'm becoming more convinced that that is what they did. Perhaps, we should do the same. Obviously an opt'in feature... 

But, I have seen enough logs and intro data, from our users, to say with some certainty that there is enough comparison in the data between user encoding,  that a database could be compiled as an online resource.. in the future. Just sayin'...

 

 

totally a great idea. an option with "would i want to share my timings" and a save with a md5 hash and the timings.

and then if found on database no need to trigger the analysis anymore

 

It wont take long to have a quite consistent database

Edited by Sarakha63
Link to comment
Share on other sites

2 minutes ago, Sarakha63 said:

yes a kind of temporary image. And the toggle on demand will request ffmpeg at the time.

as i guess 90% of the times you want to check the timings and dont need the thumbs this will improve the user experience navigating. and if for a specific reason you want to get the thumbs you ask for them with a simple click

Interesting idea. That would most certainly help with our load time.

Note taken. 

 

Thank you @Sarakha63.

If you notice anything else. Please let us know. :)

Link to comment
Share on other sites

Sarakha63
Just now, chef said:

Interesting idea. That would most certainly help with our load time.

Note taken. 

 

Thank you @Sarakha63.

If you notice anything else. Please let us know. :)

i will ;)

 

Thank you all of you for your listening ears ;)

Link to comment
Share on other sites

rbjtech

We had this debate very early on in the project - the issue is there may be many sources of the file in question - the only thing which you can quickly tell is which version you have based on the duration of the file.  ie, one with a 'recap' will be longer than one without.  However, a 'TV' version may be different to a 'DVD/BluRay' version.  Recorded versions will be all sort of lengths - some record 'N*****x' at the start, others don't.     If you need to analyse all of that to see what 'version' you have, then you may as well Fingerprint/Chromaprint it in the first place .. ;)   

Link to comment
Share on other sites

Sarakha63
3 minutes ago, rbjtech said:

We had this debate very early on in the project - the issue is there may be many sources of the file in question - the only thing which you can quickly tell is which version you have based on the duration of the file.  ie, one with a 'recap' will be longer than one without.  However, a 'TV' version may be different to a 'DVD/BluRay' version.  Recorded versions will be all sort of lengths - some record 'N*****x' at the start, others don't.     If you need to analyse all of that to see what 'version' you have, then you may as well Fingerprint/Chromaprint it in the first place .. ;)   

you can be sure with a md5sum of the file (the only thing that can be stored in the database by the way)

the question would be how long would it take to make the md5sum of the file comparing to the chromaprint and the detection. to see if i could be worht it or not

Link to comment
Share on other sites

15 minutes ago, rbjtech said:

We had this debate very early on in the project - the issue is there may be many sources of the file in question - the only thing which you can quickly tell is which version you have based on the duration of the file.  ie, one with a 'recap' will be longer than one without.  However, a 'TV' version may be different to a 'DVD/BluRay' version.  Recorded versions will be all sort of lengths - some record 'N*****x' at the start, others don't.     If you need to analyse all of that to see what 'version' you have, then you may as well Fingerprint/Chromaprint it in the first place .. ;)   

This is true different encoding do cause discrepancies.

Do you think Plex did that though? If users are saying that, all of a sudden one day a skip intro button appeared and it worked, then I'm wondering if they created a database with a whole bunch of data (like we do) and then created a weighting system (like we do) to find the best results for each kind of encoding.

It just makes me wonder if there is a private resource out there, and if Plex users were aware of their data being used to create it, and (if it's true) if they were charged a premium fee for the use of it. Ya know?

There could have very well been small print somewhere in the install that allowed the company to extract that data initially, and then use it to build a feature, that they then charged their clients for.

Obviously I don't know for certain, but...

😳

Edited by chef
  • Haha 1
Link to comment
Share on other sites

rbjtech
5 minutes ago, Sarakha63 said:

you can be sure with a md5sum of the file (the only thing that can be stored in the database by the way)

the question would be how long would it take to make the md5sum of the file comparing to the chromaprint and the detection. to see if i could be worht it or not

Sure - I fully get you - it's effectively a crowd sourced 'anonymous' db of MD5 hashes with an intro Start/End time - if you query it and get a match with your file, then you have a 100% hit and can move onto the next episode.

The questions start arising when creating the dB in the first place - who has authority to say that MD5 for that episode is 'correct' ?

Edited by rbjtech
Link to comment
Share on other sites

Sarakha63
5 minutes ago, rbjtech said:

Sure - I fully get you - it's effectively a crowd sourced 'anonymous' db of MD5 hashes with an intro Start/End time - if you query it and get a match with your file, then you have a 100% hit and can move onto the next episode.

exactly that ;)

Edited by Sarakha63
Link to comment
Share on other sites

Sarakha63
12 minutes ago, rbjtech said:

Sure - I fully get you - it's effectively a crowd sourced 'anonymous' db of MD5 hashes with an intro Start/End time - if you query it and get a match with your file, then you have a 100% hit and can move onto the next episode.

The questions start arising when creating the dB in the first place - who has authority to say that MD5 for that episode is 'correct' ?

for your last remark. i would say your plugin... i will likely see if the users checks the option to share his data. the plugin itself will do that

1- check if on the database an md5 exists
2- If yes get the title and credits timings ---> finish
2- If no do the actual chromaprint fingerpriting
3- if option to share is enabled send to the database the md5 and the timings 
3- if option disabled--->finish without sending

Edited by Sarakha63
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...