#!/bin/bash # Copyright (c) 1998-2001 Robert Woodcock # Copyright (c) 2003-2006 Jesus Climent # This code is hereby licensed for public consumption under either the # GNU GPL v2 or greater, or Larry Wall's Artistic license - your choice. # # You should have received a copy of the GNU General Public License along # with this program; if not, write to the Free Software Foundation, Inc., # 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA # # Copyright for this work is to expire January 1, 2010, after which it # shall be public domain. # # $Id$ VERSION='2.5.1-UNRELEASED' usage () { echo "This is abcde v$VERSION." echo "Usage: abcde [options] [tracks]" echo "Options:" echo "-1 Encode the whole CD in a single file" echo "-a " echo " Actions to perform:" echo " cddb,read,normalize,encode,tag,move,replaygain,playlist,clean" #echo "-A Experimental actions (retag, transcode)" echo "-b Enable batch normalization" #echo "-B Disable batch replaygain (do file by file)" echo "-c " echo " Specify a configuration file (overrides system and user config files)" echo "-C " echo " Specify discid to resume from (only needed if you no longer have the cd)" echo "-d " echo " Specify CDROM device to grab (flac uses a single-track flac file)" echo "-D Debugging mode (equivalent to sh -x abcde)" echo "-e Erase encoded track information from status file" echo "-f Force operations that otherwise are considered harmful. Read \"man abcde\"" echo "-g Use \"lame --nogap\" for MP3 encoding. Disables low disk and pipes flags" echo "-h This help information" #echo "-i Tag files while encoding, when possible (local only) -NWY-" echo "-j <#> Number of encoder processes to run at once (localhost)" echo "-k Keep the wav tracks for later use" echo "-l Use low disk space algorithm" echo "-L Use local CDDB storage directory" echo "-n No lookup. Don't query CDDB, just create and use template" echo "-N Noninteractive. Never prompt for anything" echo "-m Modify playlist to include CRLF endings, to comply with some players" #echo " WARNING: Deprecated. Use \"cue\" action" #echo "-M Create a CUE file" echo "-o " echo " Output file type(s) (vorbis,mp3,flac,spx,mpc,wav,m4a). Defaults to vorbis" echo "-p Pad track numbers with 0's (if less than 10 tracks)" echo "-P Use UNIX pipes to read+encode without wav files" echo "-q " echo " Set quality level (high,medium,low)" echo "-r " echo " Also encode on these remote hosts" echo "-R Use local CDDB in recursive mode" echo "-s " echo " Show fields from the CDDB info (year,genre)" echo "-S <#> Set the CD speed" echo "-t <#> Start the track numbering at a given number" echo "-T <#> Same as -t but modifies tag numbering" echo "-U Do NOT use UNICODE (UTF8) tags and comments" echo "-v Show version number and exit" echo "-V Be a bit more verbose about what is happening behind the scenes" echo "-x Eject CD after all tracks are read" echo "-w " echo " Add a comment to the CD tracks" echo "-W <#> Concatenate CDs: -T #01 -w \"CD #\"" echo "-z Use debug CDROMREADERSYNTAX option (needs cdparanoia)" echo "" echo "Tracks is a space-delimited list of tracks to grab." echo "Ranges specified with hyphens are allowed (i.e., 1-5)." echo "" #echo "Double hyphens are used to concatenate tracks" } addstatus () { echo "$@" >> "$ABCDETEMPDIR/status" } # log [level] [message] # # log outputs the right message in a common format log () { BLURB="$1" shift case $BLURB in error) echo "[ERROR] abcde: $@" >&2 ;; warning) echo "[WARNING] $@" >&2 ;; info) echo "[INFO] $@" ;; esac } # Funtions to replace the need of seq, which is too distribution dependant. f_seq_row () { i=$1 while [ $i -ne `expr $2 + 1` ] do echo $i i=`expr $i + 1` done } f_seq_line () { i=$1 if echo $i | grep "[[:digit:]]" > /dev/null 2>&1 ; then while [ $i -ne `expr $2 + 1` ] do printf $i" " i=`expr $i + 1` done echo else log error "syntax error while processing track numbers" exit 1 fi } # Functions to replace the need of awk {print $1} and {print $NF} get_first() { if [ X"$1" = "X" ]; then for first in `cat`; do break done else first=$1 fi echo $first } get_last() { if [ X"$1" = "X" ]; then for stdin in `cat`; do last=$stdin done else for last in $@ ; do :; done fi echo $last } # checkstatus [blurb] # Returns "0" if the blurb was found, returns 1 if it wasn't # Puts the blurb content, if available, on stdout. # Otherwise, returns "". checkstatus () { # Take the last line in the status file if there's multiple matches PATTERN="^$1(=.*)?$" BLURB=$(grep -E $PATTERN "$ABCDETEMPDIR/status" | tail -n 1) if [ -z "$BLURB" ]; then # No matches found return 1 else # Matches found # See if there's a = in it if [ "$(echo $BLURB | grep -c =)" != "0" ]; then echo "$(echo $BLURB | cut -f2- -d=)" fi return 0 fi } # checkwarnings [blurb] # Returns "0" if the blurb was found (meaning there was an warning), # returns 1 if it wasn't (yes this is a little backwards). # Does not print the blurb on stdout. # Otherwise, returns "". checkwarnings () { if [ -e "$ABCDETEMPDIR/warnings" ]; then :; else return 1 fi # Take the last line in the status file if there's multiple matches PATTERN="^$1(:.*)?$" BLURB="$(grep -E $PATTERN "$ABCDETEMPDIR/warnings" | tail -n 1)" if [ -z "$BLURB" ]; then # negative, we did not have a negative... return 1 else # affirmative, we had a negative... return 0 fi } # checkerrors [blurb] # Returns "0" if the blurb was found (meaning there was an error), # returns 1 if it wasn't (yes this is a little backwards). # Does not print the blurb on stdout. # Otherwise, returns "". checkerrors () { if [ -e "$ABCDETEMPDIR/errors" ]; then :; else return 1 fi # Take the last line in the status file if there's multiple matches PATTERN="^$1(:.*)?$" BLURB="$(grep -E $PATTERN "$ABCDETEMPDIR/errors" | tail -n 1)" if [ -z "$BLURB" ]; then # negative, we did not have a negative... return 1 else # affirmative, we had a negative... return 0 fi } # page [file] # Finds the right pager in the system to display a file page () { PAGEFILE="$1" # Use the debian sensible-pager wrapper to pick the pager # user has requested via their $PAGER environment variable if [ -x "/usr/bin/sensible-pager" ]; then /usr/bin/sensible-pager "$PAGEFILE" elif [ -x "$PAGER" ]; then # That failed, try to load the preferred editor, starting # with their PAGER variable $PAGER "$PAGEFILE" # If that fails, check for less elif [ -x /usr/bin/less ]; then /usr/bin/less -f "$PAGEFILE" # more should be on all UNIX systems elif [ -x /bin/more ]; then /bin/more "$PAGEFILE" else # No bananas, just cat the thing cat "$PAGEFILE" >&2 fi } # run_command [blurb] [command...] # Runs a command, silently if necessary, and updates the status file run_command () { BLURB="$1" shift # See if this is supposed to be silent if [ "$(checkstatus encode-output)" = "loud" ]; then "$@" >&2 RETURN=$? else # Special case for SMP, since # encoder output is never displayed, don't mute echos if [ -z "$BLURB" -a "$MAXPROCS" != "1" ]; then "$@" >&2 RETURN=$? else "$@" >/dev/null 2>&1 RETURN=$? fi fi case "$1" in normalize|normalize-audio) if [ "$RETURN" = "2" ]; then # File was already normalized. RETURN=0 fi ;; esac if [ "$RETURN" != "0" ]; then # Put an error in the errors file. For various reasons we # can't capture a copy of the program's output but we can # log what we attempted to execute and the error code # returned by the program. if [ "$BLURB" ]; then TWEAK="$BLURB: " fi echo "${TWEAK}returned code $RETURN: $@" >> "$ABCDETEMPDIR/errors" return $RETURN # Do not pass go, do not update the status file fi if [ "$BLURB" ]; then echo $BLURB >> "$ABCDETEMPDIR/status" fi } # relpath() and slash() are Copyright (c) 1999 Stuart Ballard and # distributed under the terms of the GNU GPL v2 or later, at your option # Function to determine if a word contains a slash. slash () { case "$1" in */*) return 0;; *) return 1;; esac } # Function to give the relative path from one file to another. # Usage: relpath fromfile tofile # eg relpath music/Artist/Album.m3u music/Artist/Album/Song.mp3 # (the result would be Album/Song.mp3) # Output is relative path to $2 from $1 on stdout # This code has the following restrictions: # Multiple ////s are not collapsed into single /s, with strange effects. # Absolute paths and ../s are handled wrong in FR (but they work in TO) # If FR is a directory it must have a trailing / relpath () { FR="$1" TO="$2" case "$TO" in /*) ;; # No processing is needed for absolute paths *) # Loop through common prefixes, ignoring them. while slash "$FR" && [ "$(echo "$FR" | cut -d/ -f1)" = "$(echo "$TO" | cut -d/ -f1)" ] do FR="$(echo "$FR" | cut -d/ -f2-)" TO="$(echo "$TO" | cut -d/ -f2-)" done # Loop through directory portions left in FR, adding appropriate ../s. while slash "$FR" do FR="$(echo "$FR" | cut -d/ -f2-)" TO="../$TO" done ;; esac echo $TO } new_checkexec () { if [ ! "$@" = "" ]; then # Cut off any command-line option we added in X=$(echo $@ | cut -d' ' -f2) if [ "$(which $X)" = "" ]; then return 1 elif [ ! -x $(which $X) ]; then return 2 fi fi return 0 } checkexec () { if [ ! "$@" = "" ]; then # Cut off any command-line option we added in X=$(echo $@ | cut -d' ' -f2) # Test for built-in abcde.function [ "$X" != "${X#abcde.}" ] && type $X >/dev/null 2>&1 && return if [ "$(which $X)" = "" ]; then log error "$X is not in your path." >&2 log info "Define the full path to the executable if it exists on your system." >&2 if [ -e /etc/debian_release ] ; then case $X in oggenc) MISSING_PACKAGE=vorbis-tools ;; lame|flac) MISSING_PACKAGE=$X ;; esac log info "Hint: apt-get install $MISSING_PACKAGE" >&2 fi exit 1 elif [ ! -x "$(which $X)" ]; then log error "$X is not executable." >&2 exit 1 fi fi } # diffentries , # max_value: the range of entries goes from 1 to diffentries () { FILENAME=$1 shift local CDDBDIFFCHOICES=$1 shift local CDDBDIFFCHOICE="$@" if [ ! X"$DIFF" = "X" ]; then PARSECHOICE1=$(echo $CDDBDIFFCHOICE | cut -d"," -f1 | xargs printf %d 2>/dev/null) PARSECHOICE2=$(echo $CDDBDIFFCHOICE | cut -d"," -f2 | xargs printf %d 2>/dev/null) if [ $PARSECHOICE1 -lt 1 ] || [ $PARSECHOICE1 -gt $CDDBDIFFCHOICES ] || \ [ $PARSECHOICE2 -lt 1 ] || [ $PARSECHOICE2 -gt $CDDBDIFFCHOICES ] || \ [ $PARSECHOICE1 -eq $PARSECHOICE2 ]; then echo "Invalid diff range. Please select two comma-separated numbers between 1 and $CDDBDIFFCHOICES" >&2 else # We parse the 2 choices to diff, store them in temporary files and diff them. for PARSECHOICE in $(echo $CDDBDIFFCHOICE | tr , \ ); do do_cddbparse "$ABCDETEMPDIR/$FILENAME.$PARSECHOICE" > "$ABCDETEMPDIR/$FILENAME.parsechoice.$PARSECHOICE" done echo "Showing diff between choices $PARSECHOICE1 and $PARSECHOICE2..." > "$ABCDETEMPDIR/$FILENAME.diff" $DIFF $DIFFOPTS "$ABCDETEMPDIR/$FILENAME.parsechoice.$PARSECHOICE1" "$ABCDETEMPDIR/$FILENAME.parsechoice.$PARSECHOICE2" >> "$ABCDETEMPDIR/$FILENAME.diff" if [ $(cat "$ABCDETEMPDIR/$FILENAME.diff" | wc -l) -ge 24 ]; then page "$ABCDETEMPDIR/$FILENAME.diff" else cat "$ABCDETEMPDIR/$FILENAME.diff" >&2 fi fi else echo "The diff program was not found in your path. Please choose a number between 0 and $CDDBDIFFCHOICES." >&2 fi } # getcddbinfo # Finds an specific field from cddbinfo getcddbinfo() { case $1 in TRACKNAME1) TRACKNAME="$(grep ^TTITLE$CDDBTRACKNUM= "$CDDBDATA" | head -n 1 | cut -f2- -d= | tr -d \[:cntrl:\] | sed 's/\ \+$//')" ;; TRACKNAME) TRACKNAME="$(grep ^TTITLE$CDDBTRACKNUM= "$CDDBDATA" | cut -f2- -d= | tr -d \[:cntrl:\] | sed 's/\ \+$//')" ;; TRACK-INFO) grep ^EXTT$CDDBTRACKNUM= "$CDDBDATA" | cut -f2- -d= | tr -d \[:cntrl:\] | sed 's/\\n/\n/g' ;; esac } # gettracknum # Get the track number we are going to use for different actions gettracknum() { if [ -n "$STARTTRACKNUMBER" ] ; then # Get the trackpadding from the current track CURRENTTRACKPADDING=$(echo -n $UTRACKNUM | wc -c) TRACKNUM=$( printf %0.${CURRENTTRACKPADDING}d $(expr ${UTRACKNUM} + ${STARTTRACKNUMBER} - $FIRSTTRACK )) else TRACKNUM=${UTRACKNUM} fi } # makeids # # Calculate cddb disc ids without requiring specialized helper programs. # largely copied from cd-discid and musicbrainz examples. some of the steps # don't make sense, but they're necessary to match the ids generated by other # programs. # ## FIXME ## Right now, we get 2 frames more than with cue2discid ?? # data@petit:~$ sh /tmp/cue2discid /home/data/tmp/flac/01.Roisin_Murphy--Ruby_Blue.flac # processing offsetimes 00:00:00 04:47:10 08:20:37 11:46:46 17:45:36 21:41:57 27:32:21 32:03:73 35:39:28 38:27:33 43:50:38 44:42:34 # 980b4b0c 12 150 21685 37687 53146 80061 97782 124071 144448 160603 173208 197438 201334 2895 # data@petit:~$ metaflac --export-cuesheet-to=- /home/data/tmp/flac/01.Roisin_Murphy--Ruby_Blue.flac| python /home/data/sources/abcde/trunk/examples/cue2discid # 980b4b0c 12 150 21685 37687 53146 80061 97782 124071 144448 160603 173208 197438 201334 2893 # # Variables: OFFSETS, TRACKS, LEADOUT, [LEADIN] makeids () { if [ X"$LEADOUT" = "X" ]; then log warning "Error trying to calculate disc ids without lead-out information." exit 1 fi # default to a two second lead-in IDMAGICNUM=150 LEADIN=${LEADIN:=150} # number of cdframes per second CDFRAMES=75 # reset cddb checksum for cddb disc-id calululation CDDBCKSUM=0 COOKEDOFFSETS="" for OFFSET in $(echo $OFFSETS) do COOKEDOFFSETS="${COOKEDOFFSETS} $(($OFFSET + $LEADIN))" OFFSETTIME=$(( ($OFFSET + $LEADIN) / $CDFRAMES )) while [ $OFFSETTIME -gt 0 ]; do CDDBCKSUM=$(($CDDBCKSUM + $OFFSETTIME % 10)) OFFSETTIME=$(($OFFSETTIME / 10)) done done COOKEDOFFSETS="${COOKEDOFFSETS:1}" # eat the leading space PREGAP=$(($(echo $OFFSETS | cut -f1 -d' '))) TOTALTIME=$(( (($LEADOUT + $LEADIN + $PREGAP) / $CDFRAMES) - (($LEADIN + $PREGAP) / $CDFRAMES))) case "$CDDBMETHOD" in cddb) printf -v DISCID "%08lx" $(( ($CDDBCKSUM % 0xff) << 24 | $TOTALTIME << 8 | $TRACKS)) ;; musicbrainz) # FIXME: don't assume the first track is 1 echo "dasd: 1 $TRACKS $LEADIN $LEADOUT $OFFSETS " DISCID=$($MUSICBRAINZ --command calcid --discinfo 1 $TRACKS $LEADIN $LEADOUT $OFFSETS) ;; esac TRACKINFO="${DISCID} $((TRACKS)) ${COOKEDOFFSETS} $((($LEADOUT + $LEADIN + $IDMAGICNUM) / $CDFRAMES))" } do_replaygain() { if checkstatus replaygain; then :; else run_command "" echo "Adding replaygain information..." for TMPOUTPUT in $( echo $OUTPUTTYPE | tr , \ ) do case $TMPOUTPUT in vorbis|ogg) OUTPUT=$OGGOUTPUTCONTAINER ;; flac) OUTPUT=$FLACOUTPUTCONTAINER ;; *) OUTPUT=$TMPOUTPUT ;; esac OUTPUTFILES="" REPLAYINDEX=0 for UTRACKNUM in $TRACKQUEUE do CDDBTRACKNUM=$(expr $UTRACKNUM - 1) getcddbinfo TRACKNAME splitvarious TRACKFILE="$(mungefilename "$TRACKNAME")" ARTISTFILE="$(mungefilename "$TRACKARTIST")" ALBUMFILE="$(mungefilename "$DALBUM")" GENRE="$(mungegenre "$GENRE")" YEAR=${CDYEAR:-$CDYEAR} gettracknum if [ "$ONETRACK" = "y" ]; then if [ "$VARIOUSARTISTS" = "y" ]; then OUTPUTFILE="$(eval echo \""$VAONETRACKOUTPUTFORMAT"\")" else OUTPUTFILE="$(eval echo \""$ONETRACKOUTPUTFORMAT"\")" fi else if [ "$VARIOUSARTISTS" = "y" ]; then OUTPUTFILE="$(eval echo \""$VAOUTPUTFORMAT"\")" else OUTPUTFILE="$(eval echo \""$OUTPUTFORMAT"\")" fi fi OUTPUTFILES[$REPLAYINDEX]="$OUTPUTDIR/$OUTPUTFILE.$OUTPUT" (( REPLAYINDEX = $REPLAYINDEX + 1 )) done case "$OUTPUT" in flac) run_command replaygain-flac nice $ENCNICE $METAFLAC $FLACGAINOPTS "${OUTPUTFILES[@]}" #run_command replaygain-flac true ;; vorbis|ogg) run_command replaygain-vorbis nice $ENCNICE $VORBISGAIN $VORBISGAINOPTS "${OUTPUTFILES[@]}" ;; mp3) run_command replaygain-mp3 nice $ENCNICE $MP3GAIN $MP3GAINOPTS "${OUTPUTFILES[@]}" ;; mpc) run_command replaygain-mpc nice $ENCNICE $MPPGAIN --auto "${OUTPUTFILES[@]}" ;; *);; esac done if checkerrors "replaygain-.