Summary of Features
Detailed Features
What's New?
Download
Installation & Getting
Started
License & acknowledgements
Project Info
User's Guide
Library Window
Playlist Window
Track Editor Window
Customization
Advanced Topics
Regular Expressions
Trouble Shooting
Tables
ID3 Frame Descriptions
Utilities
mp3 Management Scripts
Tag Backup
Tag Update
Text to Speech
Speech XML Format
Speech Phonemes
Author
Peter Bennett
My Other Projects
Peter's Backup
Proxy and debug tool
|
Back Up Your ID3V1 and ID3V2 Tags
If you have a lot of information in your mp3 tags, you want to
protect it against accidental deletion. You could back up your mp3
files every time you update the tags but that will take a lot of
disks. Using the Tag Backup program you can back up just the tags
from your mp3 files and restore them into the files at any time.
Tag Backup can restore tags to your mp3 files even if the mp3 files
have been renamed or moved to different directories.
Tag Backup is a command line program, it comes in versions for
Windows and Cygwin. For Linux or other systems you should be able
to build it from the source using the included make file. The
windows and cygwin versions are installed with Jampal into the
directory selected for Jampal.
To back up your tags, create a separate directory where your tag
backup will be stored on your hard drive. Running the tag backup
program will copy all of the tags from your mp3 files into files in
that directory.
You can run tag backup again and use the same backup directory to
update your backup with files whose tags have changed. Do this on a
regular basis to make sure you have backed up the most current
data.
Include your tag backup directory in the list of directories backed
up with your regular backup process. I recommend Peter's Backup for that. When
doing incremental backups with Peter's Backup only tags that have
been modified since the last full backup will be backed up.
Tag Backup also includes an option to copy tags from one mp3 to
another or to copy them to empty mp3 files of the same name in
another directory. You can also strip the tags off a file (use the
-cm option, see below).
Files created
When creating a backup, the program creates files with names like
tag_486c9c23c09d87c8_000.mp3. The long string of hexadecimal digits
identifies a checksum of the actual mp3 sound data. This tag file
is a real mp3 file containing a few milliseconds of silence, you
can examine it with a tag editing program (e.g. load it into the
jampal library). In Windows XP you can display the song titles,
artists, etc. in the backup files by selecting options on the title
bar.
In the same directory is the file "tagbkup.log". This contains a
list of the backup file names and the original files they came
from.
Warning: If you have two or more copies of the same song
file (not just the same title, but copies of the same mp3 file),
tag backup will recognize them as the same. If you are backing up
both in the same run, it will name the second one with an _001 in
the file name, the third with _002 and so on. If these have
different tag data you will not be able to correctly restore them,
they would all be restored from the backup file with _000 in the
file name. Therefor it is a good idea to look through your backup
directories for any files called tag_xxx_001.mp3 and eliminate the
duplicates from your library.
If you have two copies of the same song file and you back them up
in separate tagbkup runs, the second will overwrite the first,
unless you use the "-n" option (see below).
The -c1, -c2, -cm options can be used to combine two mp3 files
together, as long as they have the same characteristics, for
example:
tagbkup -c2 "song_2.mp3" "combined.mp3"
tagbkup -cm "song_1.mp3" "combined.mp3"
tagbkup -cm -c1 "song_2.mp3" "combined.mp3"
This assumes the combined file gets the tags from song 2. Remember that
the ID3V2 tag must be at the front of the file, then the music data, then the
ID3V1 tag.
Usage Typing the tagbkup command with no parameters will
list out the available options.
Usage: tagbkup [options] filename ... backupdir
Option
|
Explanation
|
-b
|
Backup tags from mp3 files to the directory that is mentioned
in the last parameter. Any tags that are already up to date in
that directory are not copied.
|
-r
|
Restore tags from the directory that is mentioned in the last
parameter to the mp3 files listed before it. Any mp3 file whose
tag was not backed up to that directory will not be restored.
|
-c
|
Copy tags from mp3 files to like named mp3 files in backupdir
or copy tags from an mp3 file to another mp3 file, where the
output mp3 file name is supplied instead of backupdir.
|
-c1
|
Append ID3V1 tag from mp3 file to output file.
|
-c2
|
Copy ID3V2 tag from mp3 file to output file, overwriting
existing file.
|
-cm
|
Append mp3 data without tags from mp3 file to output file. If
the output file does not exist it is created. This can be used
to strip tags off a file.
|
-d dir
|
Copy restored mp3 files to this directory. Since restoring tags
always involves copying the whole mp3 file you can save time
here. If you have your mp3 files on a cd and want to restore
them to the hard drive and restore their tags you can do this
at the same time. In this case the file listing of mp3 files
would be the contents of the CD and the desired hard drive
place would be in the '-d' parameter.
|
-t
|
Test - no copying of data. This is useful to get a preview of
what will be done in a backup or restore
|
-v
|
Verbose. This lists every input file and whether it was backed
up, restored, etc.
|
-u
|
Report updates done. This lists only the files that were backed
up. In cases where the latest data was already in the output
directory they are not listed.
|
-n
|
No overwriting of backup files, append log file. If this option
is selected and the tag data is already backed up in the output
directory an additional copy of it is created. This is useful
if you are creating a first backup with several runs of
tagbkup. If you are creating a second backup of updated tags it
will start creating files with _001, _002 and these files are
never restored when running a restore.
|
-f fn
|
Input filename list is in file fn. If you have a large number
of mp3 files, create a file with the names in it first and use
this option. For example dir /b/s *.mp3 > listfile (under
windows), or find . -name '*.mp3' > listfile under unix or
cygwin
|
-y
|
Continue if there are errors, skipping failed files. If you are
doing a restore and you know that a lot of your tags were not
backed up, this saves time.
|
-s
|
File signature based on front part of file instead of end. The
program creates a backup file name for the tag based on the
contents of the music near the end of the file. Using this
option changes it to base the backup file name on the music
near the start of the file. If this option is used for backup
it must also be used for restore. Use this option if you have
faulty tag updating software that corrupts the end of the file.
My experience is that faulty tag editing software is more
likely to corrupt the start of the file. Note that these
corruptions of the file do not normally affect the listening
experiencing because they only delete a couple of thousandths
of a second from the start or end of the file. I recommend
using Jampal as your tag editor because it does not corrupt the
beginning or end of the file.
|
-x
|
Tags in 256 subdirectories. The tag backup files will be placed
in subdirectories named 00,01, ... up to ff, (256 in all). This
has the following advantages: (1) It overcomes a limitation in
the FAT32 file system of 16384 files per directory, so if you
have more than that number of mp3 files and you are using FAT32
you will have to use the -x option. (2) It makes it easier for
Windows Explorer - If you click on a directory with 16000 files
in it, Windows explorer can hang for a minute or so while it
retrieves the information. If you use the -x option when
creating a backup, you must use it when restoring tags and also
when updating the backup. I recommend using the -x option even
if you are using a file system that handles such large
directories, beacuse the smaller directories are more
manageable.
|
|