Acid2 - Row 1 (again) and Row 14

First, some clarifications. Some of you may wonder why the fixes I mentioned in my last post aren't in 8.01 Preview 1, build 7583, since they were made in build 7570. That preview is based on a different working branch, where the Acid2 fixes weren't merged.

Also in my last post, I mentioned that the fix for max/min-width problems won't be in 8.0x. While fixing Acid2 bugs is a priority, so is avoiding regressions and Web site compatibility problems. The fewer changes we include, the lower the probability of regressions. That fix will be in next major rendering engine update (which we're already working on). Some of the Acid2 fixes will also have to wait until that update, as they require major code updates.

Build 7586 brought two more Acid2 fixes. The first fix is again for row 1. The W3C HTML 4.01 spec. says that end tags for P elements are optional. In Opera, if a TABLE element followed a P element, the P wasn't closed and the TABLE was considered a child of the P, rather than a sibling. Row 1 now displays correctly. Check this test case to see how your browser handles it.

The second fix in this build is related to anonymous table object creation. In Opera, if a table existed in an anonymous table row, then the table row was terminated after that table. Any additional cells that should have been in that row created a new anonymous table row. Check this test case to see how your browser handles it. Row 14 is now looking a lot better, but still isn't quite right.

Here's the updated rendering with these two bugs fixed:

Screenshot of the Acid2 test