{3,6}"; then :; else run_command replaygain true fi fi } # This code splits the a Various Artist track name from one of the following # forms: # # forward: Artist / Track # forward-dash: Artist - Track # reverse: Track / Artist # reverse-dash: Track - Artist # colon: Artist: Track # trailing-paren: Artist (Track) # # variables used: # VARIOUSARTISTS, VARIOUSARTISTSTYLE, TRACKNAME, TRACKARTIST splitvarious () { if [ "$VARIOUSARTISTS" = "y" ] && [ ! "$ONETRACK" = "y" ]; then case "$VARIOUSARTISTSTYLE" in forward) DTITLEARTIST="$(echo "$TRACKNAME" | sed 's- / -~-g')" TRACKARTIST="$(echo "$DTITLEARTIST" | cut -f1 -d~)" TRACKNAME="$(echo "$DTITLEARTIST" | cut -f2 -d~)" ;; forward-dash) DTITLEARTIST="$(echo "$TRACKNAME" | sed 's, - ,~,g')" TRACKARTIST="$(echo "$DTITLEARTIST" | cut -f1 -d~)" TRACKNAME="$(echo "$DTITLEARTIST" | cut -f2 -d~)" ;; reverse) DTITLEARTIST="$(echo "$TRACKNAME" | sed 's- / -~-g')" TRACKARTIST="$(echo "$DTITLEARTIST" | cut -f2 -d~)" TRACKNAME="$(echo "$DTITLEARTIST" | cut -f1 -d~)" ;; reverse-dash) DTITLEARTIST="$(echo "$TRACKNAME" | sed 's, - ,~,g')" TRACKARTIST="$(echo "$DTITLEARTIST" | cut -f2 -d~)" TRACKNAME="$(echo "$DTITLEARTIST" | cut -f1 -d~)" ;; colon) DTITLEARTIST="$(echo "$TRACKNAME" | sed 's-: -~-g')" TRACKARTIST="$(echo "$DTITLEARTIST" | cut -f1 -d~)" TRACKNAME="$(echo "$DTITLEARTIST" | cut -f2 -d~)" ;; trailing-paren) DTITLEARTIST="$(echo "$TRACKNAME" | sed 's,^\(.*\) (\(.*\)),\1~\2,')" TRACKARTIST="$(echo "$DTITLEARTIST" | cut -f2 -d~)" TRACKNAME="$(echo "$DTITLEARTIST" | cut -f1 -d~)" ;; esac elif [ "$VARIOUSARTISTS" = "y" ] && [ "$ONETRACK" = "y" ]; then TRACKARTIST="Various" else TRACKARTIST="$DARTIST" fi } do_getgenreid () { local genre=$(echo "${@}" | tr '[A-Z]' '[a-z]') local id="" case ${genre} in "blues") id=0 ;; "classic rock") id=1 ;; "country") id=2 ;; "dance") id=3 ;; "disco") id=4 ;; "funk") id=5 ;; "grunge") id=6 ;; "hip-hop") id=7 ;; "jazz") id=8 ;; "metal") id=9 ;; "new age") id=10 ;; "oldies") id=11 ;; "other") id=12 ;; "pop") id=13 ;; "r&b") id=14 ;; "rap") id=15 ;; "reggae") id=16 ;; "rock") id=17 ;; "techno") id=18 ;; "industrial") id=19 ;; "alternative") id=20 ;; "ska") id=21 ;; "death metal") id=22 ;; "pranks") id=23 ;; "soundtrack") id=24 ;; "euro-techno") id=25 ;; "ambient") id=26 ;; "trip-hop") id=27 ;; "vocal") id=28 ;; "jazz+funk") id=29 ;; "fusion") id=30 ;; "trance") id=31 ;; "classical") id=32 ;; "instrumental") id=33 ;; "acid") id=34 ;; "house") id=35 ;; "game") id=36 ;; "sound clip") id=37 ;; "gospel") id=38 ;; "noise") id=39 ;; "alt. rock") id=40 ;; "bass") id=41 ;; "soul") id=42 ;; "punk") id=43 ;; "space") id=44 ;; "meditative") id=45 ;; "instrum. pop") id=46 ;; "instrum. rock") id=47 ;; "ethnic") id=48 ;; "gothic") id=49 ;; "darkwave") id=50 ;; "techno-indust.") id=51 ;; "electronic") id=52 ;; "pop-folk") id=53 ;; "eurodance") id=54 ;; "dream") id=55 ;; "southern rock") id=56 ;; "comedy") id=57 ;; "cult") id=58 ;; "gangsta") id=59 ;; "top 40") id=60 ;; "christian rap") id=61 ;; "pop/funk"|"pop / funk") id=62 ;; "jungle") id=63 ;; "native american") id=64 ;; "cabaret") id=65 ;; "new wave") id=66 ;; "psychadelic") id=67 ;; "rave") id=68 ;; "showtunes") id=69 ;; "trailer") id=70 ;; "lo-fi") id=71 ;; "tribal") id=72 ;; "acid punk") id=73 ;; "acid jazz") id=74 ;; "polka") id=75 ;; "retro") id=76 ;; "musical") id=77 ;; "rock & roll") id=78 ;; "hard rock") id=79 ;; "folk") id=80 ;; "folk/rock") id=81 ;; "national folk") id=82 ;; "swing") id=83 ;; "fusion") id=84 ;; "bebob") id=85 ;; "latin") id=86 ;; "revival") id=87 ;; "celtic") id=88 ;; "bluegrass") id=89 ;; "avantgarde") id=90 ;; "gothic rock") id=91 ;; "progress. rock") id=92 ;; "psychadel. rock") id=93 ;; "symphonic rock") id=94 ;; "slow rock") id=95 ;; "big band") id=96 ;; "chorus") id=97 ;; "easy listening") id=98 ;; "acoustic") id=99 ;; "humour") id=100 ;; "speech") id=101 ;; "chanson") id=102 ;; "opera") id=103 ;; "chamber music") id=104 ;; "sonata") id=105 ;; "symphony") id=106 ;; "booty bass") id=107 ;; "primus") id=108 ;; "porn groove") id=109 ;; "satire") id=110 ;; "slow jam") id=111 ;; "club") id=112 ;; "tango") id=113 ;; "samba") id=114 ;; "folklore") id=115 ;; "ballad") id=116 ;; "power ballad") id=117 ;; "rhythmic soul") id=118 ;; "freestyle") id=119 ;; "duet") id=120 ;; "punk rock") id=121 ;; "drum solo") id=122 ;; "a capella") id=123 ;; "euro-house") id=124 ;; "dance hall") id=125 ;; "goa") id=126 ;; "drum & bass") id=127 ;; "club-house") id=128 ;; "hardcore") id=129 ;; "terror") id=130 ;; "indie") id=131 ;; "britpop") id=132 ;; "negerpunk") id=133 ;; "polsk punk") id=134 ;; "beat") id=135 ;; "christian gangsta rap") id=136 ;; "heavy metal") id=137 ;; "black metal") id=138 ;; "crossover") id=139 ;; "contemporary christian")id=140 ;; "christian rock") id=141 ;; "merengue") id=142 ;; "salsa") id=143 ;; "thrash metal") id=144 ;; "anime") id=145 ;; "jpop") id=146 ;; "synthpop") id=147 ;; "rock/pop"|"rock / pop") id=148 ;; *) return 1 ;; esac echo ${id} return 0 } # do_tag [tracknumber] # id3 tags a filename # variables used: # TRACKS, TRACKNAME, TRACKARTIST, TAGGER, TAGGEROPTS, VORBISCOMMENT, METAFLAC, ATOMICPARSLEY # COMMENT, DALBUM, DARTIST, CDYEAR, CDGENRE do_tag () { COMMENTOUTPUT="$(eval echo ${COMMENT})" if [ -z "$COMMENTOUTPUT" ]; then COMMENTOUTPUT="$(getcddbinfo TRACK-INFO)" fi if [ "$CDDBMETHOD" = "cddb" ]; then CDDBDISCID=$(echo $TRACKINFO | cut -d' ' -f1) fi run_command '' echo "Tagging track $1 of $TRACKS: $TRACKNAME..." # If we want to start the tracks with a given number, we need to modify the # TRACKNUM value before evaluation if [ -n "$STARTTRACKNUMBERTAG" ] ; then gettracknum fi for OUTPUT in $(echo $OUTPUTTYPE | tr , \ ) do case "$OUTPUT" in mp3) # id3v2 v0.1.9 claims to have solved the -c bug, so we merge both id3 and id3v2 GENREID=$(do_getgenreid "${CDGENRE}") # Set TPE2 in case we have a Various Artists rip. TPE2="" if [ "$VARIOUSARTISTS" = "y" ]; then TPE2="Various" fi case "$ID3SYNTAX" in id3);; eyed3) # FIXME # track numbers in mp3 come with 1/10, so we cannot # happily substitute them with $TRACKNUM run_command tagtrack-$OUTPUT-$1 nice $ENCNICE $TAGGER $TAGGEROPTS \ --comment=::"$COMMENTOUTPUT" -A "$DALBUM" \ -a "$TRACKARTIST" -t "$TRACKNAME" -Y "$CDYEAR" \ -G "$GENREID" -n "${TRACKNUM:-$1}" \ "${TRACKNUM:+-N $TRACKS}" \ "${ENCODING:+--set-encoding=$ENCODING}" \ "${TPE2:+--set-user-text-frame=TPE2:$TPE2}" \ "$ABCDETEMPDIR/track$1.$OUTPUT" ;; # FIXME # Still not activated... id3ed) run_command tagtrack-$OUTPUT-$1 nice $ENCNICE \ $TAGGER $TAGGEROPTS -c "$COMMENTOUTPUT" \ -a "$DALBUM" -n "$TRACKARTIST" -s "$TRACKNAME" \ -y "$CDYEAR" -g "$GENREID" -k "${TRACKNUM:-$1}" \ "$ABCDETEMPDIR/track$1.$OUTPUT" ;; *) # FIXME # track numbers in mp3 come with 1/10, so we cannot # happily substitute them with $TRACKNUM run_command tagtrack-$OUTPUT-$1 nice $ENCNICE \ $TAGGER $TAGGEROPTS -c "$COMMENTOUTPUT" \ -A "$DALBUM" -a "$TRACKARTIST" -t "$TRACKNAME" \ -y "$CDYEAR" -g "$GENREID" \ -T "${TRACKNUM:-$1}/$TRACKS" \ ${TPE2:+--TPE2 "$TPE2"} \ "$ABCDETEMPDIR/track$1.$OUTPUT" ;; esac ;; vorbis|ogg) case "$OGGENCODERSYNTAX" in vorbize|oggenc) # vorbiscomment can't do in-place modification, mv the file first if [ -f "$ABCDETEMPDIR/track$1.$OGGOUTPUTCONTAINER" -a ! -f "$ABCDETEMPDIR/track$1.uncommented.$OGGOUTPUTCONTAINER" ]; then mv "$ABCDETEMPDIR/track$1.$OGGOUTPUTCONTAINER" "$ABCDETEMPDIR/track$1.uncommented.$OGGOUTPUTCONTAINER" fi ( # These are from # http://www.xiph.org/ogg/vorbis/doc/v-comment.html echo ARTIST="$TRACKARTIST" echo ALBUM="$DALBUM" echo TITLE="$TRACKNAME" if [ -n "$CDYEAR" ]; then echo DATE="$CDYEAR" fi if [ -n "$CDGENRE" ]; then echo GENRE="$CDGENRE" fi echo TRACKNUMBER=${TRACKNUM:-$1} if [ -n "$DISCNUMBER" ]; then echo DISCNUMBER="$DISCNUMBER" fi echo CDDB=$CDDBDISCID if [ "$(eval echo ${COMMENT})" != "" ]; then case "$COMMENTOUTPUT" in *=*) echo "$COMMENTOUTPUT";; *) echo COMMENT="$COMMENTOUTPUT";; esac fi ) | run_command tagtrack-$OUTPUT-$1 nice $ENCNICE \ $VORBISCOMMENT $VORBISCOMMENTOPTS -w \ "$ABCDETEMPDIR/track$1.uncommented.$OGGOUTPUTCONTAINER" "$ABCDETEMPDIR/track$1.$OGGOUTPUTCONTAINER" # Doublecheck that the commented file was created # successfully before wiping the original if [ -f "$ABCDETEMPDIR/track$1.$OGGOUTPUTCONTAINER" ]; then rm -f "$ABCDETEMPDIR/track$1.uncommented.$OGGOUTPUTCONTAINER" else mv "$ABCDETEMPDIR/track$1.uncommented.$OGGOUTPUTCONTAINER" "$ABCDETEMPDIR/track$1.$OGGOUTPUTCONTAINER" fi ;; esac ;; flac) ( echo ARTIST="$TRACKARTIST" echo ALBUM="$DALBUM" echo TITLE="$TRACKNAME" if [ -n "$CDYEAR" ]; then echo DATE="$CDYEAR" fi if [ -n "$CDGENRE" ]; then echo GENRE="$CDGENRE" fi echo TRACKNUMBER="${TRACKNUM:-$1}" if [ -n "$DISCNUMBER" ]; then echo DISCNUMBER="$DISCNUMBER" fi echo CDDB="$CDDBDISCID" if [ "$(eval echo ${COMMENT})" != "" ]; then case "$COMMENTOUTPUT" in *=*) echo "$COMMENTOUTPUT";; *) echo COMMENT="$COMMENTOUTPUT";; esac fi ) | run_command tagtrack-$OUTPUT-$1 nice $ENCNICE $METAFLAC $METAFLACOPTS ${IMPORTCUESHEET:+--import-cuesheet-from="$ABCDETEMPDIR/$CUEFILE"} --import-tags-from=- "$ABCDETEMPDIR/track$1.$FLACOUTPUTCONTAINER" ;; spx) run_command tagtrack-$OUTPUT-$1 true ;; mpc) run_command tagtrack-$OUTPUT-$1 true ;; m4a) # Use a temp-file of our choice. --overWrite seems to # case core dumps with AtomicParsley 0.9.0 ATOMICTEMPFILE="$ABCDETEMPDIR/track$1.m4a-atomic" VARIOUSBOOL="false" if [ "$VARIOUSARTISTS" = "y" ]; then VARIOUSBOOL="true" fi #It has to be command file opts for AtomicParsley run_command tagtrack-$OUTPUT-$1 nice $ENCNICE $ATOMICPARSLEY $ABCDETEMPDIR/track$1.m4a --artist "$TRACKARTIST" --album "$DALBUM" --title "$TRACKNAME" --tracknum ${TRACKNUM:-$1} --year "$CDYEAR" --genre "$CDGENRE" --compilation $VARIOUSBOOL --comment "$COMMENTOUTPUT" --output $ATOMICTEMPFILE if [ -f $ATOMICTEMPFILE ]; then mv "$ATOMICTEMPFILE" "$ABCDETEMPDIR/track$1.m4a" fi ;; wav) run_command tagtrack-$OUTPUT-$1 true ;; esac done if checkerrors "tagtrack-(.{3,6})-$1"; then :; else run_command tagtrack-$1 true fi } # do_nogap_encode # variables used: # OUTPUTTYPE, {FOO}ENCODERSYNTAX, ENCNICE, ENCODER, ENCODEROPTS do_nogap_encode () { # The commands here don't go through run_command because they're never # supposed to be silenced echo "Encoding gapless MP3 tracks: $TRACKQUEUE" for OUTPUT in $(echo $OUTPUTTYPE | tr , \ ) do case "$OUTPUT" in mp3) case "$MP3ENCODERSYNTAX" in lame|toolame) ( cd "$ABCDETEMPDIR" TRACKFILES= for UTRACKNUM in $TRACKQUEUE do TRACKFILES="$TRACKFILES track$UTRACKNUM.wav" done nice $ENCNICE $MP3ENCODER $MP3ENCODEROPTS --nogap $TRACKFILES RETURN=$? if [ "$RETURN" != "0" ]; then echo "nogap-encode: $ENCODER returned code $RETURN" >> errors else for UTRACKNUM in $TRACKQUEUE do run_command encodetrack-$OUTPUT-$UTRACKNUM true #run_command encodetrack-$UTRACKNUM true done fi ) ;; esac ;; esac done if checkerrors "nogap-encode"; then :; else if [ ! "$KEEPWAVS" = "y" ] ; then if [ ! "$KEEPWAVS" = "move" ] ; then rm -f "$IN" fi fi fi # Other encoders fall through to normal encoding as the tracks have not # been entered in the status file. } # do_encode [tracknumber] [hostname] # If no hostname is specified, encode locally # variables used: # TRACKS, TRACKNAME, TRACKARTIST, DISTMP3, DISTMP3OPTS, {FOO}ENCODERSYNTAX, OUTPUTTYPE, ENCODEROPTS, DALBUM, DARTIST, ENCNICE, CDYEAR, CDGENRE, COMMENT do_encode () { if [ "$USEPIPES" = "y" ]; then case "$OUTPUT" in mp3) TEMPARG="PIPE_$MP3ENCODERSYNTAX" ;; vorbis|ogg) TEMPARG="PIPE_$OGGENCODERSYNTAX" ;; flac) TEMPARG="PIPE_$FLACENCODERSYNTAX" ;; spx) TEMPARG="PIPE_$SPEEXENCODER" ;; mpc) TEMPARG="PIPE_$MPPENCODER" ;; m4a) TEMPARG="PIPE_$MPPENCODER" ;; esac IN="$( eval echo "\$$TEMPARG" )" else IN="$ABCDETEMPDIR/track$1.wav" case "$OUTPUT" in mp3) case "$MP3ENCODERSYNTAX" in # FIXME # check if mp3enc needs -if for pipes # FIXME # I have not been able to find a working mp3enc binary mp3enc) FILEARG="-if $IN" ;; *) FILEARG="$IN" ;; esac ;; *) FILEARG="$IN" ;; esac fi # We need IN to proceed, if we are not using pipes. if [ -s "$IN" -o X"$USEPIPES" = "Xy" ] ; then for TMPOUTPUT in $(echo $OUTPUTTYPE | tr , \ ) do case "$TMPOUTPUT" in vorbis|ogg) OUTPUT=$OGGOUTPUTCONTAINER ;; flac) OUTPUT=$FLACOUTPUTCONTAINER ;; *) OUTPUT=$TMPOUTPUT ;; esac OUT="$ABCDETEMPDIR/track$1.$OUTPUT" if [ "$NOGAP" = "y" ] && checkstatus encodetrack-$OUTPUT-$1 ; then continue fi if [ X"$USEPIPES" = "Xy" ]; then RUN_COMMAND="" # We need a way to store the creation of the files when using PIPES RUN_COMMAND_PIPES="run_command encodetrack-$OUTPUT-$1 true" # When pipping it does not make sense to have a higher nice for # reading than for encoding, since it will be hold by the # encoding process. Setting an effective nice, to calm down a # bit the reading process. EFFECTIVE_NICE=$READNICE else run_command '' echo "Encoding track $1 of $TRACKS: $TRACKNAME..." RUN_COMMAND="run_command encodetrack-$OUTPUT-$1" EFFECTIVE_NICE=$ENCNICE fi case "$OUTPUT" in mp3) case "$2" in %local*%) case "$MP3ENCODERSYNTAX" in lame|toolame|gogo) $RUN_COMMAND nice $EFFECTIVE_NICE $MP3ENCODER $MP3ENCODEROPTS "$IN" "$OUT" ;; bladeenc) $RUN_COMMAND nice $EFFECTIVE_NICE $MP3ENCODER $MP3ENCODEROPTS -quit "$IN" ;; l3enc|xingmp3enc) $RUN_COMMAND nice $EFFECTIVE_NICE $MP3ENCODER "$IN" "$OUT" $MP3ENCODEROPTS ;; # FIXME # Relates to the previous FIXME since it might need the "-if" removed. mp3enc) $RUN_COMMAND nice $EFFECTIVE_NICE $MP3ENCODER -if "$IN" -of "$OUT" $MP3ENCODEROPTS ;; esac ;; *) $RUN_COMMAND nice $DISTMP3NICE $DISTMP3 $DISTMP3OPTS "$2" "$IN" "$OUT" >/dev/null 2>&1 ;; esac ;; vorbis|ogg) case "$2" in %local*%) case "$OGGENCODERSYNTAX" in vorbize) $RUN_COMMAND nice $EFFECTIVE_NICE $OGGENCODER $OGGENCODEROPTS -w "$OUT" "$IN" ;; oggenc) $RUN_COMMAND nice $EFFECTIVE_NICE $OGGENCODER $OGGENCODEROPTS -o "$OUT" "$IN" ;; esac ;; *) $RUN_COMMAND nice $DISTMP3NICE $DISTMP3 $DISTMP3OPTS "$2" "$IN" "$OUT" >/dev/null 2>&1 ;; esac ;; flac) case "$2" in %local*%) case "$FLACENCODERSYNTAX" in flac) $RUN_COMMAND nice $EFFECTIVE_NICE $FLACENCODER -f $FLACENCODEROPTS -o "$OUT" "$IN" ;; esac ;; *) vecho -n "DISTMP3:" vecho "$DISTMP3 $DISTMP3OPTS $2 $IN $OUT >/dev/null 2>&1" $RUN_COMMAND nice $DISTMP3NICE $DISTMP3 $DISTMP3OPTS "$2" "$IN" "$OUT" > /dev/null 2>&1 ;; esac ;; spx) if [ "$(eval echo ${COMMENT})" != "" ]; then case "$COMMENT" in *=*) ;; *) COMMENT="COMMENT=$COMMENT" ;; esac fi # Tag the file at encode time, as it can't be done after encoding. if [ "$DOTAG" = "y" ]; then RUN_COMMAND nice $EFFECTIVE_NICE $SPEEXENCODER $SPEEXENCODEROPTS --author "$TRACKARTIST" --title "$TRACKNAME" ${COMMENT:+--comment "$COMMENT"} "$IN" "$OUT" else $RUN_COMMAND nice $EFFECTIVE_NICE $SPEEXENCODER $SPEEXENCODEROPTS "$IN" "$OUT" fi ;; mpc) # MPP/MP+(Musepack) format (.mpc) is done locally, with inline # tagging. # I tried compiling the mppenc from corecodecs.org and got some # errors, so I have not tried it myself. ## FIXME ## Needs some cleanup to determine if an empty tag sent ## FIXME ## to the encoder ends up empty. $RUN_COMMAND nice $EFFECTIVE_NICE $MPPENCODER $MPPENCODEROPTS --artist "$TRACKARTIST" --album "$DALBUM" --title "$TRACKNAME" --track "$1" --genre "$CDGENRE" --year "$CDYEAR" --comment "$COMMENT" "$IN" "$OUT" ;; m4a) # Tag the file at encode time, as it can't be done after encoding. if [ "$DOTAG" = "y" ]; then $RUN_COMMAND nice $EFFECTIVE_NICE $AACENCODER $AACENCODEROPTS --artist "$TRACKARTIST" --album "$DALBUM" --title "$TRACKNAME" --track "$1" --genre "$CDGENRE" --year "$CDYEAR" --comment "$COMMENT" -o "$OUT" "$IN" else $RUN_COMMAND nice $ENCNICE $AACENCODER $AACENCODEROPTS -o "$OUT" "$IN" fi ;; wav) # In case of wav output we need nothing. Just keep the wavs. ;; esac $RUN_COMMAND_PIPES done # Only remove .wav if the encoding succeeded if checkerrors "encodetrack-(.