Locale fixes, and optimization, for CD-TEXT lookup

Patch 2 of 2 to address the issue where abcde fails with accented
characters from CD-TEXT.

This addresses issue 52:
https://abcde.einval.com/bugzilla/show_bug.cgi?id=53

Thanks again to Matthias Andree <matthias.andree@gmx.de> for his
contributions to abcde!

diff --git a/abcde b/abcde
index dcadbea..774580c 100755 (executable)
--- a/abcde
+++ b/abcde
@@ -2478,7 +2478,10 @@ do_cdtext ()
        fi
 
        # Do we have CD-Text on the disc (and can the drive read it?)
-       ${CDTEXT_READER} -J -N -D ${CDDA2WAVCDROM} > "$ABCDETEMPDIR/cd-text" 2>&1
+       (
+               cd "$ABCDETEMPDIR" && rm -f audio.* audio_*
+               ${CDTEXT_READER} -J -v titles -D ${CDDA2WAVCDROM} > "$ABCDETEMPDIR/cd-text" 2>&1
+       )
        grep -a -q '^CD-Text: detected' "$ABCDETEMPDIR/cd-text"
        ERRORCODE=$?
        if [ $ERRORCODE -ne 0 ]; then
@@ -2494,21 +2497,25 @@ do_cdtext ()
        echo "done." >> "$ABCDETEMPDIR/cddbchoices"
        echo cddb-read-1-complete >> "$ABCDETEMPDIR/status"
        echo cddb-choice=1 >> "$ABCDETEMPDIR/status"
-       ATITLE=$(grep -ae '^Album title:' "${ABCDETEMPDIR}/cd-text" | cut -c14- )
+       #ATITLE=$(grep -ae '^Album title:' "${ABCDETEMPDIR}/cd-text" | cut -c14- | sed "s~'\(.*\)'.*\[from \(.*\)]$~\2 / \1~")
+       ATITLE=$(grep -ae '^DTITLE=' "${ABCDETEMPDIR}/audio.cddb" | 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 $TRACKS)
-       do
-               TRACKM1=$(($TRACK - 1))
-               TITLE="$(grep -aE ^Track\ +$TRACK: "$ABCDETEMPDIR/cd-text" | tr -d \\r\\n | sed 's~^Track ..: .~~g;'"s~'$~~g")"
-               echo "$TRACK: $TITLE" >> "$ABCDETEMPDIR/cddbchoices"
-               sed "s~^TTITLE${TRACKM1}=.*~TTITLE${TRACKM1}=${TITLE}~" "$ABCDETEMPDIR/cddbread.1" > "$ABCDETEMPDIR/cddbread.new" && \
-                       mv -f "$ABCDETEMPDIR/cddbread.new" "$ABCDETEMPDIR/cddbread.1"
-       done
-       sed "s~^DTITLE=.*~DTITLE=${ATITLE}~" "$ABCDETEMPDIR/cddbread.1" > "$ABCDETEMPDIR/cddbread.new" && \
-               mv -f "$ABCDETEMPDIR/cddbread.new" "$ABCDETEMPDIR/cddbread.1"
-       rm -f "$ABCDETEMPDIR/cddbread.new" # In case a sed error occurred
+       echo "---- ${ATITLE} ----" >> "$ABCDETEMPDIR/cddbchoices"
+       sed -n 's~^Track \(..:\) .\(.*\).$~\1 \2~gp;' "$ABCDETEMPDIR/cd-text" >>"$ABCDETEMPDIR/cddbchoices"
+       rm -f "$ABCDETEMPDIR/cddbread.1"
+       # XXX FIXME - this is a hack and should be replaced by proper
+       # character set tracking for the CDDB data we have.
+       if test "$CDDBPROTO" -ge 6 ; then
+               # convert to Unicode
+               iconv -f iso-8859-1 -t utf-8 <"$ABCDETEMPDIR/audio.cddb" >"$ABCDETEMPDIR/cddbread.1"
+       else
+               # copy verbatim, assuming CD-TEXT is in ISO-8859-1 format
+               # apparently icedax/cdda2wav have no support for 16-bit
+               # characters yet, either
+               cp -p "$ABCDETEMPDIR/audio.cddb" "$ABCDETEMPDIR/cddbread.1"
+       fi
+       ( cd "$ABCDETEMPDIR" && rm -f audio_* audio.* )
        echo >> "$ABCDETEMPDIR/cddbchoices"
        echo "cdtext-readcomplete" >> "$ABCDETEMPDIR/status"
 }

diff --git a/changelog b/changelog
index cf99d9f..8016120 100644 (file)
--- a/changelog
+++ b/changelog
@@ -3,6 +3,12 @@ abcde 2.8.2
  * Add some documentation for the aged CD ripper dagrab. Thanks to 
    Teika Kazura for the notification and suggested documentation. This 
    closes Issue 50: https://abcde.einval.com/bugzilla/show_bug.cgi?id=50
+ * Work by Matthias Andree <matthias.andree@gmx.de> to address the issue
+   where abcde fails with accented characters from CD-TEXT. The issue and
+   partial fix applied here documented in Issue 53:
+   https://abcde.einval.com/bugzilla/show_bug.cgi?id=53
+   and also in the abcde mailing list:
+   https://lists.einval.com/pipermail/abcde-users/2017-January/000232.html
    
 abcde 2.8.1.