0. INTRODUCTION
   ============

Q. Who created abcde? Who develops it nowadays?
A. It was originally created by Robert Woodcook. He decided to give up
   maintaining it at some point in 2002 and Jesus Climent took over. In 2012,
   Colin Tuckley and Steve McIntyre joined in to help. Currently Steve maintains
   abcde along with Andrew Strong. Contact either with comments or suggestions
   at the following email addresses:
      Steve McIntyre <93sam@@debian.org>
      Andrew Strong <andrew.david.strong@gmail.com>


1. INSTALLATION
   ============

Q. How can I install abcde on my system?
A. To get abcde you can use a pre-packaged version, available for Debian,
   FreeBSD, OpenBSD, NetBSD and others.
   In case you want to install your own release, you need to download the
   source and install the complementary programs (abcde is just a frontend
   and its functionality is provided by other programs).

   abcde needs cd-discid. Grab it from:
   http://www.hispalinux.es/~data/files/cd-discid/

   Read the README files for a complete list of requirements.

Q. I have a problem, and I reported some time ago. Have you solved it already?
A. Maybe. Try downloading the master branch of the development from
   http://git.einval.com/cgi-bin/gitweb.cgi?p=abcde.git

Q. When are you going to release a new version?
A. When it is ready. I keep on having new ideas and adding them to the trunk
   (stored in the URL above), so I dont release until all the ideas are
   finished.


2. EXECUTION
   =========

Q. I want to force abcde 2.x to behave more like 1.x. How do I do it?
A. Put these options in your abcde.conf:

   INTERACTIVE=n
   OUTPUTTYPE=mp3
   ID3TAGV=1

Q. Low on disk space?
A. abcde has different algorithms to schedule ripping and encoding - to
   optimize for disk conservation, use -l. You can also define
   WAVOUTPUTDIR=/some/other/path to your /etc/abcde.conf or ~/.abcde.conf to
   store the WAV files on another NFS shared fisk, for example.
   Also, UNIX pipes have also been implemented, using "-P", so abcde reads and
   encodes the tracks from the CD at once. However, it has the drawback that
   only one format can be encoded at a time.

Q. How can i make abcde encode faster? My CD reader is way faster than the pace
   it can encode my tracks.
A. abcde can take advantage of SMP systems, just like make. Try 'abcde -j 2' -
   it'll run two encoder jobs while it rips the next track.
   You can also make use of systems which are networked, with the help of
   distmp3. For example, MachineA has a better CPU power and MachineB has a CD
   drive. Run distmp3host (included in distmp3) on Machine A, and then run
   abcde -r MachineA from MachineB (where "MachineA" is a dns name or IP
   address). Use this in combination with '-j 0' to shift all encoding off the
   local machine.

Q. I am having problems with *my CD reading program* reading the drive as a
   user.
A. You might have to add yourself (or the users who want to use abcde) to the
   cdrom group and change the permissions of the cdrom device to 660. If you
   have a SCSI cdrom drive, check the permissions of /dev/sg* as well.

Q. I would like to give the trackname and the artist_name directly to the
   encoder (in my case oggenc), but found no documentation.
A. That is not possible, since abcde gets the information from CDDB database.
   You can create a template and fill it with the option "-n". Also, you can
   get the CDDB entry and edit it yourself. abcde has also an option to drop
   back to a template if you like none of the CDDB entries, selecting 0 from
   the CDDB options.

Q. (Thanks to Amaya) Where are those options and settings defined? Why dont
   you include a proper abcde.conf as an example?
A. I do. It should be installed under /etc/abcde.conf and contains more or
   less all the defaults abcde uses. You can use $(HOME)/.abcde.conf to
   override those defaults. More information can be obtained from the man page
   which can be consulted using "man abcde".