{3,6})-$1"; then :; else run_command encodetrack-$1 true if [ ! "$KEEPWAVS" = "y" ] ; then if [ ! "$KEEPWAVS" = "move" ] ; then rm -f "$IN" fi fi fi else run_command "" echo "HEH! The file we were about to encode disappeared:" run_command "" echo ">> $IN" run_command encodetrack-$1 false fi } # do_preprocess [tracknumber] # variables used: # TRACKS, TRACKNAME, TRACKARTIST, DISTMP3, DISTMP3OPTS, {FOO}ENCODERSYNTAX, OUTPUTTYPE, ENCODEROPTS, DALBUM, DARTIST, ENCNICE, CDYEAR, CDGENRE, COMMENT #do_preprocess () #{ # IN="$ABCDETEMPDIR/track$1.wav" # # We need IN to proceed. # if [ -s "$IN" ] ; then # for OUTPUT in $(echo $OUTPUTTYPE | tr , \ ) # do # #OUT="$ABCDETEMPDIR/track$1.$OUTPUT" # run_command '' echo "Pre-processing track $1 of $TRACKS..." # case "$POSTPROCESSFORMAT" in # all|wav*) # run_command preprocess-$OUTPUT-$1 nice $PRENICE $WAV_PRE $IF $OF ;; # mp3) # run_command preprocess-$OUTPUT-$1 nice $PRENICE $MP3_PRE $IF $OF ;; # ogg) # run_command preprocess-$OUTPUT-$1 nice $PRENICE $OGG_PRE $IF $OF ;; # flac) # run_command preprocess-$OUTPUT-$1 nice $PRENICE $FLAC_PRE $IF $OF ;; # spx) # run_command preprocess-$OUTPUT-$1 nice $PRENICE $SPX_PRE $IF $OF ;; # esac # done # # Only remove .wav if the encoding succeeded # if checkerrors "preprocess-(.{3,4})-$1"; then # run_command preprocess-$1 false # else # run_command preprocess-$1 true # fi # else # if [ "$(checkstatus encode-output)" = "loud" ]; then # echo "HEH! The file we were about to pre-process disappeared:" # echo ">> $IN" # fi # run_command preprocess-$1 false # fi #} # do_postprocess [tracknumber] # variables used: # TRACKS, TRACKNAME, TRACKARTIST, DISTMP3, DISTMP3OPTS, {FOO}ENCODERSYNTAX, OUTPUTTYPE, ENCODEROPTS, DALBUM, DARTIST, ENCNICE, CDYEAR, CDGENRE, COMMENT #do_postprocess () #{ # for POSTPROCESSFORMAT in $(echo $POSTPROCESSFORMATS | tr , \ ) # do # IN="$ABCDETEMPDIR/track$1.$POSTPROCESSFORMAT" # # We need IN to proceed. # if [ -s "$IN" ] ; then # #OUT="$ABCDETEMPDIR/track$1.$OUTPUT" # run_command '' echo "Post-processing track $1 of $TRACKS..." # case "$POSTPROCESSFORMAT" in # mp3) # run_command postprocess-$OUTPUT-$1 nice $POSTNICE $MP3_POST $IF $OF ;; # ogg) # run_command postprocess-$OUTPUT-$1 nice $POSTNICE $OGG_POST $IF $OF ;; # flac) # run_command postprocess-$OUTPUT-$1 nice $POSTNICE $FLAC_POST $IF $OF ;; # spx) # run_command postprocess-$OUTPUT-$1 nice $POSTNICE $SPX_POST $IF $OF ;; # esac # # Only remove .wav if the encoding succeeded # if checkerrors "postprocess-(.{3,4})-$1"; then # run_command postprocess-$1 false # else # run_command postprocess-$1 true # fi # else # if [ "$(checkstatus encode-output)" = "loud" ]; then # echo "HEH! The file we were about to post-process disappeared:" # echo ">> $IN" # fi # run_command postprocess-$1 false # fi # done #} # do_single_gain # variables used: # FIXME # do_single_gain () { : } # do_batch_gain # variables used: # MP3GAIN, MP3GAINOPTS, VORBISGAIN, VORBISGAINOPTS, MPPGAIN, MPPGAINOPTS # FIXME # do_batch_gain () { # The commands here don't go through run_command because they're never supposed to be silenced echo "Batch analizing gain in tracks: $TRACKQUEUE" ( cd "$ABCDETEMPDIR" BLURB= TRACKFILES= for UTRACKNUM in $TRACKQUEUE do MP3FILES="$TRACKFILES track$UTRACKNUM.mp3" done # FIXME # Hard-coded batch option! $NORMALIZER -b $NORMALIZEROPTS $TRACKFILES RETURN=$? if [ "$RETURN" != "0" ]; then echo "batch-normalize: $NORMALIZER returned code $RETURN" >> errors else for UTRACKNUM in $TRACKQUEUE do echo normalizetrack-$UTRACKNUM >> status done fi ) } # do_batch_normalize # variables used: # NORMALIZER, NORMALIZEROPTS do_batch_normalize () { # The commands here don't go through run_command because they're never supposed to be silenced echo "Batch normalizing tracks: $TRACKQUEUE" ( cd "$ABCDETEMPDIR" BLURB= TRACKFILES= for UTRACKNUM in $TRACKQUEUE do TRACKFILES="$TRACKFILES track$UTRACKNUM.wav" done # XXX: Hard-coded batch option! $NORMALIZER -b $NORMALIZEROPTS $TRACKFILES RETURN=$? if [ "$RETURN" != "0" ]; then echo "batch-normalize: $NORMALIZER returned code $RETURN" >> errors else for UTRACKNUM in $TRACKQUEUE do echo normalizetrack-$UTRACKNUM >> status done fi ) } # do_normalize [tracknumber] # variables used: # TRACKS, TRACKNAME, NORMALIZER, NORMALIZEROPTS do_normalize () { IN="$ABCDETEMPDIR/track$1.wav" if [ -e "$IN" ] ; then run_command '' echo "Normalizing track $1 of $TRACKS: $TRACKNAME..." run_command normalizetrack-$1 $NORMALIZER $NORMALIZEROPTS "$IN" else if [ "$(checkstatus encode-output)" = "loud" ]; then echo "HEH! The file we were about to normalize disappeared:" echo ">> $IN" fi run_command normalizetrack-$1 false "File $IN was not found" fi } # do_move [tracknumber] # Deduces the outfile from environment variables # Creates directory if necessary # variables used: # TRACKNUM, TRACKNAME, TRACKARTIST, DALBUM, OUTPUTFORMAT, CDGENRE, CDYEAR do_move () { for TMPOUTPUT in $(echo $OUTPUTTYPE | tr , \ ) do # For now, set OUTPUT as TMPOUTPUT, and then change it once we have # defined the OUTPUTFILE: OUTPUT="$TMPOUTPUT" # Create ALBUMFILE, ARTISTFILE, TRACKFILE # Munge filenames as follows: # ' ' -> '_' # '/' -> '_' # ''' -> '' # '?' -> '' # Eat control characters ALBUMFILE="$(mungefilename "$DALBUM")" ARTISTFILE="$(mungefilename "$TRACKARTIST")" TRACKFILE="$(mungefilename "$TRACKNAME")" GENRE="$(mungegenre "$GENRE")" YEAR=${CDYEAR:-$CDYEAR} # If we want to start the tracks with a given number, we need to modify # the TRACKNUM value before evaluation gettracknum # Supported variables for OUTPUTFORMAT are GENRE, YEAR, ALBUMFILE, # ARTISTFILE, TRACKFILE, and TRACKNUM. if [ "$ONETRACK" = "y" ]; then if [ "$VARIOUSARTISTS" = "y" ]; then OUTPUTFILE="$(eval echo \""$VAONETRACKOUTPUTFORMAT"\")" else OUTPUTFILE="$(eval echo \""$ONETRACKOUTPUTFORMAT"\")" fi else if [ "$VARIOUSARTISTS" = "y" ]; then OUTPUTFILE="$(eval echo \""$VAOUTPUTFORMAT"\")" else OUTPUTFILE="$(eval echo \""$OUTPUTFORMAT"\")" fi fi if checkerrors "tagtrack-$OUTPUT-$1"; then :; else # Once we know the specific output was successful, we can change # the OUTPUT to the value containing the container case $TMPOUTPUT in vorbis|ogg) OUTPUT=$OGGOUTPUTCONTAINER ;; flac) OUTPUT=$FLACOUTPUTCONTAINER ;; *) OUTPUT=$TMPOUTPUT ;; esac # Check that the directory for OUTPUTFILE exists, if it doesn't, create it OUTPUTFILEDIR="$(dirname "$OUTPUTDIR/$OUTPUTFILE")" case $OUTPUT in wav) if [ "$DOCLEAN" != "y" ] && [ "$FORCE" != "y" ]; then # FIXME # introduce warnings? : else # mkdir -p shouldn't return an error if the directory already exists mkdir -p "$OUTPUTFILEDIR" run_command movetrack-$1 mv "$ABCDETEMPDIR/track$1.$OUTPUT" "$OUTPUTDIR/$OUTPUTFILE.$OUTPUT" if checkstatus movetrack-output-$OUTPUT; then :; else run_command movetrack-output-$OUTPUT true fi fi ;; *) # mkdir -p shouldn't return an error if the directory already exists mkdir -p "$OUTPUTFILEDIR" run_command movetrack-$1 mv "$ABCDETEMPDIR/track$1.$OUTPUT" "$OUTPUTDIR/$OUTPUTFILE.$OUTPUT" if checkstatus movetrack-output-$OUTPUT; then :; else run_command movetrack-output-$OUTPUT true fi ;; esac # Lets move the cue file if CUEFILE=$(checkstatus cuefile) >/dev/null ; then if [ -r "$ABCDETEMPDIR/$CUEFILE" ]; then if checkstatus movecue-$OUTPUT; then :; else # Silence the Copying output since it overlaps with encoding processes... #run_command '' vecho "Copying cue file to its destination directory..." if checkstatus onetrack >/dev/null ; then case $OUTPUT in wav) if [ "$DOCLEAN" != "y" ] && [ "$FORCE" != "y" ]; then # We dont have the dir, since it was not created before. : else run_command movecue-$OUTPUT cp "$ABCDETEMPDIR/$CUEFILE" "$OUTPUTDIR/$OUTPUTFILE.cue" fi ;; # NOTE: Creating a cue file with the 3-char-extension files is to comply with # http://brianvictor.tripod.com/mp3cue.htm#details [a-z0-9][a-z0-9][a-z0-9]) run_command movecue-$OUTPUT cp "$ABCDETEMPDIR/$CUEFILE" "$OUTPUTDIR/$OUTPUTFILE.cue" ;; *) run_command movecue-$OUTPUT cp "$ABCDETEMPDIR/$CUEFILE" "$OUTPUTDIR/$OUTPUTFILE.$OUTPUT.cue" ;; esac else run_command movecue-$OUTPUT cp "$ABCDETEMPDIR/$CUEFILE" "$OUTPUTFILEDIR/$CUEFILE" fi echo movecue-$OUTPUT >> "$ABCDETEMPDIR/status" fi fi fi fi done } # do_playlist # Create the playlist if wanted # Variables used: # PLAYLISTFORMAT, PLAYLISTDATAPREFIX, VAPLAYLISTFORMAT, VAPLAYLISTDATAPREFIX, # VARIOUSARTISTS, OUTPUTDIR do_playlist () { for TMPOUTPUT in $(echo $OUTPUTTYPE | tr , \ ) do case $TMPOUTPUT in vorbis|ogg) OUTPUT=$OGGOUTPUTCONTAINER ;; flac) OUTPUT=$FLACOUTPUTCONTAINER ;; *) OUTPUT=$TMPOUTPUT ;; esac # Create a playlist file for the playlist data to go into. # We used to wipe it out if it existed. Now we request permision if interactive. for LASTTRACK in $TRACKQUEUE; do :; done ALBUMFILE="$(mungefilename "$DALBUM")" ARTISTFILE="$(mungefilename "$DARTIST")" GENRE="$(mungegenre "$GENRE")" YEAR=${CDYEAR:-$CDYEAR} if [ "$VARIOUSARTISTS" = "y" ] ; then PLAYLISTFILE="$(eval echo "$VAPLAYLISTFORMAT")" else PLAYLISTFILE="$(eval echo "$PLAYLISTFORMAT")" fi FINALPLAYLISTDIR="$(dirname "$OUTPUTDIR/$PLAYLISTFILE")" mkdir -p "$FINALPLAYLISTDIR" if [ -s "$OUTPUTDIR/$PLAYLISTFILE" ]; then echo -n "Erase, Append to, or Keep the existing playlist file? [e/a/k] (e): " >&2 if [ "$INTERACTIVE" = "y" ]; then while [ "$DONE" != "y" ]; do read ERASEPLAYLIST case $ERASEPLAYLIST in e|E|a|A|k|K) DONE=y ;; "") ERASEPLAYLIST=e ; DONE=y ;; *) ;; esac done else echo e >&2 ERASEPLAYLIST=e fi # Once we erase the playlist, we use append to create the new one. [ "$ERASEPLAYLIST" = "e" -o "$ERASEPLAYLIST" = "E" ] && rm -f "$OUTPUTDIR/$PLAYLISTFILE" && ERASEPLAYLIST=a else # The playlist does not exist, so we can safelly use append to create the new list ERASEPLAYLIST=a fi if [ "$ERASEPLAYLIST" = "a" -o "$ERASEPLAYLIST" = "A" ]; then touch "$OUTPUTDIR/$PLAYLISTFILE" for UTRACKNUM in $TRACKQUEUE do # Shares some code with do_move since the filenames have to match CDDBTRACKNUM=$(expr $UTRACKNUM - 1) getcddbinfo TRACKNAME splitvarious TRACKFILE="$(mungefilename "$TRACKNAME")" ARTISTFILE="$(mungefilename "$TRACKARTIST")" ALBUMFILE="$(mungefilename "$DALBUM")" # If we want to start the tracks with a given number, we need to modify the # TRACKNUM value before evaluation gettracknum if [ "$VARIOUSARTISTS" = "y" ]; then OUTPUTFILE="$(eval echo \""$VAOUTPUTFORMAT\"")" else OUTPUTFILE="$(eval echo \""$OUTPUTFORMAT\"")" fi if [ "$VARIOUSARTISTS" = "y" ]; then if [ "$VAPLAYLISTDATAPREFIX" ] ; then echo ${VAPLAYLISTDATAPREFIX}$OUTPUTFILE.$OUTPUT >> "$OUTPUTDIR/$PLAYLISTFILE" else relpath "$PLAYLISTFILE", "$OUTPUTFILE.$OUTPUT" >> "$OUTPUTDIR/$PLAYLISTFILE" fi else if [ "$PLAYLISTDATAPREFIX" ]; then echo ${PLAYLISTDATAPREFIX}$OUTPUTFILE.$OUTPUT >> "$OUTPUTDIR/$PLAYLISTFILE" else relpath "$PLAYLISTFILE", "$OUTPUTFILE.$OUTPUT" >> "$OUTPUTDIR/$PLAYLISTFILE" fi fi done fi ## this will convert the playlist to have CRLF line-endings, if specified ## (some hardware players insist on CRLF endings) if [ "$DOSPLAYLIST" = "y" ]; then awk '{substr("\r",""); printf "%s\r\n", $0}' "$OUTPUTDIR/$PLAYLISTFILE" > "$ABCDETEMPDIR/PLAYLISTFILE.tmp" # mv -f "$ABCDETEMPDIR/PLAYLISTFILE.tmp" "$OUTPUTDIR/$PLAYLISTFILE" cat "$ABCDETEMPDIR/PLAYLISTFILE.tmp" | sed 's/\//\\/' > "$OUTPUTDIR/$PLAYLISTFILE" fi echo "playlistcomplete" >> "$ABCDETEMPDIR/status" done } # abcde.cue2discid # This function reads a cuefile on stdin and writes an extended # cddb query on stdout. Any PREGAP for track 1 is properly # handled, although cue files embedded in FLAC files do not # appear to properly store the PREGAP setting. :( abcde.cue2discid () { cddb_sum () { val=$1 ret=0 while [ $val -gt 0 ] ; do ret=$(( $ret + ( $val % 10) )) val=$(( $val / 10 )) done echo $ret } msf2lba () { OIFS="$IFS" IFS=":" set -- $1 IFS="$OIFS" local first second third first=$(expr ${1} + 0 ) second=$(expr ${2} + 0 ) third=$(expr ${3} + 0 ) echo $(( ((($first * 60) + $second) * 75) + $third )) } OFFSET=150 PREGAP=0 LEADOUT=0 LEADIN=88200 i=0 N=0 while read line ; do set -- $line case "$1" in TRACK) i=$(( i + 1 )) ;; INDEX) if [ "$2" -eq 1 ] ; then LBA=$(msf2lba $3) START=$(( $LBA + $PREGAP + $OFFSET )) eval TRACK$i=$START X=$(cddb_sum $(( $START / 75 )) ) N=$(( $N + $X )) fi ;; PREGAP) PREGAP=$(msf2lba $2) ;; REM) case "$2" in FLAC__lead-out) LEADOUT=$(( $4 / 588 )) ;; FLAC__lead-in) LEADIN=$(( $3 / 588 )) ;; esac ;; esac done TRACKS=$i LEADOUT=$(( $LEADOUT + $LEADIN )) LENGTH=$(( $LEADOUT/75 - $TRACK1/75 )) DISCID=$(( ( $N % 255 ) * 2**24 | $LENGTH * 2**8 | $TRACKS )) printf "%08x %i" $DISCID $TRACKS j=1 while [ $j -le $TRACKS ] ; do eval echo -n "\" \$TRACK$j\"" j=$((j+1)) done echo " $(( $LEADOUT / 75 ))" } # abcde.mkcue # abcde.mkcue [--wholedisk] # This creates a cuefile directly from the extended discid information # The --wholedisk option controls whether we're ripping data from the # start of track one or from the start of the disk (usually, but not # always the same thing!) # # Track one leadin/pregap (if any) handeling: # --wholedisk specified: # TRACK 01 AUDIO # INDEX 00 00:00:00 # INDEX 01 # Remaining track index values unchanged from disc TOC # # --wholedisk not specified # TRACK 01 AUDIO # PREGAP # INDEX 01 00:00:00 # Remaining track index values offset by # # Variables used: # TRACKINFO abcde.mkcue () { echomsf () { printf "$1%02i:%02i:%02i\n" $(($2/4500)) $((($2/75)%60)) $(($2%75)) } local MODE DISCID TRACKS local i OFFSET LBA local CUEWAVFILE if [ "$1" = --wholedisc ] ; then MODE=INDEX else MODE=PREGAP fi vecho "One track is $ONETRACK" TRACKFILE="$(mungefilename "$TRACKNAME")" ARTISTFILE="$(mungefilename "$TRACKARTIST")" ALBUMFILE="$(mungefilename "$DALBUM")" if [ "$ONETRACK" = "y" ]; then if [ "$VARIOUSARTISTS" = "y" ]; then CUEWAVFILE="$(eval echo \""$VAONETRACKOUTPUTFORMAT"\" | sed -e 's@^.*/@@').$OUTPUT" else CUEWAVFILE="$(eval echo \""$ONETRACKOUTPUTFORMAT"\" | sed -e 's@^.*/@@').$OUTPUT" fi vecho "Cue wav file is $CUEWAVFILE" else CUEWAVFILE="dummy.wav" fi set -- $TRACKINFO DISCID=$1 TRACKS=$2 shift 2 echo REM DISCID $DISCID echo FILE \""$CUEWAVEFILE"\" WAVE if [ $1 -ne 150 ] && [ $MODE = "PREGAP" ] ; then OFFSET=$1 else OFFSET=150 fi i=1 while [ $i -le "$TRACKS" ] ; do LBA=$(( $1 - $OFFSET )) printf " TRACK %02i AUDIO\n" $i if [ $i -eq 1 -a $1 -ne 150 ] ; then if [ $MODE = PREGAP ] ; then echomsf " PREGAP " $(($OFFSET-150)) else echo " INDEX 00 00:00:00" fi fi echomsf " INDEX 01 " $LBA i=$(($i+1)) shift done } # do_discid # This essentially the start of things do_discid () { # Query the CD to get the track info, unless the user specified -C # or we are using some actions which do not need the CDDB data at all #if [ ! X"$EXPACTIONS" = "X" ]; then # : #elif [ -z "$DISCID" ]; then if [ -z "$DISCID" ]; then vecho -n "Getting CD track info... " # In OSX, unmount the disc before a query if [ "$OSFLAVOUR" = "OSX" ]; then disktool -u ${CDROM#/dev/} fi case "$CDROMREADERSYNTAX" in flac) if $METAFLAC $METAFLACOPTS --export-cuesheet-to=- "$CDROM" > /dev/null 2>&1 ; then case "$CUE2DISCID" in # FIXME # right now we have 2 cue2discid internal # implementations: builtin and abcde.cue2discid. Test # both of them and decide which one we want to use. builtin) #vecho "Using builtin cue2discid implementation..." CUESHEET="$(metaflac $METAFLACOPTS --export-cuesheet-to=- "$CDROM")" #TRACKS=$(echo $CUESHEET | grep -E "TRACK \+[[:digit:]]\+ \+AUDIO" |wc -l) #TRACKS=0 OFFSETTIMES=( $(echo "$CUESHEET" | sed -n -e's/\ *INDEX 01\ \+//p' ) ) TRACKS=${#OFFSETTIMES[@]} unset OFFSETS #echo "processing offsetimes ${OFFSETTIMES[@]}" for OFFSETTIME in ${OFFSETTIMES[@]}; do OFFSETS="$OFFSETS $(( 10#${OFFSETTIME:0:2} * 4500 + 10#${OFFSETTIME:3:2} * 75 + 10#${OFFSETTIME:6:2} ))" #OFFSETS[${#OFFSETS[*]}]=$(( 10#${OFFSETTIME:0:2} * 4500 + 10#${OFFSETTIME:3:2} * 75 + 10#${OFFSETTIME:6:2} )) done LEADOUT=$(( $(echo "$CUESHEET" | grep lead-out | get_last) * 75 / 44100 )) LEADIN=$(( $(echo "$CUESHEET" | grep lead-in | get_last) * 75 / 44100 )) makeids ;; *) #vecho "Using external python cue2discid implementation..." TRACKINFO=$($METAFLAC $METAFLACOPTS --export-cuesheet-to=- "$CDROM" | $CUE2DISCID) ;; esac else log error "the input flac file does not contain a cuesheet." exit 1 fi ;; # cdparanoia|debug) # CDPARANOIAOUTPUT="$( $CDROMREADER -$CDPARANOIACDROMBUS "$CDROM" -Q --verbose 2>&1 )" # RET=$? # if [ ! "$RET" = "0" ];then # log warning "something went wrong while querying the CD... Maybe a DATA CD?" # fi # # TRACKS="$(echo "$CDPARANOIAOUTPUT" | grep -E '^[[:space:]]+[[:digit:]]' | tail -n 1 | get_first | tr -d "." | tr '\n' ' ')" # CDPARANOIAAUDIOTRACKS="$TRACKS" # # LEADOUT="$(echo "$CDPARANOIAOUTPUT" | grep -Eo '^TOTAL[[:space:]]+([[:digit:]]+)' | get_last)" # OFFSETS="$(echo "$CDPARANOIAOUTPUT" | sed -n -e's/^ .