A workaround to python_thai_ocr crashing on some of my pdfs. Inside a folder of images, OCR each image and append the output to the argument passed to the script. Example usage: ~/scripts/thai_images2ocr ocr.txt

thai_images2ocr

#!/bin/bash

# first argument is the output file
output=$1

# remove the output file if it already exists
rm -f "$output"

# create the output file
touch "$output"

# page counter
i=0

# process each image in the current directory
shopt -s nocaseglob
shopt -s nullglob
for f in *.{jpg,jpeg,tiff,bmp,png}
do
    # increase the page counter
    i=$((i+1))

    # process the image and save output to temporary file
    # uses https://github.com/nanonymoussu/python_thai_ocr
    python $HOME/python_thai_ocr/main.py "$f" temp

    # add page header
    echo -e "✧˖°─ .✦──── ･ ｡ﾟ⟡ ☽ Page ${i} ☾ ⟡ ˚｡ ･ ────✦.─ °˖✧\n" | cat >> "$output" # cute version (more at https://emojicombos.com/divider)
    # echo -e "===============Page ${i}===============\n"  | cat >> "$output" # normal version

    # add page content
    cat temp >> "$output"

    # add empty lines at end of page
    echo -e "\n\n" | cat >> "$output"

done

# remove temporary file
rm temp