Q. I keep on getting files with ".ogg.ogg" extension. What am I doing wrong?
A. The code for multiple-output adds automatically the extension of the
   different outputs you select with the "-o" extension (or with OUTPUTTYPE
   variable). Erase the ".${OUTPUTTYPE}" part from the OUTPUTFORMAT variable
   in /etc/abcde.conf or ~/.abcde.conf

Q. I have modified some of the options, and now CDDB has stopped working.
A. Check that you have modified everything in the right way. For instance, if
   you modify the HTTPGET program you might want to set some options of your
   own. If you use the predefined ones (wget, curl and fetch) abcde will try
   to use some defaults. Keep in mind that the output should go to the output
   as standard output, to be saved in a file for later use.

   If abcde seems to be ignoring your configuration options or not running
   commands such as the encoder, you may have misquoted something. Config
   options such as the following do not work:

   LAMEOPTS=-h -k

   Try this instead:

   LAMEOPTS='-h -k'

Q. How can I separate the different output files I get using multiple-output
   support?
A. Use the OUTPUT variable in PLAYLISTFORMAT and OUTPUTFORMAT. It holds the
   different output file types you passed to abcde (i.e., ogg, mp3, flac)
   during playlist creation and file/directory creation.

Q. I have a CD with a data track, and abcde complaints it cannot read/encode
   it.
A. From version 2.2 onwards, abcde includes some checkings with cdparanoia, to
   try to get this right.
   (It is not easy to find a data track on a CD. Right now, I can only think
   of getting the track info from CDDB and ask the user for continuing should
   a "data" track name is found. But the solution is far from been optimal.
   For now, if you find a "data" track and you know the number, restart abcde
   specifying the tracks to be encoded, leaving out the data one).

Q. I am requested to have "eject" when setting the speed although I do not use
   it for anything.
A. If you do not use cdparanoia, eject is used for setting the speed of the
   cdrom drive. You can substitute it for "setcd" with "-x" as the argument,
   but you have to install it manually (setcd is, at least, available in
   Debian).
   Another way to get the same results is using the pre_read function, defined
   in your /etc/abcde.conf file.

Q. I have a CD set with 1+ discs, and want to have them in the same directory,
   sorted properly. What can abcde do for me?
A. Set abcde to rip the first CD and give it the option "-t 101". Use a generic
   name for the CD, and reuse it with the rest of the CDs, editing the CDDB
   entries. Use "-t 201" for the second CD and so on. Use "-w <comment>" to
   add a comment about the CD you are ripping (-w "disc #"). Alternatively you
   can use only the "-t ###" option and then move all the files to the same
   directory, but the tag/comment information on the files will differ. If you
   want the tagged track number to be set to the modified number you should use
   "-T ###".
   Even better! If you use "-W <cd_number>", abcde will put a comment to every
   CD set ("CD <number>") and modify the number of the tracks both in the file
   name and the tag/comment information.

Q. I have a live concert. I want to encode it in a single file. How do i do it?
A. Use "-1" and it will be encoded in a single file. Use "cue" as an action to
   make a CUE sheet file where the information about the tracks is stored.

Q. Can I use abcde to take a backup of my CD collection?
A. Yes. Use "abcde -1 -o flac -a default,cue" and it will create a single-track
   FLAC file with an embedded cuesheet. Later on, you can use the command
   "abcde -o vorbis -d flac-file.flac" to extract the individual tracks in
   Ogg/Vorbis format. Of course you can select whatever format you want.


3. CDDB and Musicbrainz
   ====================

Q. I need to go through an HTTP proxy for CDDB access.
A. No problem, just export your http_proxy variable first so wget/fetch/curl
   can see it.

Q. I already store CDDB entries in my hard disc. Can I make use of them?
A. Piece of cake. Just edit CDDBLOCALDIR to point to the repository and give
   abcde the "-L" flag to make use of it.

Q. Fine, now it uses local CDDB, but I have no network connection. How can I
   disable CDDB checks?
A. Use the "-L -n" combination. It will try to use local CDDB entries, and if
   nothing can be found, it will use a template. You will be also given the
   choice to edit such template.

