ocrd-galley/ppn2ocr

#!/bin/bash
# Get OCR results as a OCR-D workspace for a given PPN

set -e  # Abort on error


show_help() {
  cat <<-EOH
Usage: $0 PPN77164308X

  Get OCR results as a OCR-D workspace for a given PPN

Options:
  --help                          Show this message and exit.
EOH
}

make_workspace () {
    ppn=$1
    workspace=$2

    # Make workspace directory
    mkdir "$workspace"
    cd "$workspace"

    # Get METS from OAI-PMH
    oai_identifier="oai%3Adigital.staatsbibliothek-berlin.de%3A$ppn"
    oai_url="https://digital.staatsbibliothek-berlin.de/oai?verb=GetRecord&metadataPrefix=mets&identifier=$oai_identifier"
    echo "$oai_url"
    curl "$oai_url" | xmlstarlet sel -t -c '//*[local-name()="mets"]' > mets.xml

    # Fix 'file:/' URLs to 'file:///'
    sed -i 's#file:/\([^/]\)#file:///\1#' mets.xml

    # Patch mets.xml to use our NFS mount
    sed -i 's#file:///goobi/tiff001/sbb/#file:///srv/digisam_images/sbb/#g' mets.xml

    # Remove LOCAL file group as we do not have access to the files
    ocrd workspace remove-group -rf --keep-files LOCAL

    # Validate workspace
    ocrd workspace validate mets.xml | grep -v "<notice>Won't download remote image"
}


# Command line parameters
OPTS=`getopt -o h --long help -- "$@"`
eval set -- "$OPTS"
while true; do
    case "$1" in
        -h|--help) show_help; exit; shift;;
        --) shift; break;;
        *) break;;
    esac
done
if [ -z "$1" ]; then show_help; exit; fi
ppn=$1


# From here, the magic happens
self_dir=`dirname $0`
self_dir=`realpath $self_dir`

make_workspace $ppn $ppn
$self_dir/run-docker-hub -I PRESENTATION --skip-validation


# TODO

# * README: Users must configure their proxy properly via environment variables;
#   This includes setting no_proxy (e.g. for use at SBB).

# my_ocrd_workflow
# ----------------
# * Need option to add volumes e.g. /srv/digisam_images

# File bugs in OCR-D
# ------------------
# * <error>PAGE-XML OCR-D-IMG-BINPAGE/OCR-D-IMG-BINPAGE_0001.xml : imageFilename '/srv/digisam_images/sbb/PPN719671574/00000001.tif' not found in METS</error>
#   -> had to use relative file names
# * Should be able to disable in workspace validate   <notice>Won't download remote image <http://content.staatsbibliothek-berlin.de/dms/PPN719671574/800/0/00000335.jpg></notice>


# sbb_textline_detector
# ---------------------
# * sbb_textline_detector slow
#   -> Support loading the models once so the OCR-D processor can profit from  processing multiple pages
# * Check what happens with the skewed textlines in SEG_LINE_0019