ViewVC Help
View File | Revision Log | View Changeset | Root Listing
root/Oni2/Validate External Links/validate_external_links.sh
(Generate patch)

Comparing Validate External Links/validate_external_links.sh (file contents):
Revision 1190 by iritscen, Sun Jun 23 23:36:00 2024 UTC vs.
Revision 1193 by iritscen, Thu Oct 31 22:03:44 2024 UTC

# Line 883 | Line 883 | for LINE in `cat "$LINKS_FILE"`; do
883     # URL ends in a suffix
884     HAS_SUFFIX=0
885  
886 <   # If the URL ends in something like ".php?foo=bar", strip everything from the '?' onward
886 >   # If the URL ends in a query string like ".php?foo=bar", strip everything from the '?' onward
887     CLEAN_URL=${URL%%\?*}
888  
889 <   # If the URL ends in something like "#section_15", strip everything from the '#' onward
889 >   # If the URL ends in an anchor link like "#section_15", strip everything from the '#' onward
890     CLEAN_URL=${CLEAN_URL%%\#*}
891  
892     # 'sed' cannot handle Unicode in my Bash shell, so skip non-ASCII URL and make reader check it
# Line 1025 | Line 1025 | for LINE in `cat "$LINKS_FILE"`; do
1025              let OK_LINKS+=1
1026  
1027              # If this is a YouTube link, we have to look at the actual page source to know if the video
1028 <            # is good or not; override the link's info if it's actually NG
1028 >            # is good or not; override the link's info if it's actually NG. Also see RD_CODES section
1029 >            # below for duplicative code.
1030              if [[ $URL == *www.youtube.com* ]]; then
1031                 PAGE_TEXT=$(curl --silent --insecure --user-agent '$AGENT' --max-time $TIMEOUT $URL)
1032                 CURL_ERR=$(echo $?)
# Line 1034 | Line 1035 | for LINE in `cat "$LINKS_FILE"`; do
1035                    CURL_RESULT="000-$CURL_ERR"
1036                    let OK_LINKS-=1
1037                    let NG_LINKS+=1
1038 <               elif [[ "$PAGE_TEXT" =~ "simpleText\":\"Video unavailable" ]]; then
1038 >               elif [[ "$PAGE_TEXT" =~ "simpleText\":\"Video unavailable" ]] | [[ "$PAGE_TEXT" =~ "simpleText\":\"Private video" ]]; then
1039                    STATUS="NG"
1040                    CURL_CODE="404"
1041                    CURL_RESULT=$CURL_CODE
# Line 1122 | Line 1123 | for LINE in `cat "$LINKS_FILE"`; do
1123                 let OK_LINKS+=1
1124                 let SKIP_SLASH_ADD+=1
1125              elif [ $YOUTU_BE -eq 1 ]; then
1126 <               # We have to look at the actual page source to know if a YouTube video is good or not
1127 <               PAGE_TEXT=$(curl --silent --insecure --user-agent '$AGENT' --max-time $TIMEOUT $NEW_URL | grep "\"simpleText\":\"Video unavailable\"")
1126 >               # We have to look at the actual page source to know if a YouTube video is good or not.
1127 >               # Also see OK_CODES section above for duplicative code.
1128 >               PAGE_TEXT=$(curl --silent --insecure --user-agent '$AGENT' --max-time $TIMEOUT $NEW_URL | grep -e "simpleText\":\"Video unavailable" -e "simpleText\":\"Private video")
1129                 if [ ! -z "$PAGE_TEXT" ]; then
1130                    STATUS="NG"
1131                    let NG_LINKS+=1

Diff Legend

– Removed lines
+ Added lines
< Changed lines (old)
> Changed lines (new)