* \([0-9]\+\) \[.*/\1/p')" # makeids # ;; *) case "$CDDBMETHOD" in cddb) TRACKINFO=$($CDDISCID "$CDROM") ;; musicbrainz) TRACKINFO=$($MUSICBRAINZ --command id --device "$CDROM") ;; esac ;; esac # Make sure there's a CD in there by checking cd-discid's return code if [ ! "$?" = "0" ]; then if [ "$CDROMREADERSYNTAX" = "flac" ] ; then log error "cuesheet information from the flac file could not be read." log error "Perhaps the flac file does not contain a cuesheet?." exit 1 else log error "CD could not be read. Perhaps there's no CD in the drive?" exit 1 fi fi # In OSX, remount the disc again if [ "$OSFLAVOUR" = "OSX" ]; then disktool -m ${CDROM#/dev/} fi WEHAVEACD=y DISCID=$(echo $TRACKINFO | cut -f1 -d' ') else TRACKINFO=$(cat "$WAVOUTPUTDIR/abcde.$DISCID/discid") fi # Get a full enumeration of tracks, sort it, and put it in the TRACKQUEUE. # This needs to be done now because a section of the resuming code will need # it later. # get the number of digits to pad TRACKNUM with - we'll use this later # a CD can only hold 99 tracks, but since we support a feature for starting # numbering the tracks from a given number, we might need to set it as a # variable for the user to define... or obtain it somehow. if [ "$PADTRACKS" = "y" ] ; then TRACKNUMPADDING=2 fi ABCDETEMPDIR="$WAVOUTPUTDIR/abcde.$(echo $TRACKINFO | cut -f1 -d' ')" if [ -z "$TRACKQUEUE" ]; then if [ ! "$STRIPDATATRACKS" = "n" ]; then case "$CDROMREADERSYNTAX" in cdparanoia|debug) if [ "$WEHAVEACD" = "y" ]; then vecho "Querying the CD for audio tracks..." CDPARANOIAOUTPUT="$( $CDROMREADER -$CDPARANOIACDROMBUS "$CDROM" -Q --verbose 2>&1 )" RET=$? if [ ! "$RET" = "0" ];then log warning "something went wrong while querying the CD... Maybe a DATA CD?" fi TRACKS="$(echo "$CDPARANOIAOUTPUT" | grep -E '^[[:space:]]+[[:digit:]]' | tail -n 1 | get_first | tr -d "." | tr '\n' ' ')" CDPARANOIAAUDIOTRACKS="$TRACKS" else # Previous versions of abcde would store the tracks on a file, instead of the status record. if [ -f "$ABCDETEMPDIR/cdparanoia-audio-tracks" ]; then echo cdparanoia-audio-tracks=$( cat "$ABCDETEMPDIR/cdparanoia-audio-tracks" ) >> "$ABCDETEMPDIR/status" rm -f "$ABCDETEMPDIR/cdparanoia-audio-tracks" fi if [ -f "$ABCDETEMPDIR/status" ] && TRACKS=$(checkstatus cdparanoia-audio-tracks); then :; else TRACKS=$(echo $TRACKINFO | cut -f2 -d' ') fi fi ;; *) TRACKS=$(echo $TRACKINFO | cut -f2 -d' ') ;; esac else TRACKS=$(echo $TRACKINFO | cut -f2 -d' ') fi if echo "$TRACKS" | grep "[[:digit:]]" > /dev/null 2>&1 ;then :;else log info "The disc does not contain any tracks. Giving up..." exit 0 fi echo -n "Grabbing entire CD - tracks: " if [ ! "$PADTRACKS" = "y" ] ; then TRACKNUMPADDING=$(echo -n $TRACKS | wc -c | tr -d ' ') fi TRACKS=$(printf "%0.${TRACKNUMPADDING}d" $TRACKS) X=0 while [ "$X" -ne "$TRACKS" ] do X=$(printf "%0.${TRACKNUMPADDING}d" $(expr $X + 1)) TRACKQUEUE=$(echo $TRACKQUEUE $X) done echo $TRACKQUEUE else TRACKS=$(echo $TRACKINFO | cut -f2 -d' ') # User-supplied track queue. # Weed out non-numbers, whitespace, then sort and weed out duplicates TRACKQUEUE=$(echo $TRACKQUEUE | sed 's-[^0-9 ]--g' | tr ' ' '\n' | grep -v ^$ | sort -n | uniq | tr '\n' ' ' | sed 's- $--g') # Once cleaned, obtain the highest value in the trackqueue for number padding for LASTTRACK in $TRACKQUEUE; do :; done if [ ! "$PADTRACKS" = "y" ] ; then TRACKNUMPADDING=$(echo -n $LASTTRACK | wc -c | tr -d ' ') fi # Now we normalize the trackqueue for TRACK in $TRACKQUEUE ; do TRACKNUM=$(printf %0.${TRACKNUMPADDING}d $(expr ${TRACK} + 0 )) PADTRACKQUEUE=$(echo $PADTRACKQUEUE $TRACKNUM) done TRACKQUEUE=$PADTRACKQUEUE echo Grabbing tracks: "$TRACKQUEUE" fi QUEUEDTRACKS=$(echo $TRACKQUEUE | wc -w | tr -d ' ') # We have the discid, create a temp directory after it to store all the temp # info if [ -e "$ABCDETEMPDIR" ]; then echo -n "abcde: attempting to resume from $ABCDETEMPDIR" # It already exists, see if it's a directory if [ ! -d "$ABCDETEMPDIR" ]; then # This is a file/socket/fifo/device/etc, not a directory # Complain and exit echo >&2 echo "abcde: file $ABCDETEMPDIR already exists and does not belong to abcde." >&2 echo "Please investigate, remove it, and rerun abcde." >&2 exit 1 fi echo -n . # It's a directory, let's see if it's writable by us if [ ! -r "$ABCDETEMPDIR" ] || [ ! -w "$ABCDETEMPDIR" ] || [ ! -x "$ABCDETEMPDIR" ]; then # Nope, complain and exit echo >&2 echo "abcde: directory $ABCDETEMPDIR already exists and is not writeable." >&2 echo "Please investigate, remove it, and rerun abcde." >&2 exit 1 fi echo . # See if it's populated if [ ! -f "$ABCDETEMPDIR/discid" ]; then # Wipe and start fresh echo "abcde: $ABCDETEMPDIR/discid not found. Abcde must remove and recreate" >&2 echo -n "this directory to continue. Continue [y/N]? " >&2 if [ "$INTERACTIVE" = "y" ]; then read ANSWER else echo y >&2 ANSWER=y fi if [ "$ANSWER" != "y" ]; then exit 1 fi rm -rf "$ABCDETEMPDIR" || exit 1 mkdir -p "$ABCDETEMPDIR" if [ "$?" -gt "0" ]; then # Directory already exists or could not be created echo "abcde: Temp directory $ABCDETEMPDIR could not be created." >&2 exit 1 fi else # Everything is fine. Check for ^encodetracklocation- # and encode-output entries in the status file and # remove them. These are not relevant across sessions. if [ -f "$ABCDETEMPDIR/status" ]; then mv "$ABCDETEMPDIR/status" "$ABCDETEMPDIR/status.old" grep -v ^encodetracklocation- < "$ABCDETEMPDIR/status.old" \ | grep -v ^encode-output > "$ABCDETEMPDIR/status" fi # Remove old error messages if [ -f "$ABCDETEMPDIR/errors" ]; then rm -f "$ABCDETEMPDIR/errors" fi fi else # We are starting from scratch mkdir -p "$ABCDETEMPDIR" if [ "$?" -gt "0" ]; then # Directory already exists or could not be created echo "abcde: Temp directory $ABCDETEMPDIR could not be created." >&2 exit 1 fi cat /dev/null > "$ABCDETEMPDIR/status" # Store the abcde version in the status file. echo "abcde-version=$VERSION" >> "$ABCDETEMPDIR/status" fi if [ X"$DOCUE" = "Xy" -a X"$WEHAVEACD" = "Xy" ]; then if checkstatus cuefile > /dev/null 2>&1 ; then :; else CUEFILE=cue-$(echo "$TRACKINFO" | cut -f1 -d' ').txt vecho "Creating cue file..." case $CDROMREADERSYNTAX in flac) if $METAFLAC --export-cuesheet-to=- "$CDROM" > "$ABCDETEMPDIR/$CUEFILE"; then echo cuefile=$CUEFILE >> "$ABCDETEMPDIR/status" else log warning "the input flac file does not contain a cuesheet." fi ;; *) if $CUEREADER $CUEREADEROPTS > "$ABCDETEMPDIR/$CUEFILE"; then echo cuefile=$CUEFILE >> "$ABCDETEMPDIR/status" else log warning "reading the CUE sheet is still considered experimental" log warning "and there was a problem with the CD reading. abcde will continue," log warning "but consider reporting the problem to the abcde author" fi ;; esac fi fi # If we got the CDPARANOIA status and it is not recorded, save it now if [ -n "$CDPARANOIAAUDIOTRACKS" ]; then if checkstatus cdparanoia-audio-tracks > /dev/null 2>&1; then :; else echo cdparanoia-audio-tracks=$CDPARANOIAAUDIOTRACKS >> "$ABCDETEMPDIR/status" fi fi # Create the discid file echo "$TRACKINFO" > "$ABCDETEMPDIR/discid" if checkstatus cddbmethod > /dev/null 2>&1 ; then :; else echo "cddbmethod=$CDDBMETHOD" >> "$ABCDETEMPDIR/status" fi } # do_cleancue # Create a proper CUE file based on the CUE file we created before. do_cleancue() { if CUEFILE_IN="$ABCDETEMPDIR"/$(checkstatus cuefile); then CUEFILE_OUT=$CUEFILE_IN.out ### FIXME ### checkstatus cddb if [ -e "$CDDBDATA" ]; then vecho "Adding metadata to the cue file..." # FIXME It doesn't preserve spaces! Why? # FIXME parse $track into PERFORMER and TITLE - abcde already has code for this? n=1 echo "PERFORMER \"$DARTIST\"" >> "$CUEFILE_OUT" echo "TITLE \"$DALBUM\"" >> "$CUEFILE_OUT" # Set IFS to to prevent read from swallowing spaces and tabs OIFS="$IFS" IFS=' ' cat "$CUEFILE_IN" | while read line do if echo "$line" | grep "INDEX 01" > /dev/null 2>&1 ; then # FIXME # Possible patch: remove the line above, uncomment the 2 lines below. # echo "$line" >> "$CUEFILE_OUT" # if echo "$line" | grep "^[[:space:]]*TRACK" > /dev/null 2>&1 ; then eval track="\$TRACK$n" n=$(expr $n + 1) echo " TITLE \"$track\"" >> "$CUEFILE_OUT" # When making a single-track rip, put the # actual file name into the file declaration # in the cue file so that it is usable by # music players and the like elif [ "$ONETRACK" = "y" ] && echo "$line" | grep '^FILE "dummy.wav" WAVE' > /dev/null 2>&1 ; then TRACKFILE="$(mungefilename "$TRACKNAME")" ARTISTFILE="$(mungefilename "$TRACKARTIST")" ALBUMFILE="$(mungefilename "$DALBUM")" if [ "$VARIOUSARTISTS" = "y" ]; then OUTPUTFILE="$(eval echo \""$VAONETRACKOUTPUTFORMAT"\" | sed -e 's@^.*/@@').$OUTPUT" else OUTPUTFILE="$(eval echo \""$ONETRACKOUTPUTFORMAT"\" | sed -e 's@^.*/@@').$OUTPUT" fi echo "FILE \"$OUTPUTFILE\" WAVE" >> "$CUEFILE_OUT" continue fi # FIXME # If the lines above are uncommented, remove the line below. echo "$line" >> "$CUEFILE_OUT" done IFS="$OIFS" mv "$CUEFILE_OUT" "$CUEFILE_IN" echo "cleancuefile" >> "$ABCDETEMPDIR/status" fi fi } # do_cddbparse # Parses a CDDB file and outputs the title and the track names. # Variables: CDDBFILE do_cddbparse () { CDDBPARSEFILE="$1" # List out disc title/author and contents if [ "$ONETRACK" = "y" ]; then vecho "ONETRACK mode selected: displaying only the title of the CD..." fi echo "---- $(grep DTITLE "${CDDBPARSEFILE}" | cut '-d=' -f2- | tr -d \\r\\n ) ----" if [ X"$SHOWCDDBYEAR" = "Xy" ]; then PARSEDYEAR=$(grep DYEAR "${CDDBPARSEFILE}" | cut '-d=' -f2-) if [ ! X"$PARSEDYEAR" = "X" ]; then echo "Year: $PARSEDYEAR" fi fi if [ X"$SHOWCDDBGENRE" = "Xy" ]; then PARSEDGENRE=$(grep DGENRE "${CDDBPARSEFILE}" | cut '-d=' -f2-) if [ ! X"$PARSEDGENRE" = "X" ]; then echo "Genre: $PARSEDGENRE" fi fi if [ ! "$ONETRACK" = "y" ]; then for TRACK in $(f_seq_row 1 $TRACKS) do echo $TRACK: "$(grep ^TTITLE$(expr $TRACK - 1)= "${CDDBPARSEFILE}" | cut -f2- -d= | tr -d \\r\\n)" done fi } # do_localcddb # Check for a local CDDB file, and report success do_localcddb () { if checkstatus cddb-readcomplete && checkstatus cddb-choice >/dev/null; then :; else CDDBLOCALSTATUS="notfound" CDDBDISCID=$(echo $TRACKINFO | cut -d' ' -f1) USELOCALRESP="y" if [ "$CDDBLOCALRECURSIVE" = "y" ]; then CDDBLOCALRESULTS="$(find ${CDDBLOCALDIR} -name "${CDDBDISCID}" -type f 2> /dev/null)" if [ ! "${CDDBLOCALRESULTS}" = "" ]; then if (( $(echo "${CDDBLOCALRESULTS}" | wc -l) == 1 )); then CDDBLOCALFILE="${CDDBLOCALRESULTS}" CDDBLOCALMATCH=single elif (( $(echo "${CDDBLOCALRESULTS}" | wc -l) > 1 )); then CDDBLOCALMATCH=multiple fi else CDDBLOCALMATCH=none fi elif [ "$CDDBLOCALMATCH" = "none" ] && [ -r "${CDDBLOCALDIR}/${CDDBDISCID}" ]; then CDDBLOCALFILE="${CDDBLOCALDIR}/${CDDBDISCID}" CDDBLOCALMATCH=single else CDDBLOCALMATCH=none fi # If the user has selected to check a local CDDB repo, we proceed with it case $CDDBLOCALMATCH in multiple) echo "Processing multiple matching CDDB entries..." > "$ABCDETEMPDIR/cddblocalchoices" X=0 echo "$CDDBLOCALRESULTS" | while read RESULT ; do X=$(expr $X + 1) # List out disc title/author and contents CDDBLOCALREAD="$ABCDETEMPDIR/cddblocalread.$X" cat "$RESULT" > "${CDDBLOCALREAD}" { echo -n "#$X: " do_cddbparse "${CDDBLOCALREAD}" echo "" ##FIXME## QUICK HACK !!!! if [ ! "$INTERACTIVE" = "y" ]; then break ; fi } >> "$ABCDETEMPDIR/cddblocalchoices" done if [ $(cat "$ABCDETEMPDIR/cddblocalchoices" | wc -l) -ge 24 ] && [ "$INTERACTIVE" = "y" ]; then page "$ABCDETEMPDIR/cddblocalchoices" else # It's all going to fit in one page, cat it cat "$ABCDETEMPDIR/cddblocalchoices" >&2 fi CDDBLOCALCHOICES=$( echo "$CDDBLOCALRESULTS" | wc -l ) # Setting the choice to an impossible integer to avoid errors in the numeric comparisons CDDBLOCALCHOICENUM=-1 if [ "$INTERACTIVE" = "y" ]; then while [ $CDDBLOCALCHOICENUM -lt 0 ] || [ $CDDBLOCALCHOICENUM -gt $CDDBLOCALCHOICES ]; do echo -n "Locally cached CDDB entries found. Which one would you like to use (0 for none)? [0-$CDDBLOCALCHOICES]: " >&2 read CDDBLOCALCHOICE [ x"$CDDBLOCALCHOICE" = "x" ] && CDDBLOCALCHOICE="1" # FIXME # Introduce diff's if echo $CDDBLOCALCHOICE | grep -E "[[:space:]]*[[:digit:]]+,[[:digit:]]+[[:space:]]*" > /dev/null 2>&1 ; then diffentries cddblocalread "$CDDBLOCALCHOICES" "$CDDBLOCALCHOICE" elif echo $CDDBLOCALCHOICE | grep -E "[[:space:]]*[[:digit:]]+[[:space:]]*" > /dev/null 2>&1 ; then # Make sure we get a valid choice CDDBLOCALCHOICENUM=$(echo $CDDBLOCALCHOICE | xargs printf %d 2>/dev/null) if [ $CDDBLOCALCHOICENUM -lt 0 ] || [ $CDDBLOCALCHOICENUM -gt $CDDBLOCALCHOICES ]; then echo "Invalid selection. Please choose a number between 0 and $CDDBLOCALCHOICES." >&2 fi fi done else ### FIXME ### #echo "Selected ..." CDDBLOCALRESP=y CDDBLOCALCHOICENUM=1 fi if [ ! "$CDDBLOCALCHOICENUM" = "0" ]; then #echo "Using local copy of CDDB data" echo "# DO NOT ERASE THIS LINE! Added by abcde to imitate cddb output" > "$ABCDETEMPDIR/cddbread.1" cat "$ABCDETEMPDIR/cddblocalread.$CDDBLOCALCHOICENUM" >> "$ABCDETEMPDIR/cddbread.1" echo 999 > "$ABCDETEMPDIR/cddbquery" # Assuming 999 isn't used by CDDB echo cddb-readcomplete >> "$ABCDETEMPDIR/status" do_cddbparse "$ABCDETEMPDIR/cddbread.1" > "$ABCDETEMPDIR/cddbchoices" echo cddb-choice=1 >> "$ABCDETEMPDIR/status" CDDBLOCALSTATUS="found" else #echo "Not using local copy of CDDB data" CDDBLOCALSTATUS="notfound" fi ;; single) # List out disc title/author and contents do_cddbparse "${CDDBLOCALFILE}" #if [ "$CDROMREADERSYNTAX" = "flac" ] ; then # echo -n "Embedded cuesheet entry found, use it [Y/n]? " >&2 #else echo -n "Locally cached CDDB entry found, use it [Y/n]? " >&2 #fi if [ "$INTERACTIVE" = "y" ]; then read USELOCALRESP while [ "$USELOCALRESP" != "y" ] && [ "$USELOCALRESP" != "n" ] && [ "$USELOCALRESP" != "" ] ; do echo -n 'Invalid selection. Please answer "y" or "n": ' >&2 read USELOCALRESP done [ x"$USELOCALRESP" = "x" ] && USELOCALRESP="y" else echo "y" >&2 fi if [ "$USELOCALRESP" = "y" ]; then #echo "Using local copy of CDDB data" echo "# DO NOT ERASE THIS LINE! Added by abcde to imitate cddb output" > "$ABCDETEMPDIR/cddbread.1" cat "${CDDBLOCALFILE}" >> "$ABCDETEMPDIR/cddbread.1" echo 999 > "$ABCDETEMPDIR/cddbquery" # Assuming 999 isn't used by CDDB echo cddb-readcomplete >> "$ABCDETEMPDIR/status" do_cddbparse "${CDDBLOCALFILE}" > "$ABCDETEMPDIR/cddbchoices" echo cddb-choice=1 >> "$ABCDETEMPDIR/status" CDDBLOCALSTATUS="single" else #echo "Not using local copy of CDDB data" CDDBLOCALSTATUS="notfound" fi ;; none) CDDBLOCALSTATUS="notfound" ;; esac fi } # do_musicbrainz # Work with the musicbrainz WS API, then transform the results here so # they look (very) like the results from CDDB. Maybe not the best way # to go, but it Works For Me (TM) do_musicbrainz () { if checkstatus musicbrainz-readcomplete; then :; else vecho "Obtaining Musicbrainz results..." # If MB is to be used, interpret the query results and read all # the available entries. rm -f "$ABCDETEMPDIR/cddbchoices" CDDBCHOICES=1 # Overridden by multiple matches MBDISCID=$(echo $TRACKINFO | cut -d' ' -f1) ${MUSICBRAINZ} --command data --discid "$MBDISCID" --workdir $ABCDETEMPDIR # The helper script will write disc matches out to # cddbread.