makepdf

#!/bin/bash
# makepdf, makes a pdf from an image series
VERSION="1.0"
SCRIPTDIR="$(dirname "$(which "${0}")")"
. "${SCRIPTDIR}/mmfunctions" || { echo "Missing '${SCRIPTDIR}/mmfunctions'. Exiting." ; exit 1 ;};
DEPENDENCIES=(ffmpeg pdfjoin tesseract)
_initialize_make

_usage(){
    echo
    echo "$(basename "${0}") ${VERSION}"
    echo "This application will create a small pdf file (suitable for sharing online) from a collection of image file or package input with the following options."
    echo "Dependencies: ${DEPENDENCIES[@]}"
    echo "Usage: $(basename "${0}") [ -o /path/to/deliver/to/ ] fileorpackage1 [ fileorpackage2 ...]"
    echo "  -o directory ( directory to write the resulting file to )"
    echo "  -e emailaddress ( send an email about the delivery, only valid if -d is used )"
    echo "  -E emailaddress ( send an email about process outcome )"
    echo "  -h ( display this help )"
    echo
    exit
}
[ "${#}" = 0 ] && _usage

# command-line options to set mediaid and original variables
OPTIND=1
while getopts ":o:e:E:h" OPT ; do
    case "${OPT}" in
        o) OUTPUTDIR_FORCED="${OPTARG}" && _check_outputdir_forced ;;
        e) EMAILADDRESS_DELIVERY="${OPTARG}" ;;
        E) EMAILADDRESS_OUTCOME="${OPTARG}" ;;
        h) _usage ;;
        :) echo "Option -${OPTARG} requires an argument" ; _writeerrorlog "makepdf" "The option selected required an argument and none was provided. The script had to exit." ; exit 1 ;;
        *) echo "bad option -${OPTARG}" ; _usage ;;
    esac
done
shift $(( ${OPTIND} - 1 ))

[ "${EMAILADDRESS_OUTCOME}" ] && check_emailaddress "${EMAILADDRESS_OUTCOME}"
[ "${EMAILADDRESS_DELIVERY}" ] &&  check_emailaddress "${EMAILADDRESS_DELIVERY}"

while [ "${*}" != "" ] ; do
    # get context about the input
    INPUT="${1}"
    shift
    if [ -z "${OUTPUTDIR_FORCED}" ] ; then
        [ -d "${INPUT}" ] && { OUTPUTDIR="${INPUT}/objects/access/pdf_1" && LOGDIR="${INPUT}/metadata/logs" ;};
        [ -f "${INPUT}" ] && { OUTPUTDIR=$(dirname "${INPUT}")"/access/pdf_1" && LOGDIR="$(dirname "${INPUT}")/access/logs" ;};
        [ ! "${OUTPUTDIR}" ] && { OUTPUTDIR="${INPUT}/objects/access/pdf_1" && LOGDIR="${INPUT}/metadata/logs" ;};
    else
        OUTPUTDIR="${OUTPUTDIR_FORCED}"
        LOGDIR="${OUTPUTDIR}/logs"
    fi
    OUTPUTDIRTEXT="${INPUT}/objects/access/txt_1"
    _run mkdir -p "${LOGDIR}"
    exec > >(tee "${LOGDIR}/$(basename "${0}")_$(_get_iso8601_c)_$(basename "${0}")_${VERSION}.txt")
    exec 2>&1

    #_find_input "${INPUT}"
    if [ -d "${INPUT}/objects" ] ; then
        SOURCEDIR="${INPUT}/objects"
    else
        _report -wt "No source objects folder"
        _writeerrorlog "makepdf" "No source objects folder provided. The script could not continue."
        exit 1
    fi
    MEDIAID=$(basename "${INPUT}" | cut -d. -f1)

    # set up output
    _log -b
    OUTPUT="${OUTPUTDIR}/${MEDIAID%.*}.pdf"
    [ -s "${OUTPUT}" ] && { _report -wt "WARNING ${OUTPUT} already exists, skipping process" ; shift ; continue ;};
    _run mkdir -p "${OUTPUTDIR}"

