Changeset 1193 for Validate External Links

Timestamp:

Oct 31, 2024, 11:03:44 PM (5 weeks ago)

Author:

iritscen

Message:

ValExtLinks now recognizes private YT videos.

Location:

Validate External Links

Files:

• Development/YouTube - video private.txt (added)
• Development/YouTube bad link detection.rtf (modified) ( previous)
• validate_external_links.sh (modified) (4 diffs)

Validate External Links/validate_external_links.sh

@@ -884 +884 @@
    HAS_SUFFIX=0
 
-   # If the URL ends in something like ".php?foo=bar", strip everything from the '?' onward
+   # If the URL ends in a query string like ".php?foo=bar", strip everything from the '?' onward
    CLEAN_URL=${URL%%\?*}
 
-   # If the URL ends in something like "#section_15", strip everything from the '#' onward
+   # If the URL ends in an anchor link like "#section_15", strip everything from the '#' onward
    CLEAN_URL=${CLEAN_URL%%\#*}
 
@@ -1026 +1026 @@
 
             # If this is a YouTube link, we have to look at the actual page source to know if the video
-            # is good or not; override the link's info if it's actually NG
+            # is good or not; override the link's info if it's actually NG. Also see RD_CODES section
+            # below for duplicative code.
             if [[ $URL == *www.youtube.com* ]]; then
                PAGE_TEXT=$(curl --silent --insecure --user-agent '$AGENT' --max-time $TIMEOUT $URL)
@@ -1035 +1036 @@
                   let OK_LINKS-=1
                   let NG_LINKS+=1
-               elif [[ "$PAGE_TEXT" =~ "simpleText\":\"Video unavailable" ]]; then
+               elif [[ "$PAGE_TEXT" =~ "simpleText\":\"Video unavailable" ]] | [[ "$PAGE_TEXT" =~ "simpleText\":\"Private video" ]]; then
                   STATUS="NG"
                   CURL_CODE="404"
@@ -1123 +1124 @@
                let SKIP_SLASH_ADD+=1
             elif [ $YOUTU_BE -eq 1 ]; then
-               # We have to look at the actual page source to know if a YouTube video is good or not
-               PAGE_TEXT=$(curl --silent --insecure --user-agent '$AGENT' --max-time $TIMEOUT $NEW_URL | grep "\"simpleText\":\"Video unavailable\"")
+               # We have to look at the actual page source to know if a YouTube video is good or not.
+               # Also see OK_CODES section above for duplicative code.
+               PAGE_TEXT=$(curl --silent --insecure --user-agent '$AGENT' --max-time $TIMEOUT $NEW_URL | grep -e "simpleText\":\"Video unavailable" -e "simpleText\":\"Private video")
                if [ ! -z "$PAGE_TEXT" ]; then
                   STATUS="NG"