Q. How can I use some other CDDB information provider?
A. abcde now uses FreeDB by default. To use a different FreeDB mirror or
   another CDDB service, change the CDDBURL option in your abcde.conf.

Q. After requesting CDDB data, I received several answers which seem to be
   quite alike. Can I find out the difference between some of them?
A. Yes. When asked which one you want to select, use "X,Y" where X and Y are
   the numbers of the selections you want to find the difference between.

Q. I don't like CDDB/FreeDB. How can I use Musicbrainz instead?
A. Set CDDBMETHOD=musicbrainz and try it!


4. abcde and Album Art
   ===================

Q. Can abcde download album art?   
A. As of abcde 2.7 it is possible to download a suitable cover
   image from within abcde that can be used by applications such
   as Audacious and vlc to illustrate playback with an album cover.
   Many hardware playback devices such as iPODs will also use this
   image during playback and show it on the device's display.

Q. How does album art downloading work?
A. The details are contained in abcde within a new function called
   'getalbumart' which can be called either from the commandline or 
   from within an ~/.abcde.conf file. The technique works in 3 possible
   ways:

    1. If CDDBMETHOD is musicbrainz the album art is downloaded from
       coverartarchive by MBID (MusicBrainz Identifier)
    2. If 1 fails, download is tried from amazon by ASIN (Amazon 
       Standard Identification Number)
    3. If if 1 and 2 fail or CDDBMETHOD is cddb the album art will 
       be downloaded using glyrc by artist + album title

   You  will need to install your distro's package of glyr, a music 
   related metadata search engine, and it is strongly advised that
   you use 'musicbrainz' for your cddb lookups. To access the more 
   advanced features of getalbumart it is also strongly advised that
   you install the ImageMagick package of your distro. getalbumart 
   will work without ImageMagick but functionality will be reduced.

Q. What settings and options should I select?
A. If you want to operate abcde from the commandline only you will
   get reasonable results with something like either of the following:

     abcde -o mp3:-V2 -a default,getalbumart
     abcde -o mp3:-V2 -G

   But to experience all of the options of getalbumart you should also
   consider setting the following in your ~/.abcde.conf file:

    #----------------------------------------------------------------#
    CDDBMETHOD=musicbrainz       # Use MusicBrainz for CDDB 
    ACTIONS=default,getalbumart  # Select the album art action + default

    # Give the path to the metadata search application glyrc. Add any
    # glyrc options:
    GLYRC=glyrc
    GLYRCOPTS= 

    # Give the path to ImageMagick's 'identify' which is used to give a 
    # commandline summary of the album art and also assist in any conversions.
    # Add any 'identify' options:
    IDENTIFY=identify
    IDENTIFYOPTS=

    # Give the path to ImageMagick's 'display' which will show the album art. 
    # The options below are the abcde default, feel free to add your own:  
    DISPLAYCMD=display
    DISPLAYCMDOPTS="-resize 512x512 -title abcde_album_art"

    # Give the path to ImageMagick's 'convert' which is used to convert album 
    # art to the format specified in ALBUMARTTYPE. Add any 'convert' options:   
    CONVERT=convert
    CONVERTOPTS=
    # If you want to run CONVERTOPTS on all files (including those specified
    # in ALBUMARTTYPE) set the following to "y":
    ALBUMARTALWAYSCONVERT="n"

    # By default the filename will be cover.jpg and the file format will
    # be jpeg. 'convert' will automagically change other formats to jpeg:
    ALBUMARTFILE="cover.jpg"
    ALBUMARTTYPE="JPEG"
    #----------------------------------------------------------------#

   There are more hints concerning commandline options in the sample 
   abcde.conf file included with your abcde package. Good news is that
   the existing defaults will work well without any extra adjustments :).

Q. Can I manually select album art?
A. If you have set INTERACTIVE=y (this is the default) then you will have
   the opportunity to manually select the album art of your choice either
   from online by typing in a URL or by typing in the path to a local file.
   
