See related bug https://bugs.webkit.org/show_bug.cgi?id=133198. Fixing this requires a somewhat heavyweight refactoring.
<rdar://problem/17166620>
Created attachment 233097 [details] Patch
Comment on attachment 233097 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=233097&action=review > Source/WebCore/platform/text/ios/LocalizedDateCache.mm:96 > + if (m_font == font && m_maxWidthMap.contains(key)) { > + cachedResult = m_maxWidthMap.get(key); > + return Vector<String>(); > + } This conflation of caching max widths per type (globally)? and the getting of list of strings is super confusing, and I don't understand the global per-type cache.
Created attachment 233255 [details] Patch
Attachment 233255 [details] did not pass style-queue: ERROR: Source/WebCore/rendering/RenderThemeIOS.mm:566: This { should be at the end of the previous line [whitespace/braces] [4] Total errors found: 1 in 16 files If any of these errors are false positives, please file a bug against check-webkit-style.
Comment on attachment 233255 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=233255&action=review > Source/WebCore/platform/text/ios/LocalizedDateCache.h:43 > + virtual ~MeasureTextClient() { } Not sure you want this. See https://bugs.webkit.org/show_bug.cgi?id=133908#c3
Just talked to Anders, we decided to keep it in. http://trac.webkit.org/changeset/170072