Advanced link checker. Takes in a path to a directory, then finds all urls in all the files inside this directory and all it's subdirectories. Then checks if urls are valid. Has nice progressbar.

check_links.sh

#!/bin/bash

if [[ -z "$1" ]]
then
  echo "Specify path as argument!"
  exit
fi

ERRORS=()
COUNT=1

function showProgress {
  TEXT=""
  for i in $(eval echo "{1..$2}"); do
    if [[ "$i" -le "$1" ]]; then
      TEXT+="|"
    else
      TEXT+="."
    fi
  done
  echo -ne "Checking links $1/$2 [$TEXT]\r"
}

function showErrors {
  if [ ${#ERRORS[@]} -eq 0 ]; then
    echo -e "\n/---NO ERRORS FOUND!---/\n"
  else
    echo -e "\n/---THERE ARE [${#ERRORS[@]}] ERRORS---/\n"
    for err in "${ERRORS[@]}"; do
      echo -e "$err"
    done
  fi
}

function parseFiles {
  FILES=($(find $1 -type f))
  for f in "${FILES[@]}"; do
    if [[ -d $f ]]; then
      echo "||| DIRECTORY ||| $f"
      parseFiles $f
    fi

    echo -e "<--\tFile [ $f ]\t-->"
    URLS=($(grep -oE '\b(https?|ftp|file)://[-A-Za-z0-9+&@#/%?=~_|!:,.;]*[-A-Za-z0-9+&@#/%=~_|]' $f))
    COUNT=1
    for i in "${URLS[@]}"
    do
      makeRequest $i $f
    done
    wait
    echo -ne "\n"
  done

  wait
  showErrors
}

function makeRequest {
  if [[ $1 == *http* ]]
  then
    # request without following the redirects
    RESPONSE=($(curl -s -w "%{http_code} %{url_effective}\\n" $1 -o /dev/null))
    if [[ ${RESPONSE::1} == "4" ]]
    then
      echo -ne "\n"
      echo -e "$RESPONSE \tURL[ $1 ]"
      ERRORS+=("$RESPONSE \n\tURL[ $1 ] \n\tIN [$2] ")
    fi
    # To follow redirects use
    #curl -sL -w "%{http_code} %{url_effective}\\n" $i -o /dev/null
  fi
  showProgress $COUNT ${#URLS[@]}

  COUNT=$((COUNT+1))
}

parseFiles $1