outlinks: simplify outlink parsing

The outlinks are collected as HTMLAnchorElement objects. The previous
version handled stringifying them by collecting the entire set of
objects into a single newline-delimited string, then splitting it back
up again in Python. It seems easier to just send back a JSON array of
strings and have Python iterate over them that way.

brozzler/browser.py

@@ -784,7 +784,11 @@ class Browser:
         # Now we actually do outlink extraction
         msg_id = self.send_to_chrome(
             method="Runtime.evaluate",
-            params={"expression": "__brzl_outlinksString()"},
+            params={
+                "expression": "__brzl_extractOutlinks()",
+                # returnByValue ensures we can receive an array response
+                "returnByValue": True,
+            },
         )
         self._wait_for(
             lambda: self.websock_thread.received_result(msg_id), timeout=timeout
@@ -798,7 +802,7 @@ class Browser:
         ):
             if message["result"]["result"]["value"]:
                 out = []
-                for link in message["result"]["result"]["value"].split("\n"):
+                for link in message["result"]["result"]["value"]:
                     try:
                         out.append(str(urlcanon.whatwg(link)))
                     except AddressValueError:

brozzler/js-templates/extract-outlinks.js

@@ -36,6 +36,6 @@ var __brzl_compileOutlinks = function(frame) {
 
     return outlinks;
 }
-var __brzl_outlinksString = function() {
-    return __brzl_compileOutlinks(window).join('\n');
+var __brzl_extractOutlinks = function() {
+    return __brzl_compileOutlinks(window).map(el => el.toString());
 }