Changeset 174222 in webkit for trunk/Source/WebCore/css/CSSGradientValue.cpp

Timestamp:

Oct 2, 2014, 11:35:30 AM (11 years ago)

Author:

[email protected]

Message:

Clean up loops in CSSGradientValue.cpp
​https://bugs.webkit.org/show_bug.cgi?id=137332

Reviewed by Andreas Kling.

This patch cleans up a couple of for loops by using C++11 syntax and
fixes some grammar in a comment.

No new tests, no behavior change.

• css/CSSGradientValue.cpp:

(WebCore::CSSGradientValue::gradientWithStylesResolved): Move for loops to C++11.
(WebCore::CSSGradientValue::addStops): Update grammar in comment for gradient midpoints.

File:

• trunk/Source/WebCore/css/CSSGradientValue.cpp (modified) (3 diffs)

trunk/Source/WebCore/css/CSSGradientValue.cpp

@@ -108 +108 @@
 {
     bool derived = false;
-    for (unsigned i = 0; i < m_stops.size(); i++) {
-        if (!m_stops[i].isMidpoint && styleResolver->colorFromPrimitiveValueIsDerivedFromElement(m_stops[i].m_color.get())) {
-            m_stops[i].m_colorIsDerivedFromElement = true;
+    for (auto& stop : m_stops) {
+        if (!stop.isMidpoint && styleResolver->colorFromPrimitiveValueIsDerivedFromElement(stop.m_color.get())) {
+            stop.m_colorIsDerivedFromElement = true;
             derived = true;
             break;
@@ -128 +128 @@
     }
 
-    for (unsigned i = 0; i < result->m_stops.size(); i++) {
-        if (!result->m_stops[i].isMidpoint)
-            result->m_stops[i].m_resolvedColor = styleResolver->colorFromPrimitiveValue(result->m_stops[i].m_color.get());
+    for (auto& stop : result->m_stops) {
+        if (!stop.isMidpoint)
+            stop.m_resolvedColor = styleResolver->colorFromPrimitiveValue(stop.m_color.get());
     }
 
@@ -277 +277 @@
     // Stops on the side with the most stops start midway because the curve approximates
     // a line in that region. We then add 5 more color stops on that side to minimize the change
-    // how the luminance changes at each of the colorstops. We don't have to add as many on the other side
-    // since it becomes small which increases the differentation of luminance which hides the colorstops.
-    // Even with 4 extra colorstops, it *is* possible to discern the steps when the gradient is large and has
+    // how the luminance changes at each of the color stops. We don't have to add as many on the other side
+    // since it becomes small which increases the differentation of luminance which hides the color stops.
+    // Even with 4 extra color stops, it *is* possible to discern the steps when the gradient is large and has
     // large luminance differences between midpoint and color stop. If this becomes an issue, we can consider
     // making this algorithm a bit smarter.