I do have a further suggestion regarding my gapless problem... I am currently going through the painful process of fixing all my mp3s (roughly half of my 76Gb of MP3s are affected and foobar is little help in identifying them!).

It occured to me, why are you parsing the "accurate stream data" from the file yourself? Surely it would be much faster to use the cached data that foobar can provide. In order to add files to the iPod, they must be in a playlist which means foobar has already parsed them. That data is available via the TAGZ engine and presumably there are internal hooks in the API as well. Not only would it circumvent problems like mine, but it should also be much faster...?

Also, according to this multiple id3 tags are to be expected.

Well, that is kind of more applicable to radio (or other) streams I guess. I could fix this on my side, but I think you would be hard pressed to argue you intended your files to be like this, or that there is a legitimate reason to have two id3v2 tags separated by null padding only. The iPod itself apparently doesn't always play files with double id3v2 tags if you look over the thread history a bit. I think you should complain to the EAC author instead for the brain dead tagging.

The ID3v2 tag header, which should be the first information in the file, is 10 bytes as follows:

  The default location of an ID3v2 tag is prepended to the audio so
  that players can benefit from the information when the data is
  streamed. It is however possible to append the tag, or make a
  prepend/append combination. When deciding upon where an unembedded
  tag should be located, the following order of preference SHOULD be
  considered.
 
    1. Prepend the tag.

    2. Prepend a tag with all vital information and add a second tag at
        the end of the file, before tags from other tagging systems. The
        first tag is required to have a SEEK frame.
     
    3. Add a tag at the end of the file, before tags from other tagging
        systems.

Yes. You can set it to write v2.3 tags right? If you do that, does it still add a whole new tag?

It was the second tag in your file of course, so not the first information in the file.

Oh well, I guess I'll have to write an app that scans all my MP3s and spits out an m3u of tracks that suffer from this. Maybe with a view to adding a function to strip the tags too.

OK you got me there. If I untick the EAC box for ID3v2.4 then it still writes an extra ID3 tag. That certainly looks like an EAC bug.

Are you suggesting that reversed tags would work with your app (i.e. 2.3 before 2.4)?

iTunes copes with multiple tags perfectly well as does every mp3 player I've ever tried (including the audio player your plugin is based upon) - they are all perfectly capable of finding the beginning of the audio stream and determining gapless data.

I think it would at least be worth noting in your FAQ that your plugin doesn't support files like these.

I really don't understand why you are so against adding this to the list of things to do. It really can't be more than a few lines of code. I'm not demanding an instant fix / workaround, merely requesting that you add it to the TODO list.

Quote from: Dunc-uk on 2008-01-04 19:09:25

Oh well, I guess I'll have to write an app that scans all my MP3s and spits out an m3u of tracks that suffer from this. Maybe with a view to adding a function to strip the tags too.

Why not just use MP3Tag to write all tags to a text file, have it remove all ID3v2 tags then re-apply them from the text file?

No need to re-invent the wheel, and TBH it's more comfortable than Foobar2000 when it comes to editing tags.

As a matter of pragmatism it may prove useful to some people if you could put handling for garbage data in their mp3s rather than refuse to parse them...

you could also spit out a warning so that people are aware, so they can fix them if they choose.

Yes I will try and do that

Thanks! I've a feature suggestion to bounce off you... not all of my tracks have gapless information included. I've no particular interest in having the majority of them play gaplessly anyway, but iTunes has an annoying habit of trying to estimate the gaps for these tracks whenever I use it to add a video or podcast, which I think screws up their album art in the process.

Would it be easy to have an option to add ineffective / null gapless data to the database for these tracks, so iTunes doesn't think it needs to do this?

Excellent plugin! I love the way it interfaces seamlessly with foobars playlists. Goodbye bloated iTunes.

I want to report some strange behavior on my 6G ipod classic which I am not certain was intended.

If for example from a blank state (no mp3's) I synchronize my media library and playlists, it will copy the songs that are in the playlists over twice into two separate files! That is, it seems to first see an mp3 in the media library and copy it over, and then when it sees it again on a playlist it will copy it over a second time.
Browsing on the ipod, it will confirm that there are in fact duplicate songs of anything that was on a playlist.
If, after this occurs, I synchronize again with the media library and same playlists selected however, it will report that it has removed some files, and indeed they are the duplicate entries.

So it's not a huge deal, I just synchronize twice whenever I want to update the ipod. This seems to me like unintended behavior, so I thought I should report it.

edit - nevermind; upgraded from foobar 0.9.4.5 to 0.9.5 and that seemed to fix it! Nice plugin!

[Well I would still need to seek through the file to calculate the value of the field I mentioned before. Basically, I don't think it is worth it, because I believe iTunes will still decide to 'correct' all the metadata/artwork when you try to add a file.

