Pulling Sort Order from metadata

[Luke 42077]

I'm going to put this in with the option to prefer embedded titles, so if you enable that, then it will pull the sort title from within the file.

[rbjtech 5284]

1 minute ago, Luke said:

I'm going to put this in with the option to prefer embedded titles, so if you enable that, then it will pull the sort title from within the file.

..but PLEASE do not overwrite any existing emby metadata Sort Title - or you'll likely have Crusher to contend with as no doubt he/she has manually crafted those Sort titles !  

[Carlo 4560]

I'm thinking what Luke suggested is pretty safe.

Most people are not going to have sort title in their media files, so even if they did have Title there would be no change for them.
If anyone has sort titles in their media files but doesn't have "prefer embedded titles" they too are safe.

Only those people with "prefer embedded titles" enabled would read the Title and Sort Title if populated.

Now I do have a question for @Luke.

Regardless of the setting "prefer embedded titles" can we have the NFO trump both title and sort title if populated in the NFO?
This way a user could edit the file if they wanted or could use the meta editor in Emby.

But if there is a conflict between the title and sort title in files/NFO let the NFO trump it.  Anyone not liking this choice could opt to not edit in Emby but only the media.  An NFO can always be deleted to purely read from the media but not the other way around.

[Neonblue 3]

46 minutes ago, rbjtech said:

nfo not a standard ? - err.. ok .. ever heard of Kodi ?

Emby has zero rights to modify my media files (nor do I ever want it to) - thus how do you propose to keep the modified metadata ?  Yes I can do this with multiple tools of course but this need to be manageable within emby.

Thus, the only option of importing the 'sort' title is during the initial scan - where it can probably populate the emby 'sort title' it natively has (if it's empty).  This is stored in the emby dB and/or NFO file to make it portable.    But any modification to it would not (should not) update the embedded metadata. 

 

Just because Kodi uses it doesn't mean it's a standard...  Metadata is a standard, nfo is something that someone came up with to do what was already out there.  nfo is not portable other than to Kodi as far as I'm aware.  MP4 metadata is a universal standard that pretty much everyone uses. You're also talking about maintaining two files vs using just one and for what that one was designed for.  

And Plex doesn't modify the files; it updates it's local DB much like what modifying the metadata in Emby does right now.  The only difference in the process is when Emby either sees the file for the first time or is forced to update the metadata from the file.  Otherwise the current process is in play.

[Neonblue 3]

27 minutes ago, cayars said:

I'm thinking what Luke suggested is pretty safe.

Most people are not going to have sort title in their media files, so even if they did have Title there would be no change for them.
If anyone has sort titles in their media files but doesn't have "prefer embedded titles" they too are safe.

Only those people with "prefer embedded titles" enabled would read the Title and Sort Title if populated.

Now I do have a question for @Luke.

Regardless of the setting "prefer embedded titles" can we have the NFO trump both title and sort title if populated in the NFO?
This way a user could edit the file if they wanted or could use the meta editor in Emby.

But if there is a conflict between the title and sort title in files/NFO let the NFO trump it.  Anyone not liking this choice could opt to not edit in Emby but only the media.  An NFO can always be deleted to purely read from the media but not the other way around.

That would work too, and if the NFO doesn't exist then still read the embedded title.

[rbjtech 5284]

26 minutes ago, Neonblue said:

Just because Kodi uses it doesn't mean it's a standard... 

lol - the only mainstream media app that doesn't use NFO is Plex - it's used by pretty much all the major media applications and specialist managers as an option for local metadata.   To me that is a 'standard' so lets agree to disagree on that.

[Neonblue 3]

3 minutes ago, rbjtech said:

lol - the only mainstream media app that doesn't use NFO is Plex - it's used by pretty much all the major media applications and specialist managers as an option for local metadata.   To me that is a 'standard' so lets agree to disagree on that.

If it's a standard, show me the RFC.  I used to use these back in the late 80's and 90's for file descriptions on a few BBSs I ran.  But there's no standards set aside.  And I can tell you, the other apps I have do not use them.

