Global link hints; revert b7535a604954b5873d825eb66bfecd08f1f2c99b.

Here's why:

- b7535a604954b5873d825eb66bfecd08f1f2c99b is complicated and complex (O(n^2)).
- With experience, it is not obviously better than what was there
  before, and in some cases it's worse.
- b7535a604954b5873d825eb66bfecd08f1f2c99b selects way too much text in
  some cases; e.g. on Google Inbox, it selects text lengths in the tens
  of thousands of characters.  That cannot be useful.
- With global hints, this extra cost (resulting from passing large
  objects between frames) is significant and noticable.
- The simpler approach of accounting for text length when scoring
  filtered hints (tweaked here: 5cbc5ad702a01a81b98f8c82edb3b6d227c2c7b5)
  works better in practice.

1 files changed, 2 insertions, 3 deletions

diff --git a/content_scripts/link_hints.coffee b/content_scripts/link_hints.coffee
index 32d67ab9..fae255c6 100644
--- a/content_scripts/link_hints.coffee
+++ b/content_scripts/link_hints.coffee
@@ -321,7 +321,6 @@ LocalHints =
 
     hint.hasHref = hint.element.href? for hint in localHints
     if Settings.get "filterLinkHints"
-      DomUtils.textContent.reset()
       @generateLabelMap()
       extend hint, @generateLinkText hint.element for hint in localHints
 
@@ -354,14 +353,14 @@ LocalHints =
         linkText = element.value
         if not linkText and 'placeholder' of element
           linkText = element.placeholder
-      # check if there is an image embedded in the <a> tag
+    # Check if there is an image embedded in the <a> tag.
     else if (nodeName == "a" && !element.textContent.trim() &&
         element.firstElementChild &&
         element.firstElementChild.nodeName.toLowerCase() == "img")
       linkText = element.firstElementChild.alt || element.firstElementChild.title
       showLinkText = true if (linkText)
     else
-      linkText = DomUtils.textContent.get element
+      linkText = (element.textContent.trim() || element.innerHTML.trim())[...512]
 
     {linkText, showLinkText}