Quote from: djkc on 2007-12-17 02:03:14

Now, a couple questions:
Currently there's no support for doing replaygain when autoconverting files.  Is this something you have planned at all?  Currently I'm converting manually and replaygain scanning before sending to iPod, but it would be great if it could be fixed to do it for me.

I also want that feature, however the problem is the lack of any proper replaygain API. I can essentially run the menu command on my side to do the scan, but that will possibly trigger popup dialogs (as well as secondary progress dialogs though there are already a few of those) and some other inefficiencies. The main issue is those user prompts that may appear though.

The good news is that if you let foo_dop convert the files itself, and manually replaygain scan the copies on your iPod those files will not be recopied on next sync (unless the local copy was modified). You can use the 'Update metadata in ipod library' command to update the soundcheck values in the database after the manual replaygain scan. (Of course if you're regularly sending files that need to be converted it's a bit tedious)

Quote from: djkc on 2007-12-17 02:03:14

Second, there's an issue that's been bugging me for a while with smart playlists and the "load library and playlists" menu option.  I use a lot of smart playlists that are based off time - "50 most played songs in the last 1 month", "50 most played songs in the last 3 months", "songs played in the last 7 days", etc.  When I load the playlists in foobar they seem to be a cached version.  Judging from the contents I think it's from the last time I actually used iTunes to send something to my iPod.  When using the playlists on my iPod they all work correctly, but in foobar they just seem to be some old cached copy instead of being calculated.  Is there something I'm doing wrong, is it a bug, or is it just not supported?

It's a bit of a quirk. Basically, for 'Live-updating' smart playlist the iPod dynamically generates content for the playlist (which you only see on your iPod). The smart playlists also have associated 'static' items which I guess are only of use (on the ipod itself) with 'Live-updating' disabled. iTunes updates those static entries whenever you use it (by basically applying the smart playlist rules to the ipod library) but foo_dop does not (I think you can imagine writing handlers to apply all of those specific rules would be pretty complicated)

1. It's currently unsorted: it is just the order they are in the iPod database. In general that is the order they were added to the iPod. I use this so I may add an option to control the sorting.

Hmm, I don't think I explained very well... when your plugin decides that a track has no gapless data, it seems to leave the database in a state which makes iTunes think it still needs to scan the track itself.

[...]

So all I would like is an option for the plugin - having already scanned the tracks itself - to mark in the database those tracks that definitely have no gapless data, in such a way that iTunes doesn't try to re-scan them. Not being familiar with the iPod database, I don't know if this is possible.

On a loosely related note, it would seem that "SharePod" does something to the iPod database that makes your plugin think it is corrupt (I've been using this to add videos, to get around the iTunes problems). Mine is a Classic 160Gb.

Whatever it does, iTunes doesn't mind and seems to fix it so your plugin works with it again... I've no idea if this is a bug in SharePod or your plugin, but I seem to remember reading earlier in the thread someone else having this problem with another iTunes style app.

SharePod website

Tedious, and there's the problem of figuring out which files need to be scanned.  I can't think of any way in foobar to sort by replaygain.  In an average week I probably send a few hundred tracks to my iPod with about half being mp3 and the other half being a mix of flac, tak, and ogg.  Even more tedious than having to manually scan the files and the update metadata is having to first check every file sent to see whether it needs to be scanned.  Something as simple as an optional dialog box that would pop up after a send operation and list which files were converted would make the process easier.  Sort of the opposite of the transcode warning someone wanted earlier, I want to know what was converted once the job is done so I can do some followup, not before.  Currently I leave covnersion off so that the files will error out, then convert them to a temp dir, replaygain scan, and send in a second batch.

Is there any chance of getting some options for controlling the order that they're sent?  This is related to my smart playlists and to using it as a hard drive full of music while at work.  In all the smart playlists that are based solely off time and not off number of plays, the songs are displayed in the order they were added to the database.  The plugin seems to be very fickle as to what order it sends them in, depending on whether the album is VA or not and how I click on them in the playlist for sending.  Sometimes I get everything in the same order as they were in my playlist, sometimes I get completely jumbled, sometimes I get some albums completely in order with others mixed together, etc etc.  It makes browsing those playlists very interesting.  Sorting by album before sending, or simply always sending in the same order as they currently are in the playlist would make me much happier.

• Houses of the Holy
• Led Zeppelin
• Led Zeppelin II
• Led Zeppelin III
• Led Zeppelin IV
• ...

• Led Zeppelin
• Led Zeppelin II
• Led Zeppelin III
• Led Zeppelin IV
• Houses of the Holy
• ...

I don't think any such flag to indicate a file has been scanned for gapless data and it wasn't fond exists in the database (I looked and didn't find any). Bear in mind that iTunes likes to estimate the data. So I reckon it probably just tries to fill missing data.

I don't know if it helps you, but I finally managed to put all the pieces together to calculate the gapless data for Nero MP4 (AAC) files (recent Nero encoders only). So they will get gapless data from next version of foo_dop as well.