*. Count how many we have NUM_RESPONSES=$(echo ${ABCDETEMPDIR}/cddbread.* | wc -w) if [ "$NUM_RESPONSES" -eq 1 ] ; then # One exact match echo -n "Retrieved 1 Musicbrainz match..." >> "$ABCDETEMPDIR/cddbchoices" echo "done." >> "$ABCDETEMPDIR/cddbchoices" echo cddb-read-1-complete >> "$ABCDETEMPDIR/status" echo cddb-choice=1 >> "$ABCDETEMPDIR/status" ATITLE=$(grep -e '^DTITLE=' ${ABCDETEMPDIR}/cddbread.1 | cut -c8- ) echo "200 none ${ATITLE}" >> "$ABCDETEMPDIR/cddbquery" # List out disc title/author and contents echo ---- ${ATITLE} ---- >> "$ABCDETEMPDIR/cddbchoices" for TRACK in $(f_seq_row 1 $TRACKS) do echo $TRACK: "$(grep ^TTITLE$(expr $TRACK - 1)= "$ABCDETEMPDIR/cddbread.1" | cut -f2- -d= | tr -d \\r\\n)" >> "$ABCDETEMPDIR/cddbchoices" done echo >> "$ABCDETEMPDIR/cddbchoices" elif [ "$NUM_RESPONSES" -eq 0 ] ; then # No matches. Use the normal cddb template for the user to # fill in echo "No Musicbrainz match." >> "$ABCDETEMPDIR/cddbchoices" $CDDBTOOL template $(cat "$ABCDETEMPDIR/discid") > "$ABCDETEMPDIR/cddbread.0" # List out disc title/author and contents of template echo ---- Unknown Artist / Unknown Album ---- >> "$ABCDETEMPDIR/cddbchoices" UNKNOWNDISK=y for TRACK in $(f_seq_row 1 $TRACKS) do echo $TRACK: "$(grep ^TTITLE$(expr $TRACK - 1)= "$ABCDETEMPDIR/cddbread.0" | cut -f2- -d= | tr -d \\r\\n)" >> "$ABCDETEMPDIR/cddbchoices" done echo >> "$ABCDETEMPDIR/cddbchoices" echo cddb-read-0-complete >> "$ABCDETEMPDIR/status" echo cddb-choice=0 >> "$ABCDETEMPDIR/status" echo 503 > "$ABCDETEMPDIR/cddbquery" else echo "210 Found exact matches, list follows (until terminating .)" > "$ABCDETEMPDIR/cddbquery" echo "Multiple Musicbrainz matches:" >> "$ABCDETEMPDIR/cddbchoices" for file in $ABCDETEMPDIR/cddbread.* do X=$(echo $file | sed 's/^.*cddbread\.//g') echo cddb-read-$X-complete >> "$ABCDETEMPDIR/status" ATITLE=$(grep -e '^DTITLE=' ${ABCDETEMPDIR}/cddbread.$X | cut -c8- ) echo "none ${ATITLE}" >> "$ABCDETEMPDIR/cddbquery" # List out disc title/author and contents echo "#$X: ---- ${ATITLE} ----" >> "$ABCDETEMPDIR/cddbchoices" for TRACK in $(f_seq_row 1 $TRACKS) do echo $TRACK: "$(grep ^TTITLE$(expr $TRACK - 1)= "$ABCDETEMPDIR/cddbread.$X" | cut -f2- -d= | tr -d \\r\\n)" >> "$ABCDETEMPDIR/cddbchoices" done echo >> "$ABCDETEMPDIR/cddbchoices" done echo "." >> "$ABCDETEMPDIR/cddbquery" fi echo "musicbrainz-readcomplete" >> "$ABCDETEMPDIR/status" fi } # do_cddbstat do_cddbstat () { # Perform CDDB protocol version check if it hasn't already been done if checkstatus cddb-statcomplete; then :; else if [ "$CDDBAVAIL" = "n" ]; then ERRORCODE=no_query echo 503 > "$ABCDETEMPDIR/cddbstat" else rc=1 CDDBUSER=$(echo $HELLOINFO | cut -f1 -d'@') CDDBHOST=$(echo $HELLOINFO | cut -f2- -d'@') while test $rc -eq 1 -a $CDDBPROTO -ge 3; do vecho "Checking CDDB server status..." $CDDBTOOL stat $CDDBURL $CDDBUSER $CDDBHOST $CDDBPROTO > "$ABCDETEMPDIR/cddbstat" RESPONSECODE=$(head -n 1 "$ABCDETEMPDIR/cddbstat" | cut -f1 -d' ') case "$RESPONSECODE" in 210) # 210 OK, status information follows (until terminating `.') rc=0 ;; 501) # 501 Illegal CDDB protocol level: . CDDBPROTO=`expr $CDDBPROTO - 1` ;; *) # Try a cddb query, since freedb2.org doesn't support the stat or ver commands # FreeDB TESTCD disc-id is used for query $CDDBTOOL query $CDDBURL $CDDBPROTO $CDDBUSER $CDDBHOST 03015501 1 296 344 > "$ABCDETEMPDIR/cddbstat" RESPONSECODE=$(head -n 1 "$ABCDETEMPDIR/cddbstat" | cut -f1 -d' ') case "$RESPONSECODE" in 2??) # Server responded, everything seems OK rc=0 ;; esac ;; esac done if test $rc -eq 1; then CDDBAVAIL="n" fi fi echo cddb-statcomplete >> "$ABCDETEMPDIR/status" fi } # do_cddbquery do_cddbquery () { CDDBDISCID=$(echo $TRACKINFO | cut -d' ' -f1) CDDBLOCALFILE="${CDDBLOCALDIR}/${CDDBDISCID}" # Perform CDDB query if it hasn't already been done if checkstatus cddb-querycomplete; then :; else if [ "$CDDBAVAIL" = "n" ]; then ERRORCODE=no_query echo 503 > "$ABCDETEMPDIR/cddbquery" # The default CDDBLOCALSTATUS is "notfound" # This part will be triggered if the user CDDB repo does not # contain the entry, or if we are not trying to use the repo. else vecho "Querying the CDDB server..." CDDBUSER=$(echo $HELLOINFO | cut -f1 -d'@') CDDBHOST=$(echo $HELLOINFO | cut -f2- -d'@') $CDDBTOOL query $CDDBURL $CDDBPROTO $CDDBUSER $CDDBHOST $TRACKINFO > "$ABCDETEMPDIR/cddbquery" ERRORCODE=$? case $ERRORCODE in 0) # success ;; 12|13|14) # no match found in database, # wget/fetch error, or user requested not to use CDDB # Make up an error code (503) that abcde # will recognize in do_cddbread # and compensate by making a template echo 503 > "$ABCDETEMPDIR/cddbquery" ;; *) # strange and unknown error echo ERRORCODE=$ERRORCODE echo "abcde: $CDDBTOOL returned unknown error code" ;; esac fi echo cddb-querycomplete >> "$ABCDETEMPDIR/status" fi } # do_cddbread do_cddbread () { # If it's not to be used, generate a template. # Then, display it (or them) and let the user choose/edit it if checkstatus cddb-readcomplete; then :; else vecho "Obtaining CDDB results..." # If CDDB is to be used, interpret the query results and read all # the available entries. rm -f "$ABCDETEMPDIR/cddbchoices" CDDBCHOICES=1 # Overridden by multiple matches RESPONSECODE=$(head -n 1 "$ABCDETEMPDIR/cddbquery" | cut -f1 -d' ') case "$RESPONSECODE" in 200) # One exact match, retrieve it # 200 [section] [discid] [artist] / [title] if checkstatus cddb-read-1-complete; then :; else echo -n "Retrieving 1 CDDB match..." >> "$ABCDETEMPDIR/cddbchoices" $CDDBTOOL read $CDDBURL $CDDBPROTO $CDDBUSER $CDDBHOST $(cut -f2,3 -d' ' "$ABCDETEMPDIR/cddbquery") > "$ABCDETEMPDIR/cddbread.1" echo "done." >> "$ABCDETEMPDIR/cddbchoices" echo cddb-read-1-complete >> "$ABCDETEMPDIR/status" echo cddb-choice=1 >> "$ABCDETEMPDIR/status" fi # List out disc title/author and contents echo ---- "$(cut '-d ' -f4- "$ABCDETEMPDIR/cddbquery")" ---- >> "$ABCDETEMPDIR/cddbchoices" for TRACK in $(f_seq_row 1 $TRACKS) do echo $TRACK: "$(grep ^TTITLE$(expr $TRACK - 1)= "$ABCDETEMPDIR/cddbread.1" | cut -f2- -d= | tr -d \\r\\n)" >> "$ABCDETEMPDIR/cddbchoices" done echo >> "$ABCDETEMPDIR/cddbchoices" ;; 202|403|409|503) # No match case "$RESPONSECODE" in 202) echo "No CDDB match." >> "$ABCDETEMPDIR/cddbchoices" ;; 403|409) echo "CDDB entry is corrupt, or the handshake failed." >> "$ABCDETEMPDIR/cddbchoices" ;; 503) echo "CDDB unavailable." >> "$ABCDETEMPDIR/cddbchoices" ;; esac $CDDBTOOL template $(cat "$ABCDETEMPDIR/discid") > "$ABCDETEMPDIR/cddbread.0" # List out disc title/author and contents of template echo ---- Unknown Artist / Unknown Album ---- >> "$ABCDETEMPDIR/cddbchoices" UNKNOWNDISK=y for TRACK in $(f_seq_row 1 $TRACKS) do echo $TRACK: "$(grep ^TTITLE$(expr $TRACK - 1)= "$ABCDETEMPDIR/cddbread.0" | cut -f2- -d= | tr -d \\r\\n)" >> "$ABCDETEMPDIR/cddbchoices" done echo >> "$ABCDETEMPDIR/cddbchoices" echo cddb-read-0-complete >> "$ABCDETEMPDIR/status" echo cddb-choice=0 >> "$ABCDETEMPDIR/status" ;; 210|211) # Multiple exact, (possibly multiple) inexact matches IN= if [ "$RESPONSECODE" = "211" ]; then IN=in; fi if [ "$(wc -l < "$ABCDETEMPDIR/cddbquery" | tr -d ' ')" -eq 3 ]; then echo "One ${IN}exact match:" >> "$ABCDETEMPDIR/cddbchoices" tail -n +2 "$ABCDETEMPDIR/cddbquery" | head -n 1 >> "$ABCDETEMPDIR/cddbchoices" echo cddb-choice=1 >> "$ABCDETEMPDIR/status" else echo "Multiple ${IN}exact matches:" >> "$ABCDETEMPDIR/cddbchoices" fi vecho -n "Retrieving multiple matches... " grep -v ^[.]$ "$ABCDETEMPDIR/cddbquery" | ( X=0 read DISCINFO # eat top line while read DISCINFO do X=$(expr $X + 1) if checkstatus cddb-read-$X-complete; then :; else $CDDBTOOL read $CDDBURL $CDDBPROTO $CDDBUSER $CDDBHOST $(echo $DISCINFO | cut -f1,2 -d' ') > "$ABCDETEMPDIR/cddbread.$X" echo cddb-read-$X-complete >> "$ABCDETEMPDIR/status" fi # List out disc title/author and contents echo \#$X: ---- "$DISCINFO" ---- >> "$ABCDETEMPDIR/cddbchoices" for TRACK in $(f_seq_row 1 $TRACKS) do echo $TRACK: "$(grep ^TTITLE$(expr $TRACK - 1)= "$ABCDETEMPDIR/cddbread.$X" | cut -f2- -d= | tr -d \\r\\n)" >> "$ABCDETEMPDIR/cddbchoices" done echo >> "$ABCDETEMPDIR/cddbchoices" done ) vecho "done." CDDBCHOICES=$(expr $(cat "$ABCDETEMPDIR/cddbquery" | wc -l) - 2) ;; 999) # Using local copy. for TRACK in $(f_seq_row 1 $TRACKS) do echo $TRACK: "$(grep ^TTITLE$(expr $TRACK - 1)= "$ABCDETEMPDIR/cddbread.1" | cut -f2- -d= | tr -d \\r\\n)" >> "$ABCDETEMPDIR/cddbchoices" done echo >> "$ABCDETEMPDIR/cddbchoices" echo cddb-read-1-complete >> "$ABCDETEMPDIR/status" echo cddb-choice=1 >> "$ABCDETEMPDIR/status" ;; esac echo "cddb-readcomplete" >> "$ABCDETEMPDIR/status" fi } # do_cddbedit do_cddbedit () { if checkstatus cddb-edit >/dev/null; then CDDBDATA="$ABCDETEMPDIR/cddbread.$(checkstatus cddb-choice)" VARIOUSARTISTS="$(checkstatus variousartists)" VARIOUSARTISTSTYLE="$(checkstatus variousartiststyle)" return 0 fi if [ "$INTERACTIVE" = "y" ]; then # We should show the CDDB results both when we are not using the local CDDB repo # or when we are using it but we could not find a proper match if [ "$CDDBUSELOCAL" = "y" ] && [ "$CDDBLOCALSTATUS" = "notfound" ] || [ ! "$CDDBUSELOCAL" = "y" ]; then # Display the $ABCDETEMPDIR/cddbchoices file created above # Pick a pager so that if the tracks overflow the screen the user can still view everything if [ -r "$ABCDETEMPDIR/cddbchoices" ]; then CDDBCHOICES=$(expr $(cat "$ABCDETEMPDIR/cddbquery" | wc -l) - 2) CHOICE=$(checkstatus cddb-choice) if [ -n "$CHOICE" ] ; then case $CDDBCHOICES in -1) if head -1 "$ABCDETEMPDIR/cddbquery" | grep "^$" > /dev/null 2>&1 ; then log error "CDDB query failed!" exit 1 else cat "$ABCDETEMPDIR/cddbchoices" fi ;; 1) cat "$ABCDETEMPDIR/cddbchoices" ;; *) echo "Selected: #$CHOICE" do_cddbparse "$ABCDETEMPDIR/cddbread.$CHOICE" ;; esac else # The user has a choice to make, display the info in a pager if necessary if [ $(cat "$ABCDETEMPDIR/cddbchoices" | wc -l) -ge 24 ]; then page "$ABCDETEMPDIR/cddbchoices" else # It's all going to fit in one page, cat it cat "$ABCDETEMPDIR/cddbchoices" >&2 fi CDDBCHOICENUM="" # Setting the choice to an impossible integer to avoid errors in the numeric comparisons CDCHOICENUM=-1 # I'll take CDDB read #3 for $400, Alex while [ $CDCHOICENUM -lt 0 ] || [ $CDCHOICENUM -gt $CDDBCHOICES ]; do echo -n "Which entry would you like abcde to use (0 for none)? [0-$CDDBCHOICES]: " >&2 read CDDBCHOICE [ X"$CDDBCHOICE" = "X" ] && CDDBCHOICE=1 if echo $CDDBCHOICE | grep -E "[[:space:]]*[[:digit:]]+,[[:digit:]]+[[:space:]]*" > /dev/null 2>&1 ; then if [ ! X"$DIFF" = "X" ]; then PARSECHOICE1=$(echo $CDDBCHOICE | cut -d"," -f1 | xargs printf %d 2>/dev/null) PARSECHOICE2=$(echo $CDDBCHOICE | cut -d"," -f2 | xargs printf %d 2>/dev/null) if [ $PARSECHOICE1 -lt 1 ] || [ $PARSECHOICE1 -gt $CDDBCHOICES ] || \ [ $PARSECHOICE2 -lt 1 ] || [ $PARSECHOICE2 -gt $CDDBCHOICES ] || \ [ $PARSECHOICE1 -eq $PARSECHOICE2 ]; then echo "Invalid diff range. Please select two comma-separated numbers between 1 and $CDDBCHOICES" >&2 else # We parse the 2 choices to diff, store them in temporary files and diff them. for PARSECHOICE in $(echo $CDDBCHOICE | tr , \ ); do do_cddbparse "$ABCDETEMPDIR/cddbread.$PARSECHOICE" > "$ABCDETEMPDIR/cddbread.parsechoice.$PARSECHOICE" done echo "Showing diff between choices $PARSECHOICE1 and $PARSECHOICE2..." > "$ABCDETEMPDIR/cddbread.diff" $DIFF $DIFFOPTS "$ABCDETEMPDIR/cddbread.parsechoice.$PARSECHOICE1" "$ABCDETEMPDIR/cddbread.parsechoice.$PARSECHOICE2" >> "$ABCDETEMPDIR/cddbread.diff" if [ $(cat "$ABCDETEMPDIR/cddbread.diff" | wc -l) -ge 24 ]; then page "$ABCDETEMPDIR/cddbread.diff" else cat "$ABCDETEMPDIR/cddbread.diff" >&2 fi fi else echo "The diff program was not found in your path. Please choose a number between 0 and $CDDBCHOICES." >&2 fi elif echo $CDDBCHOICE | grep -E "[[:space:]]*[[:digit:]]+[[:space:]]*" > /dev/null 2>&1 ; then # Make sure we get a valid choice CDCHOICENUM=$(echo $CDDBCHOICE | xargs printf %d 2>/dev/null) if [ $CDCHOICENUM -lt 0 ] || [ $CDCHOICENUM -gt $CDDBCHOICES ]; then echo "Invalid selection. Please choose a number between 0 and $CDDBCHOICES." >&2 fi fi done if [ "$CDCHOICENUM" = "0" ]; then vecho "Creating empty CDDB template..." UNKNOWNDISK=y $CDDBTOOL template $(cat "$ABCDETEMPDIR/discid") > "$ABCDETEMPDIR/cddbread.0" else echo "Selected: #$CDCHOICENUM ($(grep ^DTITLE= "$ABCDETEMPDIR/cddbread.$CDCHOICENUM" | cut -f2- -d= | tr -d \\r\\n))" >&2 do_cddbparse "$ABCDETEMPDIR/cddbread.$CDCHOICENUM" fi echo "cddb-choice=$CDCHOICENUM" >> "$ABCDETEMPDIR/status" fi fi else # We need some code to show the selected option when local repository is selected and we have found a match vecho "Using cached CDDB match..." >&2 # Display the $ABCDETEMPDIR/cddbchoices file created above # Pick a pager so that if the tracks overflow the screen the user can still view everything if [ -r "$ABCDETEMPDIR/cddbchoices" ]; then CDDBCHOICES=$(expr $(cat "$ABCDETEMPDIR/cddbquery" | wc -l) - 2) CHOICE=$(checkstatus cddb-choice) if [ "$USELOCALRESP" = "y" ]; then :; else if [ -n "$CHOICE" ] ; then case $CDDBCHOICES in 0) UNKNOWNDISK=y echo "Selected template." ;; 1) cat "$ABCDETEMPDIR/cddbchoices" ;; *) echo "Selected: #$CHOICE" do_cddbparse "$ABCDETEMPDIR/cddbread.$CHOICE" ;; esac fi fi fi fi else # We're noninteractive - pick the first choice. # But in case we run a previous instance and selected a choice, use it. if [ -r "$ABCDETEMPDIR/cddbchoices" ]; then # Show the choice if we are not using the locally stored one # or when the local search failed to find a match. PREVIOUSCHOICE=$(checkstatus cddb-choice) if [ "$CDDBUSELOCAL" = "y" ] && [ "$CDDBLOCALSTATUS" = "notfound" ] || [ ! "$CDDBUSELOCAL" = "y" ]; then #if [ "$PREVIOUSCHOICE" ]; then cat "$ABCDETEMPDIR/cddbchoices" #fi fi if [ ! -z "$PREVIOUSCHOICE" ] ; then CDCHOICENUM=$PREVIOUSCHOICE else CDCHOICENUM=1 echo "cddb-choice=$CDCHOICENUM" >> "$ABCDETEMPDIR/status" fi echo "Selected: #$CDCHOICENUM ($(grep ^DTITLE= "$ABCDETEMPDIR/cddbread.$CDCHOICENUM" | cut -f2- -d= | tr -d \\r\\n))" >&2 fi fi # sanity check if checkstatus cddb-choice >/dev/null; then :; else echo "abcde: internal error: cddb-choice not recorded." >&2 exit 1 fi CDDBDATA="$ABCDETEMPDIR/cddbread.$(checkstatus cddb-choice)" echo -n "Edit selected CDDB data? [y/n] (" >&2 if [ "$INTERACTIVE" = "y" ]; then if [ "$UNKNOWNDISK" = "y" ]; then echo -n "y): " >&2 read EDITCDDB [ "$EDITCDDB" != "n" ] && EDITCDDB=y else echo -n "n): " >&2 read EDITCDDB fi else echo "n): n" >&2 EDITCDDB=n fi if [ "$EDITCDDB" = "y" ]; then CDDBDATAMD5SUM=$($MD5SUM "$CDDBDATA" | cut -d " " -f 1); # Use the debian sensible-editor wrapper to pick the editor that the # user has requested via their $EDITOR environment variable if [ -x "/usr/bin/sensible-editor" ]; then /usr/bin/sensible-editor "$CDDBDATA" elif [ -n "$EDITOR" ]; then if [ -x $(which "${EDITOR%%\ *}") ]; then # That failed, try to load the preferred editor, starting # with their EDITOR variable eval $(echo "$EDITOR") \"$CDDBDATA\" fi # If that fails, check for a vi elif which vi >/dev/null 2>&1; then vi "$CDDBDATA" elif [ -x /usr/bin/vim ]; then /usr/bin/vim "$CDDBDATA" elif [ -x /usr/bin/vi ]; then /usr/bin/vi "$CDDBDATA" elif [ -x /bin/vi ]; then /bin/vi "$CDDBDATA" # nano should be on all (modern, i.e., sarge) debian systems elif which nano >/dev/null 2>&1 ; then nano "$CDDBDATA" elif [ -x /usr/bin/nano ]; then /usr/bin/nano "$CDDBDATA" # mg should be on all OpenBSD systems elif which mg >/dev/null 2>&1 ; then mg "$CDDBDATA" elif [ -x /usr/bin/mg ]; then /usr/bin/mg "$CDDBDATA" # bomb out else log warning "no editor available. Check your EDITOR environment variable." fi # delete editor backup file if it exists if [ -w "$CDDBDATA~" ]; then rm -f "$CDDBDATA~" fi fi # Some heuristics first. Look at Disc Title, and if it starts with # "Various", then we'll assume Various Artists if [ "$(grep ^DTITLE= "$CDDBDATA" | cut -f2- -d= | grep -Eci '^(various|soundtrack|varios|sonora|ost)')" != "0" ]; then echo "Looks like a Multi-Artist CD" >&2 VARIOUSARTISTS=y else echo -n "Is the CD multi-artist [y/N]? " >&2 if [ "$INTERACTIVE" = "y" ]; then read VARIOUSARTISTS else echo n >&2 VARIOUSARTISTS=n fi fi if [ "$VARIOUSARTISTS" = "y" ] && [ ! "$ONETRACK" = "y" ]; then # Set a default DEFAULTSTYLE=1 # Need NUMTRACKS before cddb-tool will return it: NUMTRACKS=$(grep -E '^TTITLE[0-9]+=' "$CDDBDATA" | wc -l) if [ "$(grep -c "^TTITLE.