Q. What if I want my album art embedded?
A. The philosophy of getalbumart at the moment is to simply place the album
   art in the same location as the encoded files. Perhaps this will change
   one day. In the meantime you can use abcde's post_encode function in
   your ~/.abcde.conf file to embed the album art. Here is an example for
   mp3 embedding using eyeD3:

   #----------------------------------------------------------------#
   post_encode ()
   {
    ARTISTFILE="$(mungefilename "$TRACKARTIST")"
    ALBUMFILE="$(mungefilename "$DALBUM")"

   if [ "$VARIOUSARTISTS" = "y" ] ; then
     FINDPATH="$(eval echo "$VAOUTPUTFORMAT")"
   else
     FINDPATH="$(eval echo "$OUTPUTFORMAT")"
   fi

   FINALDIR="$(dirname "$OUTPUTDIR/$FINDPATH")"
   cd "$FINALDIR"

   if [ "$OUTPUTTYPE" = "mp3" ] && [ "$TAGGER" = "$EYED3" ] ; then
    vecho "Preparing to embed the album art..." >&2
   else
    vecho "Not embedding album art, you need mp3 output and eyeD3 tagging..." >&2
    return 1
   fi

    if [ -e "cover.jpg" ] ; then
      for i in *.mp3
      do
      eyeD3 --add-image cover.jpg:FRONT_COVER "$i"
      done
      mkdir backup && mv cover.jpg backup
      vecho "Your files have had the album art embedded..." >&2
    else
     vecho "No album art found so no image embedded..." >&2
    fi
   }
  #----------------------------------------------------------------#

  Variations of the 'for' loop in this example should work for other encoding 
  types. For example with flac:

   #--------------------
   for i in *.flac
   do 
    metaflac --import-picture-from=cover.jpg "$i"
   done
   #--------------------

  With other appropriate adjustments to the $OUTPUTTYPE and $TAGGER checks of 
  course...

  
4. FORMATS
   =======

Q. What formats can I encode my music to?
A. As for version 2.2, abcde includes support for MP3, Ogg/Vorbis,
   Ogg/Speex, Flac and MPP/MP+(Musepack). Support for AAC will be added once
   an FLOSS encoder/decoder is working.

Q. I want the new *put the newest and coolest codec there* format to be
   supported by abcde. Abcde is so cool but i want to encode also to this
   new format...
A. Patches are welcome! ;)
   No, seriously, if the format is usable, available for Linux and open
   source/free software, it should be fairly easy to integrate. Support for
   AAC (m4p) should be the next one to be introduced.

Q. I use Debian/RedHat/(put your Unix flavour here) and MP3 encoding is not
   working. What am I doing wrong?
A. Since MP3 is considered non-free (you get it for free, but hardware players
   and net broadcasters have to pay license fees), some release engineering
   groups and release management teams have decided not to provide MP3 encoding
   tools.
   These distributions or operative systems have decided to use Ogg/Vorbis as
   the default encoding format, since it contains no (known) patent claims
   and they are (supposed to be) completely Free (released under a BSD-like
   license).
   However, there is no strong (at the moment) hardware support, although
   some groups and companies are strongly working on getting it, real soon
   now. For that reason some people prefer to encode to MP3.

   UPDATE *** UPDATE
   There are known brands already selling Ogg/Vorbis portable players: Rio,
   iRiver, Neuros, iAudio,...
   Go buy one and you have no more reasons to use MP3.

   If you are among those individuals, you might need to add support for MP3
   encoding to your system:
   Debian : check http://ftp.nerim.net/debian-marillat to install lame in your
            system. Others prefer bladeenc. Check www.apt-get.org or google.
   RedHat : search on rpmfind.net
   FreeBSD: By default includes LAME support.
   OpenBSD: Available by specifying you want to install the port with LAME
            support, or by just installing it later from the ports tree.
   NetBSD : Available in pkgsrc.

   Others : Please, help me here.

Q. Huh! Why is MPPENCODER (with MPP) and .mpc the extension?
A. Dunno. You must ask the guys who created and defined the format. The
   standard is MPEGplus (MPP/MP+) but the files use .mpc extension.
   