What is the error message? If it is 'Unsupported format or corrupted file' it is probably some incorrect size field or something, foo_dop checks those very strictly whereas iTunes is probably a bit more liberal there. Is there something you dislike about using foo_dop for videos?

Quote from: djkc on 2008-01-15 01:36:34

Is there any chance of getting some options for controlling the order that they're sent?  This is related to my smart playlists and to using it as a hard drive full of music while at work.  In all the smart playlists that are based solely off time and not off number of plays, the songs are displayed in the order they were added to the database.  The plugin seems to be very fickle as to what order it sends them in, depending on whether the album is VA or not and how I click on them in the playlist for sending.  Sometimes I get everything in the same order as they were in my playlist, sometimes I get completely jumbled, sometimes I get some albums completely in order with others mixed together, etc etc.  It makes browsing those playlists very interesting.  Sorting by album before sending, or simply always sending in the same order as they currently are in the playlist would make me much happier.

Do you mean you are sorting your smart playlists by date added to your iPod? I see how that may cause some issues. The order they are added was just a result of my handling for duplicate items, but it should be possible to change it back to the order the items are received (order in playlist etc.) (edit: seems a bit more complicated than I thought but I'll put it on my todo)

Sorry if this was covered before, I did do a quick search.

Is it possible to have the albums in the iPod library sorted first by date and then by name, or you don't have any control over this?

<key>SortFieldsSupported</key>
<true/>

Yes, I expect that iTunes deduced the scanned status implicitly... if it is marked has having "accurate" or "estimated" gapless info then by definition it has been scanned. Here is where a workaround may be possible... I assume foo_dop doesn't use the "estimated" option for gapless as it has no mechanism for estimating gapless info (and I don't think anybody wants one). I presume iTunes estimates gapless info by just scanning for silence at the beginning and end of the track. Of course, some tracks may not have any silence at all (the track genuinely ends at the end of an MPEG frame) and so iTunes will add estimated gapless data to this effect.

What foo_dop could do is mark all tracks with no accurate data as having "estimated" gapless data that equates to "no lead in, no lead out". The iPod playback of the file would be completely unaffected but iTunes would think the file had been scanned already and would not try to rescan.

Yes I think it was that message... I only used this app as it offered more flexibility over how to classify videos (i.e. as Music Video, TV Show, Film etc.). Some of the other apps (and iTunes) allow you to edit video specific metadata about the show, like episode number, series etc. I've nothing against using foo_dop, but as foobar isn't much of a video handler itself I wasn't expecting much in that regard from foo_dop.

Had you considered a dedicated video handling dialogue, for adding video and editing the database metadata directly? Just a thought.

[edit] Oh wow, how did I never spot the video specific metadata fields in your FAQ? Doh!

OK, I have a feature request. I've fallen over a bug / feature of the new iPod classic, whereby it orders episodes in a TV series according to the "release date" and not the episode number. See:

http://discussions.apple.com/thread.jspa?messageID=6298710

Could we get a field mapping for this database field? Or is there one already?

[edit] on the other hand it is widely being reported as a bug in the latest firmware, hopefully Apple will fix it.

I'm not explicitly sorting by that, but it's the criteria for being added to most of my smart playlists.  Most of them are of the type "Date added < [1|7|30] days" or "Last played < [1|2|7] days"  Anything based off number of plays is sorted by that and is easy enough to browse that way, but all of the time based ones appear to be sorted by when they were added onto the iPod.  So when I use those playlists on my iPod, browsing gets interesting.  Also when I do a load library in foobar2000 and play songs off my ipod as an external hard drive, I generally just skip to the bottom of the list and listen to the most recently added stuff.  In foobar it's easy enough to re-sort the albums that are jumbled up, but on the iPod itself it's a pain.

Ignoring other concerns, setting the track's gapless data as 'estimated' doesn't seem to stop iTunes from rescanning it.

As a compromise you could use Quick/Mass Tagger presets.

Hmm no remapping but there are metadata fields for them. There are various sort tables for the TV shows so not sure if the Classic is not using them or they are the problem or what... iTunes 7.6 was released very close to the recent firmware update. I'll try and check it out anyway.

I do have control over this. However, messing with the sort order can get complicated, mainly due to issues with the 5G iPod. If they added support for all of the sort fields in the last firmware update for the 5G then it would make my life easier. I am not sure if they did though. It's easy to check though if you have a 5G iPod: Hold Ctrl whilst clicking the 'View iPod device information' button in foo_dop prefs. Check if there is an entry as follows:

Hmm, that is odd... somehow iTunes does know whether it has personally scanned a track. It also does avoid rescanning tracks that foo_dop has marked as having accurate data. It is a mystery.

Well, it is probably just looking at the actual data (checking if the encoder delay/padding is set to null/zero).