[speechles 2055]

7 minutes ago, Neonblue said:

If it's a standard, show me the RFC.  I used to use these back in the late 80's and 90's for file descriptions on a few BBSs I ran.  But there's no standards set aside.  And I can tell you, the other apps I have do not use them.

https://datatracker.ietf.org/doc/html/draft-seantek-text-nfo-03

Good ol file-id.diz and file.nfo to tell where the things came from. Mainly for tracking releases across BBS as they spread. Expired in 2016. I do not see another RFC even drafted to represent NFO is basically a glorified XML.

Edited May 26, 2021 by speechles

[Neonblue 3]

1 minute ago, speechles said:

https://datatracker.ietf.org/doc/html/draft-seantek-text-nfo-03

Ok, I stand corrected.  But it's redundant and unnecessary.

[Carlo 4560]

Request For Comment is exactly that and not a standard.
A standard can be per adopted specification or by adopting how someone else does something which is often what happens in industry.

 

[Carlo 4560]

Just now, Neonblue said:

Ok, I stand corrected.  But it's redundant and unnecessary.

It's not redundant because it's a STANDARD way of storing information that resides WITH the media without TOUCHING the media and is quick to edit or alter unlike some media.

[speechles 2055]

3 minutes ago, cayars said:

Request For Comment is exactly that and not a standard.
A standard can be per adopted specification or by adopting how someone else does something which is often what happens in industry.

 

A standard requires an RFC. After the RFC is ratified it can be considered a standard. A drafted RFC is not a standard. Correct. But in absense of anything standard anything can be considered a standard. Consider that. There is nothing concrete in an NFO.

Edited May 26, 2021 by speechles

[Neonblue 3]

1 minute ago, cayars said:

It's not redundant because it's a STANDARD way of storing information that resides WITH the media without TOUCHING the media and is quick to edit or alter unlike some media.

That I can agree to disagree with.  It's super easy to modify the metadata on the .mpX files and have a very portable individual file vs. maintaining multiple files.  Regardless, it looks like you will support reading in metadata sort titles.

[Neonblue 3]

4 minutes ago, speechles said:

A standard requires an RFC. After the RFC is ratified it can be considered a standard. A drafted RFC is not a standard. Correct. But in absense of anything standard anything can be considered a standard. Consider that. There is nothing concrete in an NFO.

(In other words I can see why some people would want to do this, just like I strongly feel one file should have everything)  Text editors make editing the NFOs easy.  But there are even easier apps to assign and update metadata.

 

[Carlo 4560]

31 minutes ago, speechles said:

A standard requires an RFC. After the RFC is ratified it can be considered a standard. A drafted RFC is not a standard. Correct. But in absense of anything standard anything can be considered a standard. Consider that.

It most certainly does not. Do you think the PDF format was born by RFC? How about MS Word format or Excel format.  What about CSV, SQL, C, Pascal, Python, etc
Do you think HTML or FTP came from an RFC? Many of the things just mentions were long time a standard before anyone thought to RFC them.

Many a company built things the way they wanted, physical or virtual and when others copy it, you guessed it, it becomes and industry standard.

Standard and specification are not the same thing.

Edited May 26, 2021 by cayars

[speechles 2055]

  

28 minutes ago, cayars said:

It most certainly does not. Do you think the PDF format was born by RFC? How about MS Word format or Excel format.  What about CSV, SQL, C, Pascal, Python, etc
Do you think HTML or FTP came from an RFC? Many of the things just mentions were long time a standard before anyone thought to RFC them.

Many a company built things the way they wanted, physical or virtual and when others copy it, you guessed it, it becomes and industry standard.

Standard and specification are not the same thing.

Concepts meant for the world wide web go through this consortium. Since metadata is something used on an API since it uses URI it should adhere to those standards.

The other things you mention are not standards. They are just programs or languages. They evolve to become standards by becoming the more dominant market share.

I am speaking of collaborating on building the draft to design the standard. An open internet. What you speak of is market share bear evolution where you try to force your way in. Remember BetaMax? Yeah it was the better concept but failed because it wasn't an open concept. PDF was going to fail the same way until they opened it.

 

https://datatracker.ietf.org/doc/html/rfc1866

https://datatracker.ietf.org/doc/html/rfc959

HTML and FTP as RFC. Yeah. They came from here.

and with the invention of FTP the USENET was born: https://ftp.isc.org/   Long Live NNTP! Who loves a good alt.binaries group?

