Analyze SubStationAlpha subtitles (ASS) within a MKV file to extract the used font and add them back to the MKV file as attachments

mkvattachfonts.sh

#!/usr/bin/env bash

set -e

FONTSDIR="/fonts/dir/change/me"

attachment_exists() {
    local mkv_file="$1"
    local attachment_name="$2"
    mkvmerge -i "$mkv_file" | grep -q -E "^Attachment ID [0-9]+: .+ file name '$attachment_name'$"
}

tmpdirsaferemove() {
    test -z $1 && return 1
    rm $1/*.ass
    test -f $1/fonts && rm $1/fonts
    rmdir $1
}

IFS=$'\n'
for file in $(ls "$1"); do
    # Skip non mkv file
    if ! [ -f "$1/$file" ]; then
        echo "Skipping non regular file: $file"
        continue
    fi
    clean=$(echo $file | sed -r 's/^.+\.mkv$//')
    if [ -n "$clean" ]; then
        echo "Skipping non mkv file: $file"
        continue
    fi
    # Process
    echo "Processing '$file'"
    # Get ASS subtitles
    SubStationSubsTracksIDs=$(mkvmerge -i "$1/$file" | grep -E '^Track ID [0-9]+: subtitles \(SubStationAlpha\)'  | sed -r 's/^Track ID ([0-9]+):.+$/\1/')
    nbSubs=$(echo "$SubStationSubsTracksIDs" | wc -l)
    echo -e "\tFound $(echo "$SubStationSubsTracksIDs" | wc -l) ASS subtitle(s)."
    if [ $nbSubs -eq 0 ]; then
        continue
    fi
    # Extract them
    tmpdir=$(mktemp -d)
    cmd="mkvextract tracks \"${1}/${file}\""
    for subID in $SubStationSubsTracksIDs; do
        cmd+=" ${subID}:${tmpdir}/sub${subID}.ass"
    done
    eval $cmd
    # Inspect them and extract needed fonts
    fontsfile="$tmpdir/fonts"
    for sub in $(ls "$tmpdir"); do
        echo -e "\tAnalyzing $sub"
        for font in $(grep --no-filename -E '^Style:' "$tmpdir/$sub" | sed -r 's/^Style: [^,]+,([^,]+),.+$/\1/' | sort | uniq); do
            echo "$font" >> $fontsfile
        done
    done
    fonts=$(cat $fontsfile | sort | uniq)
    nbFonts=$(echo -e "$fonts" | wc -l)
    if [ $nbFonts -eq 0 ]; then
        echo "No fonts needed: skipping"
        continue
    fi
    echo -e -n "\tNeed $nbFonts font(s):"
    for font in $fonts; do
        echo -n "  $font"
    done
    echo
    # Add fonts to file if not already present
    cmd="mkvpropedit \"$1/$file\""
    proceed=0
    for font in $fonts; do
        if [ "$font" == "Arial" ]; then
            fontfile='arial.ttf'
            if ! attachment_exists "$1/$file" "$font"; then
                echo -e "\tFont '$font' is needed and is not attached: schedulling addition"
                cmd+=" --attachment-name '$font' --attachment-description 'The $font font as a TrueType font' --add-attachment '${FONTSDIR}/$fontfile'"
                proceed=1
            else
                echo -e "\tFont '$font' is already attached."
            fi
        elif [ "$font" == "Comic Sans MS" ]; then
            fontfile='comic.ttf'
            if ! attachment_exists "$1/$file" "$font"; then
                echo -e "\tFont '$font' is needed and is not attached: schedulling addition"
                cmd+=" --attachment-name '$font' --attachment-description 'The $font font as a TrueType font' --add-attachment '${FONTSDIR}/$fontfile'"
                proceed=1
            else
                echo -e "\tFont '$font' is already attached."
            fi
        elif [ "$font" == "Times New Roman" ]; then
            fontfile='times.ttf'
            if ! attachment_exists "$1/$file" "$font"; then
                echo -e "\tFont '$font' is needed and is not attached: schedulling addition"
                cmd+=" --attachment-name '$font' --attachment-description 'The $font font as a TrueType font' --add-attachment '${FONTSDIR}/$fontfile'"
                proceed=1
            else
                echo -e "\tFont '$font' is already attached."
            fi
        elif [ "$font" == "Trebuchet MS" ]; then
            fontfile='trebuc.ttf'
            if ! attachment_exists "$1/$file" "$font"; then
                echo -e "\tFont '$font' is needed and is not attached: schedulling addition"
                cmd+=" --attachment-name '$font' --attachment-description 'The $font font as a TrueType font' --add-attachment '${FONTSDIR}/$fontfile'"
                proceed=1
            else
                echo -e "\tFont '$font' is already attached."
            fi
        elif [ "$font" == "Verdana" ]; then
            fontfile='verdana.ttf'
            if ! attachment_exists "$1/$file" "$font"; then
                echo -e "\tFont '$font' is needed and is not attached: schedulling addition"
                cmd+=" --attachment-name '$font' --attachment-description 'The $font font as a TrueType font' --add-attachment '${FONTSDIR}/$fontfile'"
                proceed=1
            else
                echo -e "\tFont '$font' is already attached."
            fi
        elif [ "$font" == "Courier New" ]; then
            fontfile='cour.ttf'
            if ! attachment_exists "$1/$file" "$font"; then
                echo -e "\tFont '$font' is needed and is not attached: schedulling addition"
                cmd+=" --attachment-name '$font' --attachment-description 'The $font font as a TrueType font' --add-attachment '${FONTSDIR}/$fontfile'"
                proceed=1
            else
                echo -e "\tFont '$font' is already attached."
            fi
        elif [ "$font" == "Rubik" ]; then
            fontfile='rubik.ttf'
            if ! attachment_exists "$1/$file" "$font"; then
                echo -e "\tFont '$font' is needed and is not attached: schedulling addition"
                cmd+=" --attachment-name '$font' --attachment-description 'The $font font as a TrueType font' --add-attachment '${FONTSDIR}/$fontfile'"
                proceed=1
            else
                echo -e "\tFont '$font' is already attached."
            fi
        elif [ "$font" == "Georgia" ]; then
            fontfile='georgia.ttf'
            if ! attachment_exists "$1/$file" "$font"; then
                echo -e "\tFont '$font' is needed and is not attached: schedulling addition"
                cmd+=" --attachment-name '$font' --attachment-description 'The $font font as a TrueType font' --add-attachment '${FONTSDIR}/$fontfile'"
                proceed=1
            else
                echo -e "\tFont '$font' is already attached."
            fi
        else
            tmpdirsaferemove $tmpdir
            echo
            echo -e "\tFont '$font' is not supported. Add the font to the fonts folder and update the script."
            exit 1
        fi
    done
    if [ $proceed -eq 1 ]; then
        eval $cmd
    else
        echo -e "\tNothing to add"
    fi
    # Done, remove tmp dir
    tmpdirsaferemove $tmpdir
    echo
done