Extract only visible and clickable links

2 files changed, 29 insertions, 4 deletions

diff --git a/crocoite/controller.py b/crocoite/controller.py
index 113c139..bc6f948 100644
--- a/crocoite/controller.py
+++ b/crocoite/controller.py
@@ -67,7 +67,7 @@ class SinglePageController:
         self.logger = logger
 
     def run (self):
-        ret = {'stats': None, 'links': None}
+        ret = {'stats': None, 'links': []}
 
         with self.service as browser:
             browser = pychrome.Browser (url=browser)
diff --git a/crocoite/data/extract-links.js b/crocoite/data/extract-links.js
index f2a37aa..5199a63 100644
--- a/crocoite/data/extract-links.js
+++ b/crocoite/data/extract-links.js
@@ -1,9 +1,34 @@
 /*	Extract links from a page
  */
+
+/* --- copy&paste from click.js --- */
+/*	Element is visible if itself and all of its parents are
+ */
+function isVisible (o) {
+	if (o === null || !(o instanceof Element)) {
+		return true;
+	}
+	let style = window.getComputedStyle (o);
+	if ('parentNode' in o) {
+		return style.display !== 'none' && isVisible (o.parentNode);
+	} else {
+		return style.display !== 'none';
+	}
+}
+
+/*	Elements are considered clickable if they are a) visible and b) not
+ *	disabled
+ */
+function isClickable (o) {
+	return !o.hasAttribute ('disabled') && isVisible (o);
+}
+/* --- end copy&paste */
+
 let x = document.body.querySelectorAll('a[href]');
 let ret = [];
-let index = 0;
-for( index=0; index < x.length; index++ ) {
-   ret.push (x[index].href);
+for (let i=0; i < x.length; i++) {
+	if (isClickable (x[i])) {
+		ret.push (x[i].href);
+	}
 }
 ret; /* immediately return results, for use with Runtime.evaluate() */