tomru/docscan
1
0
Fork 0
mirror of https://github.com/tomru/docscan.git synced 2026-03-03 06:27:21 +01:00
Code Issues Packages Projects Releases Wiki Activity

fix a few issues

Browse Source
This commit is contained in:
Thomas Ruoff
2021-01-27 23:53:37 +01:00
parent 871514a530
commit e0000e2479
1 changed files with 13 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
run.sh
View File

@@ -2,9 +2,6 @@
set -o errexit -o nounset set -o errexit -o nounset
# as suggested in https://github.com/koalaman/shellcheck/wiki/SC2044
shopt -s globstar nullglob
OCRMYPDF_CMD='docker run --rm -i ocrmypdf' OCRMYPDF_CMD='docker run --rm -i ocrmypdf'
SCR_DIR=/mnt/docscans SCR_DIR=/mnt/docscans
@@ -31,7 +28,7 @@ function ocrFile() {
--rotate-pages \ --rotate-pages \
--deskew \ --deskew \
- - \ - - \
< "$FILE" \ < "$SRC" \
> "$DST_PDF" 2> "$DST_LOG" > "$DST_PDF" 2> "$DST_LOG"
} }
@@ -41,25 +38,31 @@ flock -n 100 || { echo $LOCKFILE is locked; exit 1; }
mkdir -p "$DST_DIR" "$ARCHIVE_DIR" "$FAILED_DIR" "$LOG_DIR" mkdir -p "$DST_DIR" "$ARCHIVE_DIR" "$FAILED_DIR" "$LOG_DIR"
while IFS= read -r -d '' file while IFS= read -r -d '' FILE
do do
echo "processing $FILE ..." echo "processing $FILE ..."
if ocrFile "$FILE" "$DST_DIR" if ocrFile "$FILE"
then then
echo processed "$FILE" echo processed "$FILE"
mv "$FILE" "$ARCHIVE_DIR" mv "$FILE" "$ARCHIVE_DIR"
echo moved "$FILE" to "$ARCHIVE_DIR" echo moved "$FILE" to "$ARCHIVE_DIR"
(( PROCESSED_COUNT++ )) (( PROCESSED_COUNT+=1 ))
else else
echo failed to process "$FILE" echo failed to process "$FILE"
mv "$FILE" "$FAILED_DIR" mv "$FILE" "$FAILED_DIR"
echo moved "$FILE" to "$FAILED_DIR" echo moved "$FILE" to "$FAILED_DIR"
(( FAILED_COUNT++ )) (( FAILED_COUNT+=1 ))
fi fi
echo
done < <(find $SCR_DIR -maxdepth 1 -name '*.pdf' -print0) done < <(find $SCR_DIR -maxdepth 1 -name '*.pdf' -print0)
echo Done echo Done
echo processed $PROCESSED_COUNT PDFs echo
[ $FAILED_COUNT -gt 0 ] && echo failed on $FAILED_COUNT PDFs echo processed "$PROCESSED_COUNT" PDFs
if (( FAILED_COUNT > 0 ))
then
echo failed on "$FAILED_COUNT" PDFs
fi