dtsig wrote:Thanks everybody for any/all information.

First .. it has nothing to do with // or NULL or anything. I have searched and searched the ID3.org pages and have only found a small reference to it but nothing that states that it is a requirement.

I believe that // is not an id3 standard but something that Windows Media Player introduced and it has become a "de-facto" standard.

I *think* i tried making a genre change when set to comma and nothing happenend but will try that again tonight to be sure i have the right information here.

Thanks

dtsig wrote:I tried using the "Tools->Options->Appearance->Split Multiple-value fields with:" option to make MM use a comma (,) instead of the default semi-colon (;) but no matter how many restarts it simply ignored it.

I wonder if making a tagging changes (or merely have MM write the tags again to the file) would have worked in this scenario.

Thanks everybody for any/all information.

First .. it has nothing to do with // or NULL or anything. I have searched and searched the ID3.org pages and have only found a small reference to it but nothing that states that it is a requirement. It is called a text field.

It appears from all tests that MM automatically *knows* that a semi-colon (;) is THE muli-part delimiter. In genre, artists etc MM will automatically split, index and display individual information.

I tried using the "Tools->Options->Appearance->Split Multiple-value fields with:" option to make MM use a comma (,) instead of the default semi-colon (;) but no matter how many restarts it simply ignored it.

So using a great tool add-on, RegExp Find and Replace (found here http://www.mediamonkey.com/forum/viewtopic.php?t=28510 ). Using this I converted all commas to semi-colons in my 10.3K albums/150K songs AFTER a disc image for backup. This add-on saved me years of work.

So even though I never did get a response from my 'bug report' i posted .. heck even became a gold member to get a response .. the help here was pretty good and pointed me in the direction i needed.

Thanks everyone again

I use the (;) delimiter setting.

Sometimes my genres appear without a delimiter.
For example: HardRock Heavy BluesRock LiveConcert.
In my Genres tree node it is incorrectly listed as HardRock Heavy BluesRock LiveConcert.

To correct this I simply insert (;).

For example: HardRock; Heavy; BluesRock; LiveConcert

Hope that helps.

I been wondering if this would be handled the same way as data in the custom fields,?
And in those i find the only thing that seperates words is a blank space if i use the ( , ) it is connected to the word and that changes what mm is to look for by doing the adding. It makes rock in to rock, and that is not the genere or correct word since mm is looking for rock it ignores the word with the ( , ) in the spelling.

In doing the different words in custom tags and keeping them different i find that even a blank space then ( , ) and the word is not working as it again is attached to the word it leads, so only a blank space is useable to seperate words, if this is how you get it working in mm great. but if this works in other programs i dont know because i use mm.

dtsig wrote:Scott967 thanks for that. It is always great to get peoples opinions on things.
I have still not found anything specific at the id3 site saying on way or another ... and it really shouldn't matter there as again I am not talking about writing to the tag .. but how it is generated to the DB and how MM then displays.

But i guess i will just have to get used to it not working as it appears to have been designed (per image). The genre listing does look a little odd as


thanks a lot

I don't think MM uses any "separator" within the database. The separator that you select in the options is used when entering data into the "properties" page of an item, and in displaying the library contents in the details pane. If the separator appears within the ID3 TCON frame (for media files using ID3v2), MM when scanning the media file will interpret that as a separator and make entries for each genre. As I wrote before, MM will write the ID3 TCON tag using the separator as well. I think you will find that MM will not interpret \\ as a separator within TCON. You can verify this with a hex editor. I think \\ is used as a separator in wma, don't know if that includes genre but I never use wma so can't give any more detail on that. \\ is also used in mp3tag.

scott s.
.

dtsig wrote: I have searched the id3 group and have not found a specific definition saying that it has to be semi-colon. I probably just missed it. I will continue looking for it though.

Thanks

That's because it isn't semi-colon, that's just a visual representation of the actual separator, which is a slash.

The problem you run into when you define your own splitter prior to import is that it can make strange things happen when you have files from multiple sources, all potentially using different splitters. I think sticking to a standard, or de-facto standard, is still the best practice.

If I'm not mistaken the ID3 delimiter is \\ which MediaMonkey uses as well. The ; is just used as visual representation of multiple values in MediaMonkey, not what MediaMonkey will use to read or write multiple values. The problem if you used , as delimiter is that it's not following the ID3 standards. Doesn't the program you used to write the , in the first place allow you to change it to \\?

