A shell script to extract text from PDF on macOS -- https://doratex.hatenablog.jp/entry/20230705/1688554958

pdf2text.sh

#!/bin/bash

SCRIPTNAME=$(basename "$0")

function realpath () {
  f=$@;
  if [ -d "$f" ]; then
    base="";
    dir="$f";
  else
    base="/$(basename "$f")";
    dir=$(dirname "$f");
  fi;
  dir=$(cd "$dir" && /bin/pwd);
  echo "$dir$base"
}

function PDF2Text () {
    osascript \
       -e 'use framework "Quartz"' \
       -e "global CA" \
       -e "set CA to current application" \
       -e "on pdf2text(filePath)" \
       -e "set doc to CA's PDFDocument's alloc's initWithURL:(CA's NSURL's fileURLWithPath:filePath)" \
       -e "set pageCount to doc's pageCount" \
       -e "set resultTexts to CA's NSMutableArray's new()" \
       -e "repeat with i from 1 to pageCount" \
       -e "(resultTexts's addObject:(((doc's pageAtIndex:(i - 1))'s attributedString)'s |string|))" \
       -e "end repeat" \
       -e "return (resultTexts's componentsJoinedByString:linefeed) as text" \
       -e "end pdf2text" \
       -e "set ocrResult to my pdf2text(\"$1\")" 2>/dev/null
}

function usage() {
    echo "Usage: $SCRIPTNAME <PATH_TO_PDF_1> <PATH_TO_PDF_2> ..."
    echo
    echo "Options:"
    echo "  -h, --help      Show help"
    echo
}

# parse arguments
declare -a args=("$@")
declare -a params=()

I=0
while [ $I -lt ${#args[@]} ]; do
    OPT="${args[$I]}"
    case $OPT in
        -h | --help )
            usage
            exit 0
            ;;
        -- | -)
            I=$(($I+1))
            while [ $I -lt ${#args[@]} ]; do
                params+=("${args[$I]}")
                I=$(($I+1))
            done
            break
            ;;
        -*)
            echo "$SCRIPTNAME: illegal option -- '$(echo $OPT | sed 's/^-*//')'" 1>&2
            exit 1
            ;;
        *)
            if [[ ! -z "$OPT" ]] && [[ ! "$OPT" =~ ^-+ ]]; then
                params+=( "$OPT" )
            fi
            ;;
    esac
    I=$(($I+1))
done

# handle invalid arguments
if [ ${#params[@]} -eq 0 ]; then
    echo "$SCRIPTNAME: too few arguments" 1>&2
    echo "Try '$SCRIPTNAME --help' for more information." 1>&2
    exit 1
fi

for FILE in "${params[@]}"; do
    PDF2Text "$(realpath $FILE)"
done