Q. Why do I get .aac files instead of .m4a files when I encode with faac?
A. Faac uses mp4v2 to create aac files in an m4a or mp4 container. Some 
   Linux distros have disabled this mp4v2 support due to a licensing issue 
   and this leave faac only capable of generating mpeg-2/ADTS streams. 
   These streams should correctly have the .aac file extension and some 
   players (such as Audacious) will not even play mpeg-2/ADTS streams when
   they are placed in an .m4a container.
   As of version 2.7 abcde tests the version of faac being used for mp4v2 
   support and then appropriately  places the audio either in .m4a or .aac. 
   Note that tagging is not possible with mpeg-2/ADTS streams in .aac.
   
Q. How do I encode to Apple Lossless Audio Codec (alac)?
A. As of abcde 2.7.1 there are 3 ways to encode to alac: using qaac, refalac
   or FFmpeg. To use qaac or refalac follow the installation directions here:

   http://www.andrews-corner.org/qaac.html#installation

   and then set your ~/.abcde.conf file as follows:

   AACENCODERSYNTAX=qaac
   QAAC="$HOME/.wine/drive_c/qaac/qaac.exe"
   # Or use the Open Source alac encoder with this small hack:
   # QAAC="$HOME/.wine/drive_c/qaac/refalac.exe"
   QAACENCOPTS="--alac --threading --verbose"
   OUTPUTTYPE="m4a"

   To use FFmpeg (or avconv) set your ~/.abcde.conf file as follows:

   AACENCODERSYNTAX=ffmpeg
   FFMPEG=ffmpeg
   # Or give the path to avconv instead:
   # FFMPEG=avconv
   FFMPEGENCOPTS="-c:a alac"
   OUTPUTTYPE="m4a"

   And then sit back and enjoy the music :)

Q. Will FFmpeg be used with abcde for any other audio formats?
A. The plan is to use FFmpeg (or avconv) when FFmpeg has a good quality native
   encoder available. For example FFmpeg has a low quality native native encoder
   for Vorbis and AAC which will not be used but a great reverse engineered alac
   encoder that has been used in abcde. FFmpeg also has a good WavPack encoder
   that has been added in abcde 2.7.1 which can be added into your ~/.abcde.conf
   file as follows:

   # WVENCODERSYNTAX=wavpack
   WVENCODERSYNTAX=ffmpeg
   # Appropriate PATH:
   WAVPACK=wavpack
   FFMPEG=ffmpeg
   # Appropriate options:
   FFMPEGENCOPTS="-c:a wavpack -compression_level 6"
   WAVPACKENCOPTS="-hx3"
   # Correct output type:
   OUTPUTTYPE="wv"

   You can see in this example how easy it is to switch from one WavPack encoder to
   another by changing the comment mark '#' on the WVENCODERSYNTAX line. For mp2
   encoding with FFmpeg try the following:
   
   # MP2ENCODERSYNTAX=twolame
   MP2ENCODERSYNTAX=ffmpeg
   # Appropriate PATH:
   TWOLAME=twolame
   FFMPEG=ffmpeg
   FFMPEGENCOPTS='-c:a mp2 -b:a 320k'
   TWOLAMENCOPTS='--bitrate 320'
   # Correct output type:
   OUTPUTTYPE="mp2"
   
   So currently FFmpeg supports: alac, wavpack and mp2. FFmpeg native aac support will
   be added when it comes out of 'experimental'.