 

At the top of the NFO which Emby generated comes this:
<?xml version="1.0" encoding="utf-8" standalone="yes"?>

NFO is just fancy XML with a custom extension: https://datatracker.ietf.org/doc/html/rfc2376

Edited May 26, 2021 by speechles

[Neonblue 3]

22 minutes ago, speechles said:

  

Concepts meant for the world wide web go through this consortium. Since metadata is something used on an API since it uses URI it should adhere to those standards.

The other things you mention are not standards. They are just programs or languages. They evolve to become standards by becoming the more dominant market share.

I am speaking of collaborating on building the draft to design the standard. An open internet. What you speak of is market share bear evolution where you try to force your way in. Remember BetaMax? Yeah it was the better concept but failed because it wasn't an open concept. PDF was going to fail the same way until they opened it.

 

https://datatracker.ietf.org/doc/html/rfc1866

https://datatracker.ietf.org/doc/html/rfc959

HTML and FTP as RFC. Yeah. They came from here.

and with the invention of FTP the USENET was born: https://ftp.isc.org/   Long Live NNTP! Who loves a good alt.binaries group?

Well said.

[Carlo 4560]

If you don't think HTML, PDF, C, Pascal, Python, Word, Excel, etc are standards then I think we have a major disagreement.
Many an Internet company create standards without there ever being an RFC because they don't care how others want something to work.

What you're referring to are specifications and/or requirements which is different.

Look at the dates for what you posted for HTML and FTP and you'll see they were well a standard before that.  Same with gopher, UUNet, etc.
The Internet and it's protocols was not born from committee but by devs at labs and corporations. Specs and RFC came far later after everyone else was already using the standard way of doing things and wanted to refine levels and interoperability.

HTML wasn't adopted as an official standard until I think 1999 but was introduced in 1993 and was excepted as a standard way of doing things by other in 1995. The official standard/specification was like 4 to 5 years after everyone was adopting it as a standard. The RFC you linked to was 1995 or so.

[speechles 2055]

HTML was not originally designed as it is now. It was part of the Apple Macintosh Hypercard application and ran cards on Hyperstacks. They used a markup called Hyper Textual Markup Language. That is where HTML came from. I took that class at the college in 1995 when Hypercard came out wrote a ton of HyperStacks filled with HyperCards. I understand entirely when it evolved.

I am talking of standards. A standard evolves when you share an interpretation of such with others so they can also participate in using that format in their environment. This is where the RFC comes from. It is at that point it is designated important enough to be called a candidate for a request for comments. That is the only way to really share an open internet concept. You share it with the IETF as an RFC and see where it goes. Same with the ability to create a usenet group you must send a new group control message to the ISC.org through alt.config. There are governing bodies which help you openly discuss your contribution to make it the best it can be.

https://macintoshgarden.org/apps/hyper-stacks

Having that CD was basically the internet building blocks. Use that to construct your website.

Edited May 26, 2021 by speechles

[Neonblue 3]

47 minutes ago, cayars said:

Many an Internet company create standards without there ever being an RFC because they don't care how others want something to work.

 

That is the definition of proprietary.  They can make changes at will to what they do without worrying about how others work.  That makes others scramble to change what they are doing to support them.  Apple is an excellent example of this.

[roaku 842]

We're working on standardizing the definition of 'standard'. Please bear with us.

[rbjtech 5284]

3 hours ago, rbjtech said:

To me that is a 'standard' so lets agree to disagree on that.

wow - I thought I finished the 'standard' conversation 3 hours ago with the above statement ...

Don't you just love internet forums ! 

[ebr 16169]

So, is this going to create another issue with edits kind of like people who have images with their media but don't allow us to save images with their media and then they change an image in Emby and wonder why it vanishes on a refresh?

IOW - File has embedded sort title, user modifies sort title in Emby interface to something else.  Optionally, this sort title is also saved to an nfo.

Refresh and which sort title is used?

[PenkethBoy 2068]

Seeing as this is only (currently) a single user request (and they could use nfo to achieve the same result) - might be worth waiting to see if any others want this

as i suspect its going to cause problems

[Luke 42077]

The I've done it is that it's tied to the library option of preferring embedded titles, so if you've checked that box then that means you're expecting the media title to come from embedded metadata. We can always split out a separate similar option for sort title, but i think this is a good start.