Filter based on tag content

[Raf]

First, I must admit I am really impressed by this product. After having licensed several other upnp server software, Minimserver is actually the first server software providing the flexibility I was looking for in classical music streaming.

I did not find it in the manual, but is there a way to "filter" tracks being added to the library? As I am reorganising tags in my classical music collection, I would like to have the library limited to tracks which have a certain (custom) tag already filled, while leaving all others physically in the map but not added to the library.

And while I am asking ... Is it possible to combine tags into one new custom tag which can be used directly in tag values merge command? Something like performerinstrument.merge={$performer1^$instrument1^ [^],$performer2^$instrument2^ [^]}. So a combination made while indexing instead of while displaying (which I understand is done at the tagformat level).

Kind regards,
Raf

[simoncn]

(16-08-2014 16:04)Raf Wrote:  First, I must admit I am really impressed by this product. After having licensed several other upnp server software, Minimserver is actually the first server software providing the flexibility I was looking for in classical music streaming.

Thanks very much!

Quote:I did not find it in the manual, but is there a way to "filter" tracks being added to the library? As I am reorganising tags in my classical music collection, I would like to have the library limited to tracks which have a certain (custom) tag already filled, while leaving all others physically in the map but not added to the library.

You can't stop the other files being added to the library, but you can prevent them being shown by selecting your custom tag as the first browsing operation from the top-level menu. This will give you another similar-looking menu containing only the files you want to see.

Quote:And while I am asking ... Is it possible to combine tags into one new custom tag which can be used directly in tag values merge command? Something like performerinstrument.merge={$performer1^$instrument1^ [^],$performer2^$instrument2^ [^]}. So a combination made while indexing instead of while displaying (which I understand is done at the tagformat level).

Kind regards,
Raf

This isn't possible at present. The Artist tag has an indexFormat option, which is currently implemented using special code. It might be possible to extend indexFormat to support the creation of custom "synthetic" index tags in the future.

Also, it also isn't possible currently to use tag formatting to produce a formatted value such as Performer1 [Instrument1], Performer2 [Instrument2]. The closest you can get is Performer1, Performer2 [Instrument1, Instrument2]. It might be possible to add this enhancement to the formatting language in the future.

The combination of the above enhancements should give you what you want. I will add them to my list of requested features to be considered for a future version in the medium to long term.

[Raf]

(16-08-2014 17:51)simoncn Wrote:  You can't stop the other files being added to the library, but you can prevent them being shown by selecting your custom tag as the first browsing operation from the top-level menu. This will give you another similar-looking menu containing only the files you want to see.

Ok, that is how I was doing it now, and indeed works as a charm already.

(16-08-2014 17:51)simoncn Wrote:  This isn't possible at present. The Artist tag has an indexFormat option, which is currently implemented using special code. It might be possible to extend indexFormat to support the creation of custom "synthetic" index tags in the future.

Also, it also isn't possible currently to use tag formatting to produce a formatted value such as Performer1 [Instrument1], Performer2 [Instrument2]. The closest you can get is Performer1, Performer2 [Instrument1, Instrument2]. It might be possible to add this enhancement to the formatting language in the future.

The combination of the above enhancements should give you what you want. I will add them to my list of requested features to be considered for a future version in the medium to long term.

Thanks. It is only "nice to have" anyway.

Another nice to have in the synthetic concept would be to have an index tag populated based on logic on the path of the file. I have pop and classical music in different paths, and it would allow me to create a general "genre"-like tag containing "classical" and "pop" which at selection would allow me to have a cleaner further selection on composer/artist, genre, ... With the current solution, I guess I can obtain the same with the local "tag update" logic based on specific values of genre (I have no overlapping genres), or by setting up two MinimServers, each with their own index tag logic. (Or of course by storing it as a tag in the file itself )

[simoncn]

(16-08-2014 21:37)Raf Wrote:  Another nice to have in the synthetic concept would be to have an index tag populated based on logic on the path of the file. I have pop and classical music in different paths, and it would allow me to create a general "genre"-like tag containing "classical" and "pop" which at selection would allow me to have a cleaner further selection on composer/artist, genre, ... With the current solution, I guess I can obtain the same with the local "tag update" logic based on specific values of genre (I have no overlapping genres), or by setting up two MinimServers, each with their own index tag logic. (Or of course by storing it as a tag in the file itself )

I think the best approach for something like this is to store the information as a tag rather than trying to infer tag information from file paths.

Running multiple instances of MinimServer doesn't work on most platforms. This is on my "wish list" as well.

[Raf]

Thanks for the replies and valid suggestions.

(16-08-2014 21:48)simoncn Wrote:  Running multiple instances of MinimServer doesn't work on most platforms. This is on my "wish list" as well.

I am running it on a Debian server. I understood from another post someone runs 3 instances headless on Ubuntu, so I give it a try and see how far it gets me.

[simoncn]

(16-08-2014 22:12)Raf Wrote:  I am running it on a Debian server. I understood from another post someone runs 3 instances headless on Ubuntu, so I give it a try and see how far it gets me.

This is supported with the Linux .tar.gz download if you unpack it multiple times in different directories and make sure the port numbers don't conflict.

I haven't tested whether automatic startup works with this setup. It might start only one of the instances.

[Raf]

(17-08-2014 09:13)simoncn Wrote:  I haven't tested whether automatic startup works with this setup. It might start only one of the instances.

That also works after a manual intervention.

I copied the minimserver script in /etc/init.d, gave it another value after "Provided:" and pointed it to the second path (and made the script executable and added the symbolic links with update-rc.d).