Merge branch 'linker_tracking_wout_refs' of https://github.com/Sefari…

…a/Sefaria-Project into linker_tracking

sefaria/model/webpage.py

@@ -160,7 +160,7 @@ def add_or_update_from_linker(data):
             webpage = WebPage(data)
             existing = False
         webpage._normalize() # to remove bad refs, so pages with empty ref list aren't saved
-        if webpage.should_be_excluded():
+        if webpage.should_be_excluded() or len(data["refs"]) == 0:  # data["refs"] is empty when linker.js finds no refs on a page:
             if existing:
                 webpage.delete()
             return "excluded"

static/test/linker_test.html

@@ -134,7 +134,7 @@ <h3>Some problematic ones</h3> <a href="https://github.com/Sefaria/Sefaria-Proje
 
         <p>This paragraph appears outside of #foo: Amos 5:6 (shouldn't link when `selector` is #foo)</p>
 
-    <script type="text/javascript" charset="utf-8" src="http://localhost:8000/linker.v2.js"></script>
+    <script type="text/javascript" charset="utf-8" src="/linker.v2.js"></script>
 
     <script>
         sefaria.link({

templates/js/linker.v1.js

@@ -232,6 +232,7 @@
 
         if (matchedTitles.length == 0) {
             console.log("No book titles found to link to Sefaria.");
+            ns._trackPage();
             return;
         }
 
@@ -363,8 +364,15 @@
     };
 
     ns._trackPage = function() {
-        if (ns.trackedMatches.length == 0) { return; }
-
+        if (ns.trackedMatches.length === 0 && ns.matches.length > 0) {
+            // we want to track page in two cases:
+            // 1) if there are trackedMatches found
+            // or 2) if there are no trackMatches found AND no matches found, as we want to send a message
+            // to api/linker_track so that the backend will delete the webpage.
+            // however, this if statement is a third case:
+            // when there are no trackMatches found but there are matches found, we don't want to track page
+            return;
+        }
         var robots = document.head.querySelector("meta[name~=robots]");
         if (robots && robots.content.includes("noindex")) { return; }
 

templates/js/linker.v2.js

@@ -417,12 +417,13 @@
         const full_text = [].reduce.call(ns.elems, (prev, current) => prev + current.textContent, "");
         ns.matchedTitles = bookTitles.filter(title => full_text.indexOf(title) > -1).filter(distinct);
 
-        if (ns.matchedTitles.length === 0) {
+         if (ns.matchedTitles.length === 0) {
             //console.log("No book titles found to link to Sefaria.");
-            return;
+            ns._trackPage();
+        }
+        else {
+            ns._getRegexesThenTexts(ns.mode);
         }
-
-        ns._getRegexesThenTexts(ns.mode);
     };
 
 
@@ -617,8 +618,15 @@
     };
 
     ns._trackPage = function() {
-        if (ns.trackedMatches.length == 0) { return; }
-
+        if (ns.trackedMatches.length === 0 && ns.matches.length > 0) {
+            // we want to track page in two cases:
+            // 1) if there are trackedMatches found
+            // or 2) if there are no trackMatches found AND no matches found, as we want to send a message
+            // to api/linker_track so that the backend will delete the webpage.
+            // however, this if statement is a third case:
+            // when there are no trackMatches found but there are matches found, we don't want to track page
+            return;
+        }
         var robots = document.head.querySelector("meta[name~=robots]");
         if (robots && robots.content.includes("noindex")) { return; }