Help find prime numbers for Primel https://converged.yt/primel/

primel.sh

#!/bin/bash

if test ! -e 5_digit_primes.txt.xz; then
  echo 5_digit_primes.txt.xz does not exist
  echo $0 requires this file
  echo Please generate it with the following command
  echo "seq 10007 99991|factor|grep '^\([^:]*\): \1$'|cut -d: -f1|xz -9e >5_digit_primes.txt.xz"
  exit 2
fi

if test -z "$REG" -a -z "$XREG" -a -z "$SKIP"; then
  echo Start with 25693
  echo
  echo Examples:
  echo REG=..491 XREG=3.... SKIP=256870 $0
  echo
  echo REG=9.... XREG=.[35]5.3 SKIP=2681 $0
  echo
  echo 'REG  = regex search - correct numbers'
  echo XREG = regex remove - misplaced numbers
  echo SKIP = numbers to remove - not in answer
  exit 1
fi

COMMAND='xzcat 5_digit_primes.txt.xz'

if test -n "$REG"; then
  COMMAND+="|grep ^$REG"
fi

if test -n "$XREG"; then
  KEEP="$(echo $XREG|tr -d '.[]^'|fold -w1|sort -u |tr -d '\n')"
  SEARCH=
  for ((i=1;i<=${#XREG};i++)); do
    if test "${XREG:$i-1:1}" = '['; then
      SEARCH+='[^'
      ((i++))
      while test "${XREG:$i-1:1}" != ']'; do
        SEARCH+="${XREG:$i-1:1}"
        ((i++))
      done
      SEARCH+=']'
    elif test "${XREG:$i-1:1}" != '.'; then
      SEARCH+='[^'${XREG:$i-1:1}']'
    else
      SEARCH+=.
    fi
  done
  if test -n "$SEARCH"; then
    COMMAND+="|grep -E '$SEARCH'"
  fi
fi

if test -n "$KEEP"; then
  COMMAND+="|awk '"
  for ((i=1;i<=${#KEEP};i++)); do
    COMMAND+="/${KEEP:$i-1:1}/"
    if test $i -ne ${#KEEP}; then
      COMMAND+=' && '
    fi
  done
  COMMAND+="'"
fi

if test -n "$SKIP"; then
  PSKIP=$(echo $SKIP|sed 's/\B/|/g')
  COMMAND+="|grep -Ev '$PSKIP'"
fi

eval $COMMAND

I don't think I would do it, but since the prime numbers are generated so quickly and everything is done in 1 pass, you could generate them on the fly each time and save 8k of disk space. To do this, delete (or comment out) lines 3 through 9. Replace the old line 25

COMMAND='xzcat 5_digit_primes.txt.xz'

with

COMMAND="seq 10007 99991|factor|grep '^\([^:]*\): \1$'|cut -d: -f1"