    TMP_MAKEPDF_DIR="${INPUT}/tmp"
    TMP_JPG_DIR="${TMP_MAKEPDF_DIR}/jpgs"
    _run mkdir -p "${TMP_MAKEPDF_DIR}" "${TMP_JPG_DIR}" "${OUTPUTDIRTEXT}"

    for TIF in $(find "${SOURCEDIR}" -maxdepth 1 -mindepth 1 -iname "*.tif" -type f | sort) ; do
        tifname="$(basename "${TIF}")"
        _report -dt "Working on ${tifname}..."
        pageno="$(echo "${tifname}" | cut -d_ -f2 | cut -d. -f1)"
        # check that tif name adheres to filename pattern and rename if not
        if [[ "${#pageno}" != "3" ]] ; then
            tifdir="$(dirname "${TIF}")"
            basename="$(echo "${tifname}" | cut -d_ -f1 | cut -d. -f2)"
            EXTENSION="${tifname##*.}"
            pageno="$(echo "0000${pageno}" | tail -c 4 | head -c 3)"
            _report -dt "Renaming to fit image sequence naming pattern."
            NEW_TIF_NAME="${tifdir}/${basename}_${pageno}.${EXTENSION}"
            mv -v -n "${TIF}" "${NEW_TIF_NAME}"
            TIF="${NEW_TIF_NAME}"
        fi
        TIF_BASE_NAME="$(basename "${TIF%.*}")"
        JPG_NAME="${TMP_JPG_DIR}/${TIF_BASE_NAME}.jpg"
        if [[ ! -s "${JPG_NAME}" ]] ; then
            ffmpeg -hide_banner -nostdin -i "${TIF}" -pix_fmt yuvj420p "${JPG_NAME}"
        fi
        TESSERACT_CONFIG=(-c tessedit_char_whitelist="abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789!@#$%^\&*(){}[]\|\"':;?/>.<,~\` " -c textord_min_linesize=2.25 -c preserve_interword_spaces=1)
        tesseract "${JPG_NAME}" "${TMP_JPG_DIR}/${TIF_BASE_NAME}" -l eng --psm 4 "${TESSERACT_CONFIG[@]}" pdf
        tesseract "${JPG_NAME}" "${TMP_JPG_DIR}/${TIF_BASE_NAME}" -l eng --psm 4 "${TESSERACT_CONFIG[@]}" txt
    done

    _report -dt "Checking for PBCore data"
    SCRIPT_TITLE=$(fmpbcore "${MEDIAID}" | xmlstarlet 'select' -N "p=http://www.pbcore.org/PBCore/PBCoreNamespace.html" -t -v "/p:pbcoreCollection/p:pbcoreDescriptionDocument/p:pbcoreTitle[@titleType='Series']" -o ": " -v "/p:pbcoreCollection/p:pbcoreDescriptionDocument/p:pbcoreTitle[@titleType='Episode']")
    if [[ -n "${SCRIPT_TITLE}" ]] ; then
        MIDDLE_OPTIONS+=(--pdftitle "${SCRIPT_TITLE}")
    fi
    SCRIPT_AUTHOR=$(fmpbcore "${MEDIAID}" | xmlstarlet 'select' -N "p=http://www.pbcore.org/PBCore/PBCoreNamespace.html" -t -m "/p:pbcoreCollection/p:pbcoreDescriptionDocument/p:pbcoreCreator" -v "p:creatorRole" -o ": " -v "p:creator" -o " ; ")
    if [[ -n "${SCRIPT_AUTHOR}" ]] ; then
        MIDDLE_OPTIONS+=(--pdfauthor "${SCRIPT_AUTHOR}")
    fi

    pdfjoin "${MIDDLE_OPTIONS[@]}" --pdfkeywords "${MEDIAID}" --fitpaper 'false' --rotateoversize 'false' --paper letter $(find "${TMP_JPG_DIR}" -name "*.pdf" | sort | xargs) --outfile "${OUTPUT}"

    find "${TMP_JPG_DIR}" -name "*.txt" | sort | while read page ; do
        PAGENAME="$(basename "${page%.*}")"
        echo "======================================" >> "${OUTPUTDIRTEXT}/${MEDIAID}.txt"
        echo "Page: ${PAGENAME}" >> "${OUTPUTDIRTEXT}/${MEDIAID}.txt"
        echo "======================================" >> "${OUTPUTDIRTEXT}/${MEDIAID}.txt"
        echo >> "${OUTPUTDIRTEXT}/${MEDIAID}.txt"
        cat "${page}" >> "${OUTPUTDIRTEXT}/${MEDIAID}.txt"
    done

    #if [ -d "${TMP_MAKEPDF_DIR}" ] ; then
    #    _run rm -rvf "${TMP_MAKEPDF_DIR}"
    #fi
    _summarize_make
    _deliver_output "${MAKEYOUTUBE_DELIVERY_EMAIL_TO}"
    _log -e
done