5. USEPIPES
   ========

Q. What is/are 'USEPIPES'??
A. This is a slightly different method of ripping and encoding with abcde
   utilising Unix pipes. Using this method a program sends as output what
   another programs expects as input without writing the information to the
   hard disk. For it to work with abcde, a ripper must send the data of the
   CD tracks which are ripped to the 'standard output', while sending other
   information and error messages to 'standard error' so that the encoder
   does not get confused. The encoder program needs to be able to encode the
   information send to it by reading from the standard input (which is what
   the ripper sends to the standard output).
   Usually Unix/Linux programs deal with the standard input/output by using
   the special file name '-', as in the following example:

   cdripper -o - | audioencoder -i - -o file.ext

   The 'cdripper' program uses the flag '-o' to identify the file it will
   output the data to. For the 'audioencoder' program to be able to read
   the output through a Unix pipe (the '|' symbol), the output needs to be
   sent to the standard output using the '-' special filename. Once the
   output has been piped to 'audioencoder' we use the same special filename
   to read the information as input with the '-i' flag and output it to
   'file.ext' with the '-o' flag.

Q. How do I use USEPIPES with abcde?
A. You can run USEPIPES in either of two ways:

     1. Add 'USEPIPES=y' to your ~/.abcde.conf file
     2. Use the commandline option 'abcde -P'

Q. Do all rippers and encoders work with USEPIPES?
A. No :). Keep in mind that every ripper and encoder differs from in each
   other in the way they interact with pipes and not all rippers and encoders
   will use or understand the '-o' and '-' options. Currently (August 2015) the
   following rippers and encoders work:

   Rippers:  cdparanoia, debug, libcdio, cdda2wav, flac, pird.
   Encoders: mp3enc, lame, bladeenc, oggenc, opusenc, flac, speexenc,
             mpcenc, wavpack, faac, neroAacEnc, fdkaac, qaac, fhgaacenc,
             ffmpeg, tta, ttaenc.

Q. Can I add a ripper or encoder which is not supported?
A. Sure! If you plan to use a ripper or encoder which is not supported you
   need to know how to use the ripper/encoder. For example, cdparanoia just
   uses '-' as the output filename and works. The program used to decode
   FLAC files (yes, abcde supports 'ripping' from a FLAC file :) uses '-c'.
   To add support for a ripper called 'cdripper' which uses the '-' special
   flag for outputting to the standard output, you need to open the config
   file (either /etc/abcde.conf or ~/.abcde.conf) and add the line:

     PIPERIPPER_cdripper="-"
   
   For an encoder called 'audioencoder' which uses '--stdin' to read the pipe
  input, you need to add:

     PIPE_audioencoder="--stdin"

  Of course, you should try your configuration and report it to the abcde
  developers so that it can be included in future versions of abcde. If
  you are familiar with POSIX shell programing, please open "abcde" and
  take a look at the lines below the one marked with "###USEPIPESSUPPORT###",
  the required syntax is reasonably clear.

6. MISC PROBLEMS
   =============

Q. What happened with normalize?
A. Normalize has changed its name under Debian, which is the GNU/Linux
   distribution for developing abcde (well, is The GNU/Linux Distribution
   i only use, at the moment). Now it is called normalize-audio.
   If you are using some other flavour of OS, you need to change the name
   of the executable in your abcde.conf file.

Q. I am using MacOSX and I am having problems to use abcde.
A. Well, I have a Mac, but I use linux on it. I have not been able to find
   a way to make it work there, since I do not have the development tools
   installed. I have placed a basic support, but some user reports are just
   complaining about abcde not encoding the ripped tracks, as if
   icedax/cdda2wav/cdparanoia/cddafs would never end reading them. If you have
   such a system and experience problems, please, report them, and I will
   try to work them out with your help.

Q. My hardware player (put it here) does not recognize the playlists created
   with abcde. What I am doing wrong?
A. Try using "-m" when creating the playlists, or setting DOSPLAYLIST as an
   option in the config file.

OBSOLETE
--------
Q. I set KEEPWAVS to "y" but abcde insists on erasing my directory, along
   with the wav tracks. What I am doing wrong?
A. The default action set includes clean, which cleans the temp directory,
   if nothing goes wrong. Take the "clean" out from the action list and you
   are done.
   UPDATE: KEEPWAVS unselects now the clean action.

--
Jesus Climent <jesus.climent@hispalinux.es>
Andrew Strong <andrew.david.strong@gmail.com>