*\/" "$CDDBDATA")" -gt "$(expr $NUMTRACKS / 2 )" ]; then # More than 1/2 tracks contain a "/", so guess forward DEFAULTSTYLE=1 elif [ "$(grep -c "^TTITLE.*\-" "$CDDBDATA")" -gt "$(expr $NUMTRACKS / 2 )" ]; then # More than 1/2 contain a "-", so guess forward-dash DEFAULTSTYLE=2 elif [ "$(grep -c "^TTITLE.*(.*)" "$CDDBDATA")" -gt "$(expr $NUMTRACKS / 2 )" ]; then # More than 1/2 contain something in parens, so guess trailing-paren DEFAULTSTYLE=6 fi echo "1) Artist / Title" >&2 echo "2) Artist - Title" >&2 echo "3) Title / Artist" >&2 echo "4) Title - Artist" >&2 echo "5) Artist: Title" >&2 echo "6) Title (Artist)" >&2 echo "7) This is a single-artist CD" >&2 echo -n "Which style of multiple artist entries is it? [1-7] ($DEFAULTSTYLE): " >&2 if [ "$INTERACTIVE" = "y" ]; then read VARIOUSARTISTSTYLE else echo $DEFAULTSTYLE >&2 VARIOUSARTISTSTYLE=$DEFAULTSTYLE fi VARIOUSARTISTSTYLE=$(echo 0$VARIOUSARTISTSTYLE | xargs printf %d) # If they press Enter, then the default style (0) was chosen while [ $VARIOUSARTISTSTYLE -lt 0 ] || [ $VARIOUSARTISTSTYLE -gt 7 ]; do echo "Invalid selection. Please choose a number between 1 and 7." echo -n "Selection [1-7]: " read VARIOUSARTISTSTYLE VARIOUSARTISTSTYLE=$(echo 0$VARIOUSARTISTSTYLE | xargs printf %d) done if [ "$VARIOUSARTISTSTYLE" = "0" ]; then VARIOUSARTISTSTYLE=$DEFAULTSTYLE fi vecho "Selected: $VARIOUSARTISTSTYLE" case "$VARIOUSARTISTSTYLE" in 1) # Artist / Title VARIOUSARTISTSTYLE=forward ;; 2) # Artist - Title VARIOUSARTISTSTYLE=forward-dash ;; 3) # Title / Artist VARIOUSARTISTSTYLE=reverse ;; 4) # Title - Artist VARIOUSARTISTSTYLE=reverse-dash ;; 5) # Artist: Title VARIOUSARTISTSTYLE=colon ;; 6) # Title (Artist) VARIOUSARTISTSTYLE=trailing-paren ;; 7) # Single Artist VARIOUSARTISTS=n ;; esac fi echo "variousartists=$VARIOUSARTISTS" >> "$ABCDETEMPDIR/status" echo "variousartiststyle=$VARIOUSARTISTSTYLE" >> "$ABCDETEMPDIR/status" if [ "$EDITCDDB" = "y" ] && [ "$CDDBMETHOD" = "cddb" ] && [ "$UNINTENTIONALLY_ANGER_THE_FREEDB_PEOPLE" = "y" ]; then if [ "$CDDBDATAMD5SUM" != "" ] && [ "$CDDBDATAMD5SUM" != "$($MD5SUM "$CDDBDATA" | cut -d " " -f 1)" ]; then # This works but does not have the necessary error checking # yet. If you are familiar with the CDDB spec # (see http://www.freedb.org/src/latest/DBFORMAT) # and can create an error-free entry on your own, then put # UNINTENTIONALLY_ANGER_THE_FREEDB_PEOPLE=y in your # abcde.conf to enable it. Put CDDBSUBMIT=email@address in # your abcde.conf to change the email address submissions are # sent to. # submit the modified file, if they want if [ "$NOSUBMIT" != "y" ]; then echo -n "Do you want to submit this entry to $CDDBSUBMIT [y/N]? " read YESNO while [ "$YESNO" != "y" ] && [ "$YESNO" != "n" ] && [ "$YESNO" != "Y" ] && \ [ "$YESNO" != "N" ] && [ "$YESNO" != "" ] do echo -n 'Invalid selection. Please answer "y" or "n": ' read YESNO done if [ "$YESNO" = "y" ] || [ "$YESNO" = "Y" ]; then echo -n "Sending..." $CDDBTOOL send "$CDDBDATA" $CDDBSUBMIT echo "done." fi fi fi fi ### FIXME ### # User CDDBLOCALPOLICY to find out if we store the file or not... # Cache edited CDDB entry in the user's cddb dir if [ "$CDDBCOPYLOCAL" = "y" ]; then # Make sure the cache directory exists mkdir -p $CDDBLOCALDIR cat "$CDDBDATA" | tail -n $(expr $(cat "$CDDBDATA" | wc -l ) - 1 ) > ${CDDBLOCALDIR}/$(echo "$TRACKINFO" | cut -d' ' -f1) fi echo "cddb-edit" >> "$ABCDETEMPDIR/status" } # do_cdread [tracknumber] # do_cdread onetrack [firsttrack] [lasttrack] # do_cdread () { # The commands here don't go through run_command because they're never supposed to be silenced # return codes need to be doublechecked anyway, however if [ "$1" = "onetrack" ]; then # FIXME # Add the possibility of grabbing ranges of tracks in onetrack # FIXME # Until then, we grab the whole CD in one track, no matter what # the user said # We need the first and last track for cdda2wav/icedax FIRSTTRACK=$2 LASTTRACK=$(expr $3 + 0) UTRACKNUM=$FIRSTTRACK case "$CDROMREADERSYNTAX" in flac) READTRACKNUMS="$FIRSTTRACK.1-$(($LASTTRACK + 1)).0" ;; cdparanoia) #XX FIXME XX # Add a variable to check if tracks are provided in command line and if not, use "0-" to rip the tracks READTRACKNUMS="$FIRSTTRACK-$LASTTRACK" ;; cdda2wav | icedax) READTRACKNUMS="$FIRSTTRACK+$LASTTRACK" ;; *) echo "abcde error: $CDROMREADERSYNTAX does not support ONETRACK mode" exit 1 ;; esac else UTRACKNUM=$1 fi CDDBTRACKNUM=$(expr $UTRACKNUM - 1) if [ "$USEPIPES" = "y" ]; then TEMPARG="PIPERIPPER_$CDROMREADERSYNTAX" FILEARG="$( eval echo "\$$TEMPARG" )" REDIR="y" PIPE_MESSAGE="and encoding " else WAVDATA="$ABCDETEMPDIR/track$UTRACKNUM.wav" case "$CDROMREADERSYNTAX" in ## FIXME ## Find the cases for dagrab and flac, to avoid exceptions flac) FILEARG="--output-name=$WAVDATA" ;; dagrab) FILEARG="-f $WAVDATA" ;; *) FILEARG="$WAVDATA" ;; esac REDIR="n" fi if [ "$1" = "onetrack" ]; then echo "Grabbing ${PIPE_MESSAGE}tracks $UTRACKNUM - $LASTTRACK as one track ..." >&2 else if [ -r "$CDDBDATA" ]; then getcddbinfo TRACKNAME echo "Grabbing ${PIPE_MESSAGE}track $UTRACKNUM: $TRACKNAME..." >&2 else echo "Grabbing ${PIPE_MESSAGE}track $UTRACKNUM..." >&2 fi fi case "$CDROMREADERSYNTAX" in ### FIXME ### use an exception for flac, since it uses -o ### FIXME ### Shall we just use -o $FILEARG ?? flac) # Avoid problems wit math expressions by unpadding the given UTRACKNUM STRIPTRACKNUM=$(expr $UTRACKNUM + 0) nice $READNICE $FLAC -d -f --cue=${READTRACKNUMS:-$STRIPTRACKNUM.1-$(($STRIPTRACKNUM + 1)).0} "$FILEARG" "$CDROM" ;; cdparanoia) if [ "$REDIR" = "y" ]; then nice $READNICE $CDROMREADER -$CDPARANOIACDROMBUS "$CDROM" ${READTRACKNUMS:-$UTRACKNUM} "$FILEARG" >&2 else nice $READNICE $CDROMREADER -$CDPARANOIACDROMBUS "$CDROM" ${READTRACKNUMS:-$UTRACKNUM} "$FILEARG" fi;; cdda2wav | icedax) if [ "$OSFLAVOUR" = "OSX" ] ; then # Hei, we have to unmount the device before running anything like cdda2wav/icedax in OSX disktool -u ${CDROM#/dev/} 0 # Also, in OSX the cdrom device for cdda2wav/icedax changes... CDDA2WAVCDROM="IODVDServices" elif [ "$OSFLAVOUR" = "FBSD" ] ; then CDDA2WAVCDROM="$CDROMID" else if [ "$CDROMID" = "" ]; then CDDA2WAVCDROM="$CDROM" else CDDA2WAVCDROM="$CDROMID" fi fi if [ "$REDIR" = "y" ]; then nice $READNICE $CDROMREADER -D $CDDA2WAVCDROM -t ${READTRACKNUMS:-$UTRACKNUM} "$FILEARG" >&2 else nice $READNICE $CDROMREADER -D $CDDA2WAVCDROM -t ${READTRACKNUMS:-$UTRACKNUM} "$FILEARG" fi ;; ## FIXME ## We have an exception for dagrab, since it uses -f ## FIXME ## Shall we just use -f $FILEARG ?? dagrab) if [ "$REDIR" = "y" ]; then nice $READNICE $CDROMREADER -d "$CDROM" -v $UTRACKNUM "$FILEARG" >&2 else nice $READNICE $CDROMREADER -d "$CDROM" -v $UTRACKNUM "$FILEARG" fi ;; cddafs) # Find the track's mounted path REALTRACKNUM=$(expr $UTRACKNUM + 0) FILEPATH=$(mount | grep "$CDROM on" | sed 's/^[^ ]* on \(.*\) (.*/\1/') FILEPATH=$(find "$FILEPATH" | grep "/$REALTRACKNUM "); # If the file exists, copy it if [ -e "$FILEPATH" ] ; then if [ "$REDIR" = "y" ]; then nice $READNICE $CDROMREADER "$FILEPATH" "$FILEARG" >&2 else nice $READNICE $CDROMREADER "$FILEPATH" "$FILEARG" fi else false fi ;; debug) if [ "$REDIR" = "y" ]; then nice $READNICE $CDROMREADER -$CDPARANOIACDROMBUS "$CDROM" -w $UTRACKNUM-[:1] "$FILEARG" >&2 else nice $READNICE $CDROMREADER -$CDPARANOIACDROMBUS "$CDROM" -w $UTRACKNUM-[:1] "$FILEARG" fi ;; esac RETURN=$? # If we get some error or we get some missing wav # (as long as we dont use pipes) if [ "$RETURN" != "0" -o \( ! -s "$WAVDATA" -a X"$USEPIPES" != "Xy" \) ]; then # Thank goodness errors is only machine-parseable up to the # first colon, otherwise this woulda sucked if [ "$RETURN" = "0" -a ! -s "$WAVDATA" ]; then RETURN=73 # fake a return code as cdparanoia return 0 also on aborted reads fi if [ "$USEPIPES" = "y" ]; then echo "readencodetrack-$UTRACKNUM: $CDROMREADER returned code $RETURN" >> "$ABCDETEMPDIR/errors" else echo "readtrack-$UTRACKNUM: $CDROMREADER returned code $RETURN" >> "$ABCDETEMPDIR/errors" fi return $RETURN else if [ "$USEPIPES" = "y" ]; then echo readencodetrack-$UTRACKNUM >> "$ABCDETEMPDIR/status" else echo readtrack-$UTRACKNUM >> "$ABCDETEMPDIR/status" fi if [ "$1" = "onetrack" ]; then echo onetrack >> "$ABCDETEMPDIR/status" fi fi } # do_cdspeed # No values accepted, only uses env variables do_cdspeed () { if "$CDSPEED" "$CDSPEEDOPTS" "$CDSPEEDVALUE" >/dev/null ; then vecho "Setting CD speed to ${CDSPEEDVALUE}x" else echo "abcde: unable to set the device speed" >&2 fi } # vecho [message] # # vecho outputs a message if EXTRAVERBOSE is selected vecho () { if [ x"$EXTRAVERBOSE" != "x" ]; then case $1 in warning) shift ; log warning "$@" ;; *) echo "$@" ;; esac fi } # decho [message] # # decho outputs a debug message if DEBUG is selected decho () { if [ x"$DEBUG" != "x" ]; then if echo $1 | grep "^\[" > /dev/null 2>&1 ; then DEBUGECHO=$(echo "$@" | tr -d '[]') echo "[DEBUG] $DEBUGECHO: `eval echo \\$${DEBUGECHO}`" else echo "[DEBUG] $1" fi fi } # User-redefinable functions # Custom filename munging: mungefilename () { #echo "$@" | sed s,:,\ -,g | tr \ /\* __+ | tr -d \'\"\?\[:cntrl:\] echo "$@" | sed s,:,\ -,g | tr \ / __ | tr -d \'\"\?\[:cntrl:\] } # Custom genre munging: mungegenre () { echo $CDGENRE | tr "[:upper:]" "[:lower:]" } # pre_read # Empty pre_read function, to be defined in the configuration file. pre_read () { : } # post_read # Empty post_read function, to be defined in the configuration file. post_read () { : } # post_encode # Empty post_encode function, to be defined in the configuration file. post_encode () { : } ############################################################################### # End of functions # # Start of execution ############################################################################### # Builtin defaults # CDDB # Defaults to FreeDB, but musicbrainz can be used too, via the abcde-musicbrainz-tool script CDDBMETHOD=cddb CDDBURL="http://freedb.freedb.org/~cddb/cddb.cgi" CDDBSUBMIT=freedb-submit@freedb.org CDDBPROTO=6 HELLOINFO="$(whoami)@$(hostname)" CDDBCOPYLOCAL="n" CDDBLOCALPOLICY="always" CDDBLOCALRECURSIVE="y" CDDBLOCALDIR="$HOME/.cddb" CDDBUSELOCAL="n" # List of fields we parse and show during the CDDB parsing... SHOWCDDBFIELDS="year,genre" INTERACTIVE=y #CDROMREADERSYNTAX=cdparanoia ENCODERSYNTAX=default MP3ENCODERSYNTAX=default OGGENCODERSYNTAX=default FLACENCODERSYNTAX=default SPEEXENCODERSYNTAX=default MPPENCODERSYNTAX=default AACENCODERSYNTAX=default NORMALIZERSYNTAX=default CUEREADERSYNTAX=default OUTPUTFORMAT='${ARTISTFILE}-${ALBUMFILE}/${TRACKNUM}.${TRACKFILE}' # Use the following VAOUTPUTFORMAT to revert to 2.0.x VA format: #VAOUTPUTFORMAT=${OUTPUTFORMAT} VAOUTPUTFORMAT='Various-${ALBUMFILE}/${TRACKNUM}.${ARTISTFILE}-${TRACKFILE}' ONETRACKOUTPUTFORMAT='${ARTISTFILE}-${ALBUMFILE}/${ALBUMFILE}' VAONETRACKOUTPUTFORMAT='Various-${ALBUMFILE}/${ALBUMFILE}' PLAYLISTFORMAT='${ARTISTFILE}-${ALBUMFILE}.${OUTPUT}.m3u' PLAYLISTDATAPREFIX='' VAPLAYLISTFORMAT='${ARTISTFILE}-${ALBUMFILE}.${OUTPUT}.m3u' VAPLAYLISTDATAPREFIX='' DOSPLAYLIST=n COMMENT='' ID3TAGV=2 ENCNICE=10 READNICE=10 DISTMP3NICE=10 VARIOUSARTISTS=n VARIOUSARTISTSTYLE=forward KEEPWAVS=n PADTRACKS=n NOGAP=n BATCHNORM=n NOCDDBQUERY=n # If using scsi devices, cdda2wav/icedax needs a CDROMID, instead of a device node # i.e. CDROMID="1,0,0" CDROMID="" # If we are using the IDE bus, we need CDPARANOIACDROMBUS defined as "d" # If we are using the ide-scsi emulation layer, we need to define a "g" CDPARANOIACDROMBUS="d" # program paths - defaults to checking your $PATH # mp3 LAME=lame TOOLAME=toolame GOGO=gogo BLADEENC=bladeenc L3ENC=l3enc XINGMP3ENC=xingmp3enc MP3ENC=mp3enc # ogg VORBIZE=vorbize OGGENC=oggenc # flac FLAC=flac # speex SPEEXENC=speexenc # mpp (Musepack) MPPENC=mppenc # m4a AACENC=faac ID3=id3 ID3V2=id3v2 EYED3=eyeD3 VORBISCOMMENT=vorbiscomment METAFLAC=metaflac AACTAG=faac ATOMICPARSLEY=AtomicParsley CDPARANOIA=cdparanoia CDDA2WAV=icedax DAGRAB=dagrab CDDAFS=cp CDDISCID=cd-discid CDDBTOOL=cddb-tool MUSICBRAINZ=abcde-musicbrainz-tool EJECT=eject MD5SUM=md5sum DISTMP3=distmp3 NORMALIZE=normalize-audio CDSPEED=eject VORBISGAIN=vorbisgain MP3GAIN=mp3gain MPPGAIN=replaygain MKCUE=mkcue MKTOC=cdrdao DIFF=diff CUE2DISCID=builtin # Options for programs called from abcde # mp3 LAMEOPTS= TOOLAMEOPTS= GOGOOPTS= BLADEENCOPTS= L3ENCOPTS= XINGMP3ENCOPTS= MP3ENCOPTS= MP3GAINOPTS="-a -k" # ogg VORBIZEOPTS= OGGENCOPTS= VORBISGAINOPTS="--album" # flac FLACOPTS= FLACGAINOPTS="--add-replay-gain" # speex SPEEXENCOPTS= # mpc MPPENCOPTS= # m4a AACENCOPTS= ID3OPTS= ID3V2OPTS= CDPARANOIAOPTS= CDDA2WAVOPTS= DAGRABOPTS= CDDAFSOPTS="-f" CDDBTOOLOPTS= EJECTOPTS= DISTMP3OPTS= NORMALIZEOPTS= CDSPEEDOPTS="-x" CDSPEEDVALUE= MKCUEOPTS= MKTOCOPTS="" VORBISCOMMENTOPTS="-R" METAFLACOPTS="--no-utf8-convert" DIFFOPTS= # Default to one process if -j isn't specified MAXPROCS=1 # List of actions to perform - by default, run to completion ACTIONS=cddb,read,encode,tag,move,clean # This option is basicaly for Debian package dependencies: # List of prefered outputs - by default, run with whatever we have in the path DEFAULT_OUTPUT_BINARIES=vorbis:oggenc,flac:flac,mp3:toolame,mp3:lame,mp3:bladeenc,spx:speex,m4a:faac # List of prefered cdromreaders - by default, run whichever we have in the path DEFAULT_CDROMREADERS="cdparanoia icedax cdda2wav" # List of quality levels associated with the encoders: DEFAULT_QUALITY_XLO="oggenc:-q -1,lame:-q 9,speex:--quality 1,m4a:" DEFAULT_QUALITY_LO="oggenc:-q 1,lame:-q 7,speex:--quality 5,m4a:" DEFAULT_QUALITY_HI="oggenc:-q 7,lame:--preset standard,speex:--quality 9,m4a:" DEFAULT_QUALITY_XHI="oggenc:-q 10,lame:--preset extreme,speex:--quality 10,m4a:" # Asume fetch if under FreeBSD. curl is used for Mac OS X. wget is used for # Linux/OpenBSD. ftp is user for NetBSD. # Let's use these checkings to determine the OS flavour, which will be used # later if [ X$(uname) = "XFreeBSD" ] ; then HTTPGET=fetch MD5SUM=md5 NEEDCDROMID=y OSFLAVOUR=FBSD elif [ X$(uname) = "XDarwin" ] ; then HTTPGET=curl OSFLAVOUR=OSX # We should have disktool in OSX, but let's be sure... NEEDDISKTOOL=y CDROMREADERSYNTAX=cddafs elif [ X$(uname) = "XOpenBSD" ] ; then HTTPGET=wget MD5SUM=md5 OSFLAVOUR=OBSD elif [ X$(uname) = "XNetBSD" ] ; then HTTPGET=ftp MD5SUM=md5 OSFLAVOUR=NBSD elif [ X$(uname) = "SunOS" ] ; then HTTPGET="" MD5SUM=md5 OSFLAVOUR=SunOS else HTTPGET=wget fi # If CDDBAVAIL is set to n, no CDDB read is done # If USEID3 is set to n, no ID3 tagging is done CDDBAVAIL=y USEID3=y USEID3V2=y # There we go.. # But before we get into business, let us chop off any GREP environmental # variables. GREP_OPTIONS="" GREP_COLOR="" if [ -z "$OUTPUTDIR" ]; then OUTPUTDIR=$(pwd) fi if [ -z "$WAVOUTPUTDIR" ]; then WAVOUTPUTDIR="$OUTPUTDIR" fi # Load system defaults if [ -r /etc/abcde.conf ]; then . /etc/abcde.conf fi # Load user preference defaults if [ -r $HOME/.abcde.conf ]; then . $HOME/.abcde.conf fi # By this time, we need some HTTPGETOPTS already defined. # If the user has defined a non-default HTTPGET method, we should not be empty. if [ "$HTTPGETOPTS" = "" ] ; then case $HTTPGET in wget) HTTPGETOPTS="-q -nv -e timestamping=off -O -";; curl) HTTPGETOPTS="-f -s";; fetch)HTTPGETOPTS="-q -o -";; ftp) HTTPGETOPTS="-a -V -o - ";; *) log warning "HTTPGET in non-standard and HTTPGETOPTS are not defined." ;; esac fi # If the CDROM has not been set yet, find a suitable one. # If this is a devfs system, default to /dev/cdroms/cdrom0 # instead of /dev/cdrom if [ "$CDROM" = "" ] ; then if [ -e /dev/cdroms/cdrom0 ]; then CDROM=/dev/cdroms/cdrom0 elif [ -e /dev/cdrom ]; then CDROM=/dev/cdrom elif [ -e /dev/cd0c ]; then CDROM=/dev/cd0c elif [ -e /dev/acd0c ]; then CDROM=/dev/acd0c elif [ -e /dev/disk1 ]; then CDROM=/dev/disk1 fi fi # Parse command line options #while getopts 1a:bc:C:d:Dehj:klLmMnNo:pPq:r:Rs:S:t:T:vVxw:W: opt ; do while getopts 1a:bBc:C:d:Defghj:klLmMnNo:pPq:r:s:S:t:T:UvVxX:w:W:z opt ; do case "$opt" in 1) ONETRACK=y ;; a) ACTIONS="$OPTARG" ;; A) EXPACTIONS="$OPTARG" ;; b) BATCHNORM=y ;; B) NOBATCHREPLAYGAIN=y ;; c) if [ -e "$OPTARG" ] ; then . "$OPTARG" ; else log error "config file \"$OPTARG\" cannot be found." ; exit 1 ; fi ;; C) DISCID="$( echo ${OPTARG#abcde.