Scott967 thanks for that. It is always great to get peoples opinions on things.

I probably wasn't clear enough but this is not about writing the tag to the mp3 file but interpreting the genre tag that is already in the mp3 file when building the MM database. As shown in the image below it appears that the ability to break apart multiple values in genre has already been thought of .. it just doesn't appear to work.


I have still not found anything specific at the id3 site saying on way or another ... and it really shouldn't matter there as again I am not talking about writing to the tag .. but how it is generated to the DB and how MM then displays.

But i guess i will just have to get used to it not working as it appears to have been designed (per image). The genre listing does look a little odd as


thanks a lot

I don't think 2.3 spec allows for the idea of multiple genre. When MM4 writes out the tag it embeds the ;<space> (assuming that is your delimiter). Other tag readers probably won't know what to do with it and just treat it as a single genre. This is (IMHO) an advantage of FLAC which uses vorbis comments. You simply tag as many GENRE= entries as you like and no need for separators. Readers can take them all, or just the last one.

Bottom line is, no assurance any given reader will understand a multiple genre tag in ID3.

The only delimiter 2.3 spec shows is "/", and that is only in certain tags (frames in spec speech). That creates the AC/DC problem. 2.4 allows ASCII/UTF-8 "null" delimiter and I think it can be used in any text field, but I don't know how widespread 2.4 is (does iTunes or WMP use it?) If I had a vote it would be for MM to read/write 2.4 ID3 tags, but you still have the problem of other readers (don't know what foobar or picard do -- don't use 'em).

scott s.
.

nohitter151 wrote:

dtsig wrote:

dtsig wrote: Wouldn't it make sense that the Scan process convert as necessary for the DB. It should not be necessary to actually change the data in the mp3, just define the delimiter before scan.

I think that's unnecessecary considering there is a defined standard exactly for that purpose for mp3 and nearly all other tag formats. It makes more sense for MM to adhere to the standard than anything else.

But MM creates a db representation of the files. It is not the tag itself. There is no rule to say that the db must have this or that. The import/scan routine of lots of db programs modify as needed for that specific system.

I have searched the id3 group and have not found a specific definition saying that it has to be semi-colon. I probably just missed it. I will continue looking for it though.

Thanks

I have a collection of 150K songs. Years and years of collecting cds and albums (and with albums ripping, correcting and hand assigning tags). All of my genre, multiple genre, use the comma ',' to delineate the genres. So for example, all songs on the album "Weird Al Yankovic" but .. yes Weird Al Yanlovic has "Comedy, Parody, Rock, Polka" in the genre field.

I have found that of course MM doesn't have any idea about the comma. From questions on the forum it was suggested first to modify the Tools/Options/Appearance field "Split Multiple Fields with" to a comma. Having done this and then restarting .. well it didn't appear to make any difference.

After a couple of other suggestions it came to converting all my songs ... changing the comma (,) to a semi-colon (;). Using a great addon for MM i tried this on a single album and it appears to work. But like i said that means changing the entire physical files. requiring a full disc image copy for backup, converting the files and rescanning all files.

I would strongly request that the above mentioned "Split Multiple Fields with" option actually be used to correctly create the create the DB as needed by MM so it can display the data as it wants. DBs are after all a representation of the physical world. It should not be necessary to change the physical file. At least to my way of thinking.

Thanks ... MM is a really great product

dtsig wrote:

dtsig wrote: Wouldn't it make sense that the Scan process convert as necessary for the DB. It should not be necessary to actually change the data in the mp3, just define the delimiter before scan.

I think that's unnecessecary considering there is a defined standard exactly for that purpose for mp3 and nearly all other tag formats. It makes more sense for MM to adhere to the standard than anything else.

dtsig wrote:Thanks for that Lowlander .. I have posted a ticket with first few seconds of example song and jpg of how MM looks.DTSig

Never did get a response from the ticket ... even as a gold member <sigh> But the help here rocks

I did try a couple of files and converting the ',' to ';' appeared to work. So will be doing an image backup tomorrow night and them modifying the entire collection and going to see if it really works. Of course the will require a rescan so hopefully will know by monday morning.

Wouldn't it make sense that the Scan process convert as necessary for the DB. It should not be necessary to actually change the data in the mp3, just define the delimiter before scan.