Update lint_urls.sh (#10919)

To match
[PyTorch](https://github.com/pytorch/pytorch/blob/main/scripts/lint_urls.sh)

Author: shoumikhin

scripts/lint_urls.sh

@@ -7,6 +7,8 @@
 
 set -euo pipefail
 
+trap 'kill 0' SIGINT
+
 status=0
 green='\e[1;32m'; red='\e[1;31m'; cyan='\e[1;36m'; yellow='\e[1;33m'; reset='\e[0m'
 user_agent="Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/110.0.0.0 Safari/537.36"
@@ -21,9 +23,11 @@ while IFS=: read -r filepath url; do
   (
     code=$(curl -k -gsLm30 --retry 3 --retry-delay 3 --retry-connrefused -o /dev/null -w "%{http_code}" -I "$url") || code=000
     if [ "$code" -lt 200 ] || [ "$code" -ge 400 ]; then
+      sleep 1
       code=$(curl -k -gsLm30 --retry 3 --retry-delay 3 --retry-connrefused -o /dev/null -w "%{http_code}" -r 0-0 -A "$user_agent" "$url") || code=000
     fi
     if [ "$code" -lt 200 ] || [ "$code" -ge 400 ]; then
+      sleep 1
       request_id=$(curl -sS -G -H 'Accept: application/json' \
         --data-urlencode "host=$url" \
         --data-urlencode "max_nodes=1" \
@@ -45,11 +49,16 @@ while IFS=: read -r filepath url; do
         done
       fi
     fi
+    # Treat Cloudflare JS-challenge and rate-limit as success.
+    if [[ "$code" == "403" || "$code" == "429" || "$code" == "503" ]]; then
+      printf "${yellow}WARN %s${reset} ${cyan}%s${reset} %s\n" "$code" "$url" "$filepath"
+      exit 0
+    fi
     if [ "$code" -lt 200 ] || [ "$code" -ge 400 ]; then
-      printf "${red}%s${reset} ${yellow}%s${reset} %s\n" "$code" "$url" "$filepath" >&2
+      printf "${red}FAIL %s${reset} ${yellow}%s${reset} %s\n" "$code" "$url" "$filepath" >&2
       exit 1
     else
-      printf "${green}%s${reset} ${cyan}%s${reset} %s\n" "$code" "$url" "$filepath"
+      printf "${green} OK  %s${reset} ${cyan}%s${reset} %s\n" "$code" "$url" "$filepath"
       exit 0
     fi
   ) &
@@ -58,7 +67,7 @@ while IFS=: read -r filepath url; do
     sleep 1
   done
  done < <(
-  pattern='(?!.*@lint-ignore)(?<!git\+)(?<!\$\{)https?://(?![^\s<>\")]*[<>\{\}\$])[^[:space:]<>")\[\]\\]+'
+  pattern='(?!.*@lint-ignore)(?<!git\+)(?<!\$\{)https?://(?![^/]*@)(?![^\s<>\")]*[<>\{\}\$])[^[:space:]<>")\[\]\\|]+'
   excludes=(
     ':(exclude,glob)**/.*'
     ':(exclude,glob)**/*.lock'
@@ -77,6 +86,7 @@ while IFS=: read -r filepath url; do
   git --no-pager grep --no-color -I -P -o "$pattern" -- "${paths[@]}" "${excludes[@]}" \
   | sed -E 's/[^/[:alnum:]]+$//' \
   | grep -Ev '://(0\.0\.0\.0|127\.0\.0\.1|localhost)([:/])' \
+  | grep -Ev '://[0-9]+\.[0-9]+\.[0-9]+\.[0-9]+' \
   | grep -Ev 'fwdproxy:8080' \
   || true
 )