} | tr -d /)" ;; d) CDROM="$OPTARG" ;; D) set -x ;; h) usage; exit ;; e) ERASEENCODEDSTATUS=y ;; E) ENCODING="$OPTARG" ;; f) FORCE=y ;; g) NOGAP=y ;; i) INLINETAG=y ;; j) MAXPROCS="$OPTARG" ;; k) KEEPWAVS=y ;; l) LOWDISK=y ;; L) CDDBUSELOCAL=y ;; n) CDDBAVAIL=n ;; N) INTERACTIVE=n ;; m) DOSPLAYLIST=y ;; M) DOCUE=y ;; o) OUTPUTTYPE="$OPTARG" ;; p) PADTRACKS=y ;; P) USEPIPES=y ;; q) QUALITYLEVEL="$OPTARG" ;; r) REMOTEHOSTS="$OPTARG" ;; R) CDDBLOCALRECURSIVE=y ;; s) SHOWCDDBFIELDS="$OPTARG" ;; S) CDSPEEDVALUE="$OPTARG" ;; t) STARTTRACKNUMBER="$OPTARG" ;; T) STARTTRACKNUMBER="$OPTARG" ; STARTTRACKNUMBERTAG="y" ;; U) CDDBPROTO=5 ;; v) echo "This is abcde v$VERSION." echo "Usage: abcde [options] [tracks]" echo "abcde -h for extra help" exit ;; V) EXTRAVERBOSE="y" ;; x) EJECTCD="y" ;; X) CUE2DISCID="$OPTARG" ;; w) COMMENT="$OPTARG" ;; W) if echo $OPTARG | grep "[[:digit:]]" > /dev/null 2>&1 ; then STARTTRACKNUMBER="${OPTARG}01" STARTTRACKNUMBERTAG="y" COMMENT="CD${OPTARG}" DISCNUMBER="${OPTARG}" else log error "argument of -W must be integer" exit 1 fi ;; z) DEBUG=y ; CDROMREADERSYNTAX=debug ; EJECTCD="n" ;; ?) usage; exit ;; esac done shift $(($OPTIND - 1)) # Here it comes the worse part of the whole thing. From here until we start # ripping, we have a mixture of sanity checks, verifications, default settigs # and other crazy stuff that interdepends, but hey, someone has to do it. # If NOCDDBQUERY is set, don't query the CDDB server. if [ "$NOCDDBQUERY" = "y" ]; then CDDBAVAIL="n" fi # If the user specified a flac file, then switch to special flac mode if echo "$CDROM" | grep -i '.flac$' > /dev/null 2>&1 ; then if [ ! -f "$CDROM" ]; then log error "the defined file for FLAC ripping cannot be found" >&2 exit 1 fi vecho warning "switching to flac CDROMREADERSYNTAX..." CDROMREADERSYNTAX=flac # We have a builtin version of cue2discid... case "$CUE2DISCID" in builtin);; *) NEEDCUE2DISCID=y;; esac NEEDMETAFLAC=y EJECTCD=n fi # If the user provided a DISCID, disable eject if [ -n "$DISCID" ] || [ "$CDROMREADERSYNTAX" = "flac" ]; then EJECTCD=n ; fi # Check the available cd rippers in the system, from the ones we know. if [ "$CDROMREADERSYNTAX" = "" ]; then for DEFAULT_CDROMREADER in $DEFAULT_CDROMREADERS; do if new_checkexec $DEFAULT_CDROMREADER; then CDROMREADERSYNTAX=$DEFAULT_CDROMREADER break fi done if [ "$CDROMREADERSYNTAX" = "" ]; then log error "no cdreader found in your PATH" log error "hints: are all dependencies installed? has the \$PATH been modified?" exit 1 fi fi # Decide if we can continue. if [ "$ONETRACK" = "y" ]; then # FIXME # remove check as soon as we find out about the other readers case "$CDROMREADERSYNTAX" in flac) ;; cdparanoia) ;; cdda2wav | icedax) ;; *) log error "$CDROMREADERSYNTAX does not support ONETRACK mode" exit 1 ;; esac if [ "$BATCHNORM" = "y" ]; then log warning "BATCHNORM mode is not compatible with ONETRACK mode. Disabling..." BATCHNORM=n fi if [ "$NOGAP" = "y" ]; then log warning "NOGAP mode is not compatible with ONETRACK mode. Disabling..." NOGAP=n fi # It does not matter how many tracks we want. In ONETRACK mode we grab them all # FIXME # allow ranges of tracks to be selected for onetrack ripping if [ $# -gt 0 ]; then log warning "ONETRACK mode selected, grabbing all tracks..." fi else while [ $# -gt 0 ]; do # Range parsing code courtesy of Vincent Ho # Cleaned up to use shell built-ins by Charles Steinkuehler if [ "${1#*[^0-9-]}" != "$1" ]; then log error "syntax error while processing track numbers" else RSTART=${1%%-*} REND=${1##*-} while [ ${RSTART:=1} -le ${REND:=0} ] ; do TRACKQUEUE="$TRACKQUEUE $RSTART" RSTART=$(( $RSTART + 1 )) done fi shift done fi # List of valid actions: cddb,read,normalize,encode,tag,move,playlist,clean # List of experimental actions: retag,transcode # Determine what actions are to be done from $ACTIONS and set the # following environment variables for them: DOCDDB=n DOREAD=n DONORMALIZE=n DOPREPROCESS=n DOENCODE=n DOPOSTPROCESS=n DOTAG=n DOMOVE=n DOREPLAYGAIN=n DOPLAYLIST=n DOCLEAN=n ## FIXME ## Lets keep compatibility with -M [ "$DOCUE" != "y" ] && DOCUE=n for ACTION in $(echo $ACTIONS | tr , \ ) do case $ACTION in default) DOCDDB=y; DOREAD=y; DOENCODE=y; DOTAG=y; DOMOVE=y; DOCLEAN=y;; cue) DOCUE=y ; MAKECUEFILE=y ;; cddb) DOCDDB=y;; read) DOREAD=y;; normalize) DONORMALIZE=y; DOREAD=y;; # preprocess) DOPREPROCESS=y; DOREAD=y;; encode) DOENCODE=y; DOREAD=y;; # postprocess) DOPREPROCESS=y; DOENCODE=y; DOREAD=y;; tag) DOTAG=y; DOREAD=y; DOENCODE=y; DOCDDB=y;; move) DOMOVE=y; DOTAG=y; DOREAD=y; DOENCODE=y; DOCDDB=y;; replaygain) DOCDDB=y; DOREAD=y; DOENCODE=y; DOTAG=y; DOMOVE=y; DOREPLAYGAIN=y;; playlist) DOCDDB=y; DOPLAYLIST=y;; clean) DOCLEAN=y;; esac done if [ "$DONORMALIZE" = "y" ] && [ "$DOREPLAYGAIN" = "y" ]; then # FIXME # should we abort on error or just inform the user? log warning "selected both normalize and replaygain actions" fi for SHOWCDDBFIELD in $(echo $SHOWCDDBFIELDS | tr , \ ); do case $SHOWCDDBFIELD in y*|Y*) SHOWCDDBYEAR="y";; g*|G*) SHOWCDDBGENRE="y";; *) ;; esac done # At this point a CDROM has to be defined, so we check it exists. if [ X"$CDROM" != "X" ] ; then if ( [ "$CDROMREADERSYNTAX" = "cdda2wav" ] || [ "$CDROMREADERSYNTAX" = "icedax" ] ) && [ "$NEEDCDROMID" = "y" ] ; then if [ "$OSFLAVOUR" = "FBSD" ]; then if echo "$CDROMID" | grep "^[0-9],[0-9],[0-9]$" >/dev/null 2>&1 ; then :; else log error "CDROMID not in the right format for $CDROMREADERSYNTAX" log error "Use \"cdrecord -scanbus\" to obtain a adecuate ID an set CDROMID accordingly" exit 1 fi fi elif [ ! -e "$CDROM" -a X"$DOREAD" = "Xy" ]; then log error "CDROM device cannot be found." exit 1 fi # avoid processing if we are not going to hit the CDROM. elif [ X"$DOREAD" = "Xy" ]; then log error "CDROM has not been defined or cannot be found" exit 1 fi # USEPIPES pre-tests, before we get into more serious stuff # Not compatible with: # - multiple outputs # - normalize # - lowdisk algorithm # - anything else? if [ X"$USEPIPES" = "Xy" ]; then if [ $(echo "$OUTPUTTYPE" | tr , \ | wc -w ) -gt 1 ]; then log error "Unix pipes not compatible with multiple outputs" exit 1 fi if [ X"$DONORMALIZE" = "Xy" ]; then log error "Unix pipes not compatible with normalizer" # FIXME # Do we need to exit or shall we just disable the mode? exit 1 fi if [ X"$BATCHNORM" = "Xy" ]; then log error "Unix pipes not compatible with BATCHNORM encoding" exit 1 fi if [ X"$NOGAP" = "Xy" ]; then log error "Unix pipes not compatible with NOGAP encoding" exit 1 fi if [ X"$DOENCODE" = "Xn" ]; then vecho warning "Disabling Unix pipes since we are not encoding!" USEPIPES=n fi if [ X"$LOWDISK" = "Xy" ]; then log error "Unix pipes not compatible with lowdisk algorithm" exit 1 fi fi # LOWDISK pre-tests, before we get into more problematic stuff # Not compatible with anything that needs all the files in the hard disc: # - BATCHNORM # - NOGAP lame mode if [ X"$LOWDISK" = "Xy" ]; then if [ X"$BATCHNORM" = "Xy" ]; then log error "Unix pipes not compatible with BATCHNORM encoding" exit 1 fi if [ X"$NOGAP" = "Xy" ]; then log error "Unix pipes not compatible with NOGAP encoding" exit 1 fi fi # BATCHNORM pre-tests, before we get into serious problems # Not compatible with if [ "$BATCHNORM" = "y" ] && [ "$DONORMALIZE" = "n" ]; then vecho warning "Disabling BATCHNORM since we are not normalizing!" BATCHNORM=n fi # Check the encoding format from the ones available in the system, if nothing has been configured. if [ X"$OUTPUTTYPE" = "X" ]; then for DEFAULT_OUTPUT in $( echo "$DEFAULT_OUTPUT_BINARIES" | tr , \ ); do DEFAULT_OUTPUT_FORMAT="$(echo $DEFAULT_OUTPUT | cut -d ":" -f 1)" DEFAULT_OUTPUT_BINARY="$(echo $DEFAULT_OUTPUT | cut -d ":" -f 2)" if [ -x $(which $DEFAULT_OUTPUT_BINARY) ] ; then OUTPUTTYPE=$DEFAULT_OUTPUT_FORMAT vecho "No default output type defined. Autoselecting $OUTPUTTYPE..." >&2 break fi done if [ X"$OUTPUTTYPE" = "X" ]; then log error "no encoder found in the PATH" log error "hints: are all dependencies installed? has the \$PATH been modified?" exit 1 fi fi # Decide which CDROM reader we're gonna use case "$CDROMREADERSYNTAX" in cdparanoia|debug) CDROMREADER="$CDPARANOIA" CDROMREADEROPTS="$CDPARANOIAOPTS" ;; cdda2wav | icedax) CDROMREADER="$CDDA2WAV" CDROMREADEROPTS="$CDDA2WAVOPTS" ;; dagrab) CDROMREADER="$DAGRAB" CDROMREADEROPTS="$DAGRABOPTS" ;; cddafs) CDROMREADER="$CDDAFS" CDROMREADEROPTS="$CDDAFSOPTS" ;; flac) CDROMREADER="$FLAC" CDROMREADEROPTS="$FLACOPTS" ;; esac # There's only one normalize... case "$NORMALIZERSYNTAX" in default|normalize) NORMALIZER="$NORMALIZE" NORMALIZEROPTS="$NORMALIZEOPTS" ;; esac # Allow -o OUTPUT(1):OPTIONS(1),...,OUTPUT(N):OPTIONS(N) mode of operation if echo "$OUTPUTTYPE" | grep ":" > /dev/null 2>&1 ; then for OUTPUT in "$(echo "$OUTPUTTYPE" | tr \ \|| tr , \ | tr \| \ )"; do case "$OUTPUT" in vorbis:*|ogg:*) OGGENCODEROPTSCLI="$( echo $OUTPUT | cut -d: -f2- )" ;; mp3:*) MP3ENCODEROPTSCLI="$( echo $OUTPUT | cut -d: -f2- )" ;; flac:*) FLACENCODEROPTSCLI="$( echo $OUTPUT | cut -d: -f2- )" ;; spx:*) SPEEXENCODEROPTSCLI="$( echo $OUTPUT | cut -d: -f2- )" ;; mpc:*) MPPENCODEROPTSCLI="$( echo $OUTPUT | cut -d: -f2- )" ;; m4a:*) AACENCODEROPTSCLI="$( echo $OUTPUT | cut -d: -f2- )" ;; esac done for OUTPUT in "$(echo "$OUTPUTTYPE" | tr , \ )"; do TEMPOUTPUT=$( echo "$OUTPUT" | cut -d: -f1 ) TEMPOUTPUTTYPE="${TEMPOUTPUTTYPE:+$TEMPOUTPUTTYPE,}$TEMPOUTPUT" done OUTPUTTYPE="$TEMPOUTPUTTYPE" fi # If nothing has been specified, use oggenc for oggs and lame for mp3s and flac # for flacs and speexenc for speex and mppenc for mpps and faac for m4as # Getting ready for multiple output changes for OUTPUT in $(echo $OUTPUTTYPE | tr , \ ) do case $OUTPUT in vorbis|ogg) [ "$OGGENCODERSYNTAX" = "default" ] && OGGENCODERSYNTAX=oggenc [ "$DOTAG" = "y" ] && NEEDCOMMENTER=y [ "$DOREPLAYGAIN" = "y" ] && NEEDVORBISGAIN=y OGGOUTPUTCONTAINER=ogg ;; mp3) [ "$MP3ENCODERSYNTAX" = "default" ] && MP3ENCODERSYNTAX=lame [ "$DOTAG" = "y" ] && NEEDTAGGER=y [ "$DOREPLAYGAIN" = "y" ] && NEEDMP3GAIN=y ;; flac) [ "$FLACENCODERSYNTAX" = "default" ] && FLACENCODERSYNTAX=flac [ "$DOTAG" = "y" ] && NEEDMETAFLAC=y [ "$DOREPLAYGAIN" = "y" ] && NEEDMETAFLAC=y [ "$ONETRACK" = "y" ] && [ "$DOCUE" = "y" ] && NEEDMETAFLAC=y ;; spx) [ "$SPEEXENCODERSYNTAX" = "default" ] && SPEEXENCODERSYNTAX=speexenc # [ "$DOREPLAYGAIN" = "y" ] && ;; mpc) [ "$MPPENCODERSYNTAX" = "default" ] && MPPENCODERSYNTAX=mppenc [ "$DOREPLAYGAIN" = "y" ] && NEEDMPPGAIN=y ;; m4a) [ "$AACENCODERSYNTAX" = "default" ] && AACENCODERSYNTAX=faac [ "$DOTAG" = "y" ] && CHECKATOMICPARSLEY=y ;; wav) if [ "$KEEPWAVS" = "y" ]; then vecho "Setting the KEEPWAVS option, since the resulting wav files were requested..." fi KEEPWAVS=move ;; *) log error "Invalid OUTPUTTYPE defined" exit 1 ;; esac done # decide which encoder case "$MP3ENCODERSYNTAX" in lame) MP3ENCODEROPTS="${MP3ENCODEROPTSCLI:-$LAMEOPTS}" MP3ENCODER="$LAME" ;; toolame) MP3ENCODEROPTS="${MP3ENCODEROPTSCLI:-$TOOLAMEOPTS}" MP3ENCODER="$TOOLAME" ;; gogo) MP3ENCODEROPTS="${MP3ENCODEROPTSCLI:-$GOGOOPTS}" MP3ENCODER="$GOGO" ;; bladeenc) MP3ENCODEROPTS="${MP3ENCODEROPTSCLI:-$BLADEENCOPTS}" MP3ENCODER="$BLADEENC" ;; l3enc) MP3ENCODEROPTS="${MP3ENCODEROPTSCLI:-$L3ENCOPTS}" MP3ENCODER="$L3ENC" ;; xingmp3enc) MP3ENCODEROPTS="${MP3ENCODEROPTSCLI:-$XINGMP3ENCOPTS}" MP3ENCODER="$XINGMP3ENC" ;; mp3enc) MP3ENCODEROPTS="${MP3ENCODEROPTSCLI:-$MP3ENCOPTS}" MP3ENCODER="$MP3ENC" ;; esac case "$OGGENCODERSYNTAX" in vorbize) OGGENCODEROPTS="${OGGENCODEROPTSCLI:-$VORBIZEOPTS}" OGGENCODER="$VORBIZE" ;; oggenc) OGGENCODEROPTS="${OGGENCODEROPTSCLI:-$OGGENCOPTS}" OGGENCODER="$OGGENC" ;; esac case "$FLACENCODERSYNTAX" in flac) FLACENCODEROPTS="${FLACENCODEROPTSCLI:-$FLACOPTS}" FLACENCODER="$FLAC" if [ "$DOREPLAYGAIN" = "y" ]; then FLACENCODEROPTS="${FLACENCODEROPTS} --replay-gain" fi # FLAC streams can be encapsulated on a Ogg transport layer if echo "$FLACENCODEROPTS" | grep -E -- "(^| )--ogg($| )" > /dev/null 2>&1 ;then log error "FLAC on an Ogg container is not yet supported" log error "due to problem with adding comments to such files" exit 1 FLACOUTPUTCONTAINER=ogg else FLACOUTPUTCONTAINER=flac fi ;; esac case "$SPEEXENCODERSYNTAX" in speexenc) SPEEXENCODEROPTS="${SPEEXENCODEROPTSCLI:-$SPEEXENCOPTS}" SPEEXENCODER="$SPEEXENC" ;; esac case "$MPPENCODERSYNTAX" in mppenc) MPPENCODEROPTS="${MPPENCODEROPTSCLI:-$MPPENCOPTS}" MPPENCODER="$MPPENC" ;; esac case "$AACENCODERSYNTAX" in faac) AACENCODEROPTS="${AACENCODEROPTSCLI:-$AACENCOPTS}" AACENCODER="$AACENC" ;; esac # and which tagger if [ "$ID3TAGV" = "1" ]; then TAGGER="$ID3" TAGGEROPTS="$ID3OPTS" else TAGGER="$ID3V2" TAGGEROPTS="$ID3V2OPTS" fi # NOGAP is specific to lame. Other encoders fail ... if [ "$NOGAP" = "y" ] && [ ! "$MP3ENCODER" = "lame" ]; then log warning "the NOGAP option is specific to lame. Deactivating..." NOGAP=n fi # Options for mkcue case "$CUEREADERSYNTAX" in default|mkcue) CUEREADEROPTS="${CDROM}" CUEREADER="$MKCUE" ;; esac # which information retrieval tool are we using? case "$CDDBTOOL" in cddb) ;; musicbrainz) ;; esac # Check if both OGGEOUTPUTCONTAINER and FLACOUTPUTCONTAINER are the same, and differentiante them if [ X"$OGGOUTPUTCONTAINER" = "Xogg" ] && [ X"$FLACOUTPUTCONTAINER" = "Xogg" ]; then log error "FLAC on an Ogg container is not yet supported" log error "due to problem with adding comments to such files" exit 1 OGGOUTPUTCONTAINER=ogg.ogg FLACOUTPUTCONTAINER=flac.ogg vecho warning "modified file endings due to conflicting transport layers in Ogg/Vorbis and Ogg/FLAC" fi # Clean up nice options (either use '-n NICELEVEL or -NICELEVEL') if [ "$ENCNICE" ]; then ENCNICE="-n $ENCNICE" fi if [ "$READNICE" ]; then READNICE="-n $READNICE" fi if [ "$DISTMP3NICE" ]; then DISTMP3NICE="-n $DISTMP3NICE" fi # Don't check for stuff if it's not needed if [ "$REMOTEHOSTS" ]; then NEEDDISTMP3=y fi if [ "$DONORMALIZE" = "y" ]; then NEEDNORMALIZER=y fi if [ "$EJECTCD" = "y" ]; then NEEDEJECT=y fi if [ ! "$CDDBAVAIL" = "n" ] && [ "$DOCDDB" = "y" ]; then if [ "$CDDBMETHOD" = "cddb" ]; then NEEDHTTPGET=y elif [ "$CDDBMETHOD" = "musicbrainz" ]; then : fi fi if [ "$DOCUE" = "y" ]; then NEEDCUEREADER=y fi if [ X"$CDSPEEDVALUE" != "X" ] && [ "$DOREAD" = "y" ]; then case "$CDROMREADERSYNTAX" in cdparanoia|debug) CDROMREADEROPTS="$CDPARANOIAOPTS -S $CDSPEEDVALUE" ;; ### FIXME ### translate "cue2discid" from python to bash flac) NEEDMETAFLAC=y ; NEEDCUE2DISCID=y ; CDSPEEDVALUE="" ;; *) NEEDCDSPEED=y ;; esac fi ###USEPIPESSUPPORT### # Rippers with USEPIPE support # FIXME # Include here all the rippers we can figure out support pipes PIPERIPPER_cdparanoia="-" PIPERIPPER_debug="-" PIPERIPPER_flac="-c " # Encoders with USEPIPE support # FIXME # Include here all the encoders we can figure out support pipes PIPE_lame="-" PIPE_bladeenc="-" PIPE_oggenc="-" PIPE_flac="-" # Figure out if we can use pipes with the ripper/encoder combination # exit otherwise if [ "$USEPIPES" = "y" ]; then PIPERIPPERSVARCHECK="PIPERIPPER_${CDROMREADERSYNTAX}" case "$OUTPUT" in mp3) PIPEENCODERSVARCHECK="PIPE_$MP3ENCODERSYNTAX" ;; vorbis|ogg) PIPEENCODERSVARCHECK="PIPE_$OGGENCODERSYNTAX" ;; flac) PIPEENCODERSVARCHECK="PIPE_$FLACENCODERSYNTAX" ;; spx) PIPEENCODERSVARCHECK="PIPE_$SPEEXENCODER" ;; mpc) PIPEENCODERSVARCHECK="PIPE_$MPPENCODER" ;; esac decho "PIPERIPPERSVARCHECK: $( eval echo "\$$PIPERIPPERSVARCHECK" )" if [ "$( eval echo "\$$PIPERIPPERSVARCHECK" )" = "$" ] || \ [ "$( eval echo "\$$PIPERIPPERSVARCHECK" )" = "" ] ; then log error "no support for pipes with given ripper" log error "read the USEPIPES file from the source tarball to get help." log error "On a Debian system, it is under /usr/share/doc/abcde/USEPIPES.gz" exit 1; fi decho "PIPEENCODERSVARCHECK: $( eval echo "\$$PIPEENCODERSVARCHECK" )" if [ "$( eval echo "\$$PIPEENCODERSVARCHECK" )" = "$" ] || \ [ "$( eval echo "\$$PIPEENCODERSVARCHECK" )" = "" ] ; then log error "no support for pipes with given encoder" log error "read the USEPIPES file from the source tarball to help" log error "on a Debian system, read /usr/share/doc/abcde/USEPIPES.gz" exit 1; fi fi # Make sure a buncha things exist for X in $CDROMREADER $CDDISCID ${NEEDTAGGER+$TAGGER} $MP3ENCODER \ $OGGENCODER $FLACENCODER $SPEEXENCODER $MPPENCODER \ $AACENCODER \ ${NEEDATOMICPARSLEY+$ATOMICPARSLEY} \ ${NEEDHTTPGET+$HTTPGET} ${NEEDDISTMP3+$DISTMP3} \ ${NEEDCOMMENTER+$VORBISCOMMENT} ${NEEDMETAFLAC+$METAFLAC} \ ${NEEDNORMALIZER+$NORMALIZER} ${NEEDEJECT+$EJECT} \ ${NEEDDISKTOOL+disktool} ${NEEDCDSPEED+$CDSPEED} \ ${NEEDVORBISGAIN+$VORBISGAIN} ${NEEDMP3GAIN+$MP3GAIN} \ ${NEEDMPPGAIN+$MPPGAIN} ${NEEDCUEREADER+$CUEREADER} \ ${NEEDCUE2DISCID+$CUE2DISCID} do checkexec "$X" done if [ "$CHECKATOMICPARSLEY" = "y" ]; then if ! new_checkexec "$ATOMICPARSLEY"; then echo "WARNING: $ATOMICPARSLEY Not Found Not Post-Tagging!" DOTAG='n' else echo "Using $ATOMICPARSLEY To Tag AAC Tracks." fi fi # And last but not least, check if we can diff between files. We do not abort, # since diffing is not critical... if [ -x $(which $DIFF) ]; then :; else vecho warning "Disabling diff since we cannot find it in the \$PATH..." DIFF="" fi ## Now that we have metaflac, check if we need cue2discid #case $CDROMREADERSYNTAX in # flac) # TRACKINFO=$($METAFLAC --show-tag=CDDB $CDROM | cut -d"=" -f2 | grep -E "[a-f0-9]{8}") # if [ "$TRACKINFO" = "" ]; then # checkexec ${NEEDCUE2DISCID+$CUE2DISCID} # fi # ;; #esac CDROMREADER="$CDROMREADER $CDROMREADEROPTS" CDDBTOOL="$CDDBTOOL $CDDBTOOLOPTS" HTTPGET="$HTTPGET $HTTPGETOPTS" # Here it used to say: # One thousand lines in, we can start doing stuff with things # Well, right now we are at line 3737 ;) # Export needed things so they can be read in this subshell export CDDBTOOL ABCDETEMPDIR TRACKQUEUE LOWDISK EJECTCD EJECT EJECTOPTS export CDROM CDDBDATA REMOTEHOSTS MAXPROCS HTTPGET MD5SUM if [ "$DOREAD" = "y" ]; then # User-definable function to set some things. Use it for # - closing the CD tray with eject -t # - set the CD speed value with eject -x vecho -n "Executing customizable pre-read function... " pre_read # Execute the user-defined pre-read funtion. Close the CD with it. vecho "done." fi # Get ABCDETEMPDIR created and status file initialized do_discid if [ "$DOCDDB" = "y" ]; then # start with a sane default: CDDBLOCALSTATUS=notfound if [ $CDDBUSELOCAL = "y" ]; then do_localcddb fi if checkstatus cddb-choice > /dev/null; then : else if [ "$CDDBLOCALSTATUS" = "notfound" ] ; then case "$CDDBMETHOD" in cddb) do_cddbstat do_cddbquery do_cddbread ;; musicbrainz) do_musicbrainz ;; esac fi fi do_cddbedit eval "$($CDDBTOOL parse "$CDDBDATA")" fi # Before reading tracks, we set the speed of the device if [ X"$CDSPEEDVALUE" != "X" ]; then case "$CDROMREADERSYNTAX" in cdparanoia|debug) ;; flac) ;; *) do_cdspeed ;; esac fi # Define the first and last track, since we might need them later in several places FIRSTTRACK=$( get_first $TRACKQUEUE ) LASTTRACK=$( get_last $TRACKQUEUE ) if [ -f "$ABCDETEMPDIR/status" ] && [ X"$ERASEENCODEDSTATUS" = "Xy" ]; then mv "$ABCDETEMPDIR/status" "$ABCDETEMPDIR/status.old" grep -v ^encodetracklocation- < "$ABCDETEMPDIR/status.old" \ | grep -v ^encode-output > "$ABCDETEMPDIR/status" fi if checkstatus onetrack ; then ONETRACK=y ; fi if [ "$ONETRACK" = "y" ]; then # Reuse the CUEFILE in case we created it (now or in a previous run) if CUEFILE=$(checkstatus cuefile); then IMPORTCUESHEET=y fi fi # Create playlist if needed (backgroundable) and start reading in tracks ( if [ ! "$ONETRACK" = "y" ]; then if [ "$DOPLAYLIST" = "y" ]; then echo Creating playlist... >&2 do_playlist fi fi # For the lowdisk option, only one program is running at once so the encoder # can be unsilenced right away. if [ "$LOWDISK" = "y" ] || [ "$ONETRACK" = "y" ]; then echo "encode-output=loud" >> "$ABCDETEMPDIR/status" fi if [ "$ONETRACK" = "y" ]; then TRACKS="$FIRSTTRACK" if [ "$USEPIPES" = "y" ]; then if checkstatus readencodetrack-$FIRSTTRACK; then :; else do_cdread onetrack $FIRSTTRACK $LASTTRACK | do_encode $FIRSTTRACK %local0% > /dev/null 2>&1 fi else if checkstatus readtrack-$FIRSTTRACK; then :; else do_cdread onetrack $FIRSTTRACK $LASTTRACK fi fi else for UTRACKNUM in $TRACKQUEUE do if [ "$DOREAD" = "y" ]; then if [ "$USEPIPES" = "y" ]; then if checkstatus readencodetrack-$UTRACKNUM; then :; else # Read, pipe, shut up! do_cdread $UTRACKNUM | do_encode $UTRACKNUM %local0% > /dev/null 2>&1 fi else if checkstatus readtrack-$UTRACKNUM; then :; else do_cdread $UTRACKNUM fi if [ "$?" != "0" ]; then # CD read failed - don't give the goahead to # the encoder echo NO exit fi fi fi if [ "$NOGAP" = "y" ] || [ "$BATCHNORM" = "y" ]; then : else # If we are not reading, set the encode output to loud already, so # that we can see the output of the first track. if [ "$MAXPROCS" = "1" ] && [ ! "$DOREAD" = "y" ]; then echo "encode-output=loud" >> "$ABCDETEMPDIR/status" fi echo NEXTTRACK # Get the encoder machine churning again if [ "$DOREAD" = "y" ]; then if [ "$LOWDISK" = "y" ] && [ "$DOENCODE" = "y" ]; then until checkstatus encodetrack-$UTRACKNUM do if checkerrors encodetrack-$UTRACKNUM; then break fi sleep 2 done fi fi fi done fi # Now that we're done the encoding can be loud again - # if we're not using SMP. if [ "$MAXPROCS" = "1" ]; then echo "encode-output=loud" >> "$ABCDETEMPDIR/status" fi # All tracks read, start encoding. if [ "$NOGAP" = "y" ] || [ "$BATCHNORM" = "y" ] || [ "$ONETRACK" = "y" ]; then echo NEXTTRACK fi # Execute the user-defined post_read funtion before ejecting CD post_read # We are now finished with the cdrom - it can be safely ejected. Note that # abcde will not have completed yet. if [ "$EJECTCD" = "y" ] && [ -x $(which $EJECT) ]; then # We check if the disk we are processing is actually the disk inside the # CD tray. If not, we do not eject the CD, since it might be so that the # user ejected it manually. #CURRENTTRACKINFO=$($CDDISCID $CDROM) #if if [ "$?" != "1" ] && [ "$CURRENTTRACKINFO" = "$TRACKINFO" ] ; then # More FreeBSD bits. if [ X"$(uname)" = X"FreeBSD" ] ; then # FreeBSD eject uses the EJECT environment variable to name the CDROM # but in this script EJECT is in the envionment and names the program eject=$EJECT unset EJECT # The FreeBSD eject needs "adc0" not "/dev/adc0c" cd="$(echo $CDROM | sed -e 's=.*/==;s=[a-h]$==;')" $eject $EJECTOPTS $cd elif [ X"$(uname)" = X"Darwin" ] ; then disktool -e ${CDROM#/dev/} 0 else $EJECT $EJECTOPTS "$CDROM" fi #fi fi ) | ( ## Do we need to pre-process #if [ x"$PREPROCESS" = "x" ] ; then # cat #else # for PRETRACKNUM in $TRACKQUEUE # do # read GOAHEAD # if [ "$GOAHEAD" = "NO" ]; then break; fi # PREPROCEED= # until [ $PREPROCEED ] # do # if checkstatus readtrack-$PRETRACKNUM; then PREPROCEED=y; break; fi # # all locations are working, wait and try again later # if [ ! $PREPROCEED ]; then sleep 3; fi # done # ( do_preprocess $PRETRACKNUM # echo "NEXTTRACK" # ) & # done #fi # #) | ( # In BATCHNORM and/or NOGAP modes, we want all tracks to be read first. #BACK if [ "$BATCHNORM" = "y" ] || [ "$NOGAP" = "y" ]; then read GOAHEAD # For blocking - will contain either "NO" or "NEXTTRACK" if [ "$GOAHEAD" = "NO" ]; then break; fi for LASTTRACK in $TRACKQUEUE; do :; done if checkstatus readtrack-$LASTTRACK; then if [ "$DONORMALIZE" = "y" ] && [ "$BATCHNORM" = "y" ]; then if checkstatus normalizetrack-$LASTTRACK; then :; else do_batch_normalize; fi if checkerrors batch-normalize; then exit 1; fi fi if [ "$DOENCODE" = "y" ] && [ "$NOGAP" = "y" ]; then if [ "$DONORMALIZE" = "y" ]; then for UTRACKNUM in $TRACKQUEUE do if checkstatus readtrack-$UTRACKNUM; then if checkstatus normalizetrack-$UTRACKNUM; then :; else do_normalize $UTRACKNUM; fi fi done fi if checkstatus encodetrack-$LASTTRACK; then :; else do_nogap_encode; fi if checkerrors nogap-encode; then exit 1; fi fi fi fi # If we are using ONETRACK, we can proceed with the normal encoding using just the $FIRSTTRACK as TRACKQUEUE if [ "$ONETRACK" = "y" ] ; then TRACKQUEUE="$FIRSTTRACK" TRACKS="$FIRSTTRACK" fi # Do the encoding, including parallelization of remote encoding # Figure out where each track is going to be encoded ENCODELOCATIONS="$(echo $REMOTEHOSTS | tr , ' ')" if [ "$MAXPROCS" != "0" ]; then for NUM in $(f_seq_row 1 "$MAXPROCS") do ENCODELOCATIONS="$ENCODELOCATIONS %local$NUM%" done fi # Strip whitespace ENCODELOCATIONS=$(echo $ENCODELOCATIONS) for UTRACKNUM in $TRACKQUEUE do # Wait for our cue read GOAHEAD # For blocking - will contain either "NO" or "NEXTTRACK" if [ "$GOAHEAD" = "NO" ]; then break; fi # find out where this track is to be encoded if [ "$DOENCODE" = "y" -a "$USEPIPES" != "y" ]; then # Make sure we have a place to encode this, if not, exit stage right if [ -z "$ENCODELOCATIONS" ]; then continue fi PROCEED= until [ $PROCEED ] do for LOCATION in $ENCODELOCATIONS do PREVIOUSTRACK="$(checkstatus encodetracklocation-$LOCATION)" # check first if a track has ever been assigned to this location if [ -z "$PREVIOUSTRACK" ]; then PROCEED=y; break; fi # If it errored out, rebuild $ENCODELOCATIONS without this location in it if checkerrors encodetrack-$PREVIOUSTRACK; then for TEMPLOCATION in $ENCODELOCATIONS do if [ "$TEMPLOCATION" != "$LOCATION" ]; then TEMPENCODELOCATIONS="$TEMPENCODELOCATIONS $TEMPLOCATION" fi done ENCODELOCATIONS=$(echo $TEMPENCODELOCATIONS) ABORT=y PROCEED=y break fi # We're still here, this location must have been previously assigned, # and last completed without error - check if it's done with the # previous track yet if checkstatus encodetrack-$PREVIOUSTRACK; then PROCEED=y; break; fi done # all locations are working, wait and try again later if [ ! $PROCEED ]; then sleep 3; fi done # Record the location we're about to encode the next track at echo "encodetracklocation-$LOCATION=$UTRACKNUM" >> "$ABCDETEMPDIR/status" fi # Don't proceed with the rest of the loop if we can't encode if [ "$ABORT" ]; then continue; fi ## FIXME ## Add here ## run_command tagtrack-$OUTPUT-$1 $METAFLAC $METAFLACOPTS ${IMPORTCUESHEET:+--import-cuesheet-from="$ABCDETEMPDIR/$CUEFILE"} --import-tags-from=- "$ABCDETEMPDIR/track$1.$FLACOUTPUTCONTAINER" # Set TRACKNUM, TRACKNAME if [ -e "$CDDBDATA" ]; then if [ "$ONETRACK" = "y" ]; then TRACKNAME="$DALBUM" TRACKNUM="$FIRSTTRACK" splitvarious else TRACKNUM=$UTRACKNUM CDDBTRACKNUM=$(expr $UTRACKNUM - 1) getcddbinfo TRACKNAME splitvarious fi fi # You can't encode a file which needs to be normalized before finishing # You can't tag a file before it's finished encoding - # thus all of this is backgrounded together ( if [ "$DONORMALIZE" = "y" ]; then if checkstatus readtrack-$UTRACKNUM; then if checkstatus normalizetrack-$UTRACKNUM; then :; else do_normalize $UTRACKNUM; fi fi fi if [ "$DOENCODE" = "y" -a "$USEPIPES" != "y" ]; then if checkstatus readtrack-$UTRACKNUM; then #if checkstatus encodetrack-$UTRACKNUM; then :; else do_encode $UTRACKNUM $LOCATION; fi if [ "$DONORMALIZE" = "y" ]; then if checkstatus normalizetrack-$UTRACKNUM; then if checkstatus encodetrack-$UTRACKNUM; then :; else do_encode $UTRACKNUM $LOCATION $OUTPUT; fi fi else if checkstatus encodetrack-$UTRACKNUM; then :; else do_encode $UTRACKNUM $LOCATION $OUTPUT; fi fi fi fi if [ "$DOTAG" = "y" ]; then if checkstatus encodetrack-$UTRACKNUM; then if checkstatus tagtrack-$UTRACKNUM; then :; else do_tag $UTRACKNUM; fi fi # Lets tag the cue file if checkstatus cleancuefile >/dev/null; then :; else if checkstatus cuefile >/dev/null ; then do_cleancue fi fi fi if [ "$DOMOVE" = "y" ]; then if checkstatus tagtrack-$UTRACKNUM; then if checkstatus movetrack-$UTRACKNUM; then :; else do_move $UTRACKNUM; fi fi fi ) & done # Go through it again and make sure there's no distmp3 stragglers, otherwise # we'll delete the files they're working on # Do NOT play ball if there is no ball to play (so ABORT if needed) ## FIXME ## Check also for files which are encoded using PIPEs. if [ "$DOENCODE" = "y" ] && [ "$USEPIPES" != "y" ] && [ ! "$ABORT" ]; then PROCEED= until [ $PROCEED ] do PROCEED=y for LOCATION in $ENCODELOCATIONS do CHECKTRACK="$(checkstatus encodetracklocation-$LOCATION)" # "How can he give us a status update, if he's DEAD?" if checkstatus encodetrack-$CHECKTRACK; then continue fi # Nothing to see here please go quietly back to your homes if [ -z "$CHECKTRACK" ]; then continue; fi # You're still here? Maybe there is something... if checkstatus encodetrack-$CHECKTRACK; then :; else PROCEED= ; break; fi done # hold up if [ ! $PROCEED ]; then sleep 5; fi done fi # If the above didn't catch the stragglers, this will wait if [ "$DOREPLAYGAIN" = "y" ]; then do_replaygain fi # Execute the user-defined post_encode function before cleaning up post_encode # Check to see if run_command logged any errors if [ -f "$ABCDETEMPDIR/errors" ]; then log error "The following commands failed to run:" cat "$ABCDETEMPDIR/errors" # Don't clean up DOCLEAN=n fi if [ "$KEEPWAVS" = "y" ];then # Don't clean up DOCLEAN=n fi # Check if we have moved all the formats we had previously encoded, if we are not using the FORCE. if [ "$DOCLEAN" = "y" ] && [ ! "$FORCE" = "y" ]; then ENCODED_FORMATS=$(grep -E "^encodetrack-(.{3,6})-(.{1,2})$" "$ABCDETEMPDIR/status" | cut -d"-" -f2 | sort -u | tr '\n' '|') MOVED_FORMATS=$(grep -E "^movetrack-output-(.{3,6})$" "$ABCDETEMPDIR/status" | cut -d"-" -f3 | sort -u | tr '\n' '|') if [ "$ENCODED_FORMATS" != "$MOVED_FORMATS" ]; then log warning "The encoded formats does not match with the moved ones" log warning "Formats encoded: $( echo $ENCODED_FORMATS | tr "|" " " )" log warning "Formats moved: $( echo $MOVED_FORMATS | tr "|" " " )" log warning "Use \"abcde -a clean -f -C $DISCID\" to force the removal of the remaining data." DOCLEAN=n fi fi if [ "$DOCLEAN" = "y" ]; then # Wipe all the evidence # Gimme gimme gimme some more time! sleep 5 rm -rf "$ABCDETEMPDIR" echo "Finished." else echo "Finished. Not cleaning $ABCDETEMPDIR." fi ) exit 0 # b:is_bash # vim:tabstop=4