VTTCue: Support 'auto' for line

LayoutTests/media/track/opera/interfaces/VTTCue/line.html

Index: LayoutTests/media/track/opera/interfaces/VTTCue/line.html
diff --git a/LayoutTests/media/track/opera/interfaces/VTTCue/line.html b/LayoutTests/media/track/opera/interfaces/VTTCue/line.html
index 7df65eb338c9f0ff164fc3873eba349debcd3177..e5441da21dac82d3354a9d4c63ff7caf536299fd 100644
--- a/LayoutTests/media/track/opera/interfaces/VTTCue/line.html
+++ b/LayoutTests/media/track/opera/interfaces/VTTCue/line.html
@@ -8,31 +8,33 @@ test(function(){
     var video = document.createElement('video');
     document.body.appendChild(video);
     var c1 = new VTTCue(0, 1, 'text1');
-    assert_equals(c1.line, -1);
+    assert_equals(c1.line, 'auto');
     var track = document.createElement('track');
     var t = track.track;
     t.addCue(c1);
-    assert_equals(c1.line, -1);
+    assert_equals(c1.line, 'auto');
     video.appendChild(track);
-    assert_equals(c1.line, -1);
+    assert_equals(c1.line, 'auto');
     t.mode = 'showing';
-    assert_equals(c1.line, -1);
+    assert_equals(c1.line, 'auto');
     var c2 = new VTTCue(0, 1, 'text2');
     var track2 = document.createElement('track');
     var t2 = track2.track;
     t2.addCue(c2);
-    assert_equals(c2.line, -1);
+    assert_equals(c2.line, 'auto');
     video.appendChild(track2);
     t2.mode = 'showing';
-    assert_equals(c2.line, -2);

[philipj_slow, 2015/01/20 14:52:31]

I see that this -2 test used to fail. Was that per spec when you implemented
this in Presto, or what? (Just curious.)

[fs, 2015/01/20 16:35:16]

I'd think that "-2" is still the expected outcome of the computed line position
algorithm [1]. I think that the issue here was that VTTCue::line() just returned
m_linePosition without validating state (see VTTCue.h).

[1] http://dev.w3.org/html5/webvtt/#dfn-text-track-cue-computed-line-position

[philipj_slow, 2015/01/20 20:53:36]

I see that the TextTrackCue.line getter in Presto was the computed line
position, so I guess the spec said this at some point.

[fs, 2015/01/21 09:00:12]

Right, I missed to acknowledge that part.

-    assert_equals(c1.line, -1);
+    assert_equals(c2.line, 'auto');
+    assert_equals(c1.line, 'auto');
     c1.line = -5;
     assert_equals(c1.line, -5);

[philipj_slow, 2015/01/20 14:52:31]

Can you add a test for setting line to -1 since that used to be special? And
NaN/Infinity somewhere.

[fs, 2015/01/20 16:35:16]

Done. (And looks like we found a bindings bug...)

-    assert_equals(c2.line, -2);
+    assert_equals(c2.line, 'auto');
     c1.line = 0;
     c1.snapToLines = false;
     assert_equals(c1.line, 0);
-    assert_equals(c2.line, -2);
+    assert_equals(c2.line, 'auto');
+    assert_throws(new TypeError, function() { c1.line = 'auto\0'; });
+    assert_equals(c1.line, 0);
 }, document.title+', script-created cue');
 
 var t_parsed = async_test(document.title+', parsed cue');
@@ -44,7 +46,7 @@ t_parsed.step(function(){
         var c1 = t.track.cues[0];
         var c2 = t.track.cues[1];
         var c3 = t.track.cues[2];
-        assert_equals(c1.line, -1);
+        assert_equals(c1.line, 'auto');
         assert_equals(c2.line, 0);
         assert_equals(c3.line, 0);