Changeset 1158

Timestamp:

Jun 13, 2021, 10:50:43 PM (4 years ago)

Author:

iritscen

Message:

ValExtLinks: Added argument to 'curl' that prevents some sites from rejecting it. Val now skips archive.is links too when skipping archive.org links.

File:

• Validate External Links/validate_external_links.sh (modified) (7 diffs)

Validate External Links/validate_external_links.sh

@@ -39 +39 @@
 SUGGEST_SNAPSHOTS_NG=0 # query the Internet Archive for a possible snapshot URL for each NG page
 SUGGEST_SNAPSHOTS_OK=0 # query the Internet Archive for an existing snapshot of each OK page
-CHECK_ARCHIVE_LINKS=0  # check URLs under the archive.org domain
+CHECK_ARCHIVE_LINKS=0  # check URLs on archive.org and archive.is
 TAKE_PAGE_SHOT=0       # take a screenshot of each OK page
 TIMEOUT=10             # time to wait for a response when querying a site
@@ -111 +111 @@
 SKIP_SLASH_ADD=0
 SKIP_YOUTU_BE=0
-SKIP_ARCHIVE_ORG=0
+SKIP_ARCHIVES=0
 FILE_LINKS=0
 PAGE_LINKS=0
@@ -186 +186 @@
                                --record-ok-links argument.
        --check-archive-links   Check links that are already pointing to a page
-                               on the Internet Archive. In theory these links
-                               should be totally stable and not need validation.
+                               on the Internet Archive or archive.is (AKA 
+                               archive.today). In theory these links should be
+                               totally stable and not need validation.
        --take-screenshots FILE Call the Google Chrome binary at this path to
                                take screenshots of each "OK" page.
@@ -532 +533 @@
    if [ $LINKS_PROCESSED -gt 0 ]; then valPrint ctrh "$LINKS_PROCESSED processed $(pluralCheckNoun link $LINKS_PROCESSED):"; fi
    if [ $LINK_ERRORS -gt 0 ]; then valPrint ctrh "- $LINK_ERRORS $(pluralCheckNoun link $LINK_ERRORS) could not be processed"; fi
-   if [ $SKIP_ARCHIVE_ORG -gt 0 ]; then valPrint ctrh "- $SKIP_ARCHIVE_ORG Archive.org $(pluralCheckNoun link $SKIP_ARCHIVE_ORG) were not checked"; fi
+   if [ $SKIP_ARCHIVES -gt 0 ]; then valPrint ctrh "- $SKIP_ARCHIVES archive.org/archive.is $(pluralCheckNoun link $SKIP_ARCHIVES) were not checked"; fi
    if [ $LINK_PROBLEMS_TOTAL -gt 0 ]; then valPrint ctrh "- $LINK_PROBLEMS_TOTAL processed $(pluralCheckNoun link $LINK_PROBLEMS_TOTAL) had $(pluralCheckAn $LINK_PROBLEMS_TOTAL)$(pluralCheckNoun issue $LINK_PROBLEMS_TOTAL)"; fi
    if [ $LINKS_EXCEPTED -gt 0 ]; then valPrint ctr "  (excepted $LINKS_EXCEPTED link $(pluralCheckNoun issue $LINKS_EXCEPTED) from report)"; valPrint h "  (excepted $LINKS_EXCEPTED $(pluralCheckNoun link $LINKS_EXCEPTED) from report)"; fi
@@ -677 +678 @@
 if [ $SHOW_YT_RD -eq 1 ]; then valPrint ctrh "No"; else valPrint ctrh "Yes"; fi
 
-valPrint ctrhn "Check archive.org links: "
+valPrint ctrhn "Check archive.org and archive.is links: "
 if [ $CHECK_ARCHIVE_LINKS -eq 1 ]; then valPrint ctrh "Yes"; else valPrint ctrh "No"; fi
 
@@ -810 +811 @@
    fi
 
-   # If we're skipping Archive.org links, see if this is one
-   if [ $CHECK_ARCHIVE_LINKS -eq 0 ] && [[ $URL == *web.archive.org* ]]; then
-      valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because I have not been asked to check Wayback Machine links."
-      let SKIP_ARCHIVE_ORG+=1
+   # If we're skipping archive links, see if this is one
+   if [ $CHECK_ARCHIVE_LINKS -eq 0 ] && [[ ( $URL == *web.archive.org* || $URL == *archive.is* ) ]]; then
+      valPrint trs "Skipping URL '$URL' (found on page '$LOCAL_PAGE_PATH') because I have not been asked to check archive links."
+      let SKIP_ARCHIVES+=1
       let PAGE_LINKS+=1
       continue
@@ -917 +918 @@
    # Get response code using 'curl' to see if this link is valid; the --insecure option avoids an
    # issue with sites that require HTTPS
-   CURL_CODE=$(curl -o /dev/null --silent --insecure --head --user-agent '$AGENT' --max-time $TIMEOUT --retry 2 --write-out '%{http_code}\n' $URL)
+   CURL_CODE=$(curl -o /dev/null --silent --insecure --compressed --head --user-agent '$AGENT' --max-time $TIMEOUT --retry 2 --write-out '%{http_code}\n' $URL)
    CURL_ERR=$(echo $?)
    CURL_RESULT=$CURL_CODE