2.50 KB, text/html
552 bytes, text/html
568 bytes, text/html
Remove quirk that text-decoration propagates to floats and absolutely positioned elements in quirks mode
8.86 KB, patch
|Details | Diff | Splinter Review|
Created attachment 8550699 [details] index.html User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:35.0) Gecko/20100101 Firefox/35.0 Build ID: 20150108202552 Steps to reproduce: i have 2 nested spans the outer defines text-decoration:underline the inner specifies no text-decoration or may specify text-decoration:none the inner span is also position:absolute; Actual results: text-decoration:underline is applied to the inner span Expected results: text-decoration:underline should neither be applied nor propagated to the inner span see http://www.w3.org/TR/CSS21/text.html#decoration see Google Chrome
I don't see any text underlined in the attached testcase, with either Firefox 35 or a current nightly. Reporter, do you see the problem in safe mode?
click on the blue Z80 to make the inner span visible, hover over the inner span. yes, also with -safe-mode.
This is because the page is in quirks mode. In quirks mode, we propagate text-decoration to absolutely-positioned and floating descendants, which we do not to in standards mode. If other browsers don't have that quirks mode behavior, we could probably remove it. (The code is at the very end of nsTextFrame::GetTextDecorations.)
Status: UNCONFIRMED → NEW
Ever confirmed: true
OS: Mac OS X → All
Hardware: x86 → All
Summary: text-decoration:underline is propagated to absolutely positioned inner span → text-decoration is propagated to absolutely positioned descendants in quirks mode
Chrome and WebKit don't have the quirk; we do. It might be worth testing IE.
Summary: text-decoration is propagated to absolutely positioned descendants in quirks mode → text-decoration is propagated to absolutely positioned and floated descendants in quirks mode
thanks for the quick examination. what makes my test case fall into "quirks mode"? I'd like to fix this then. the test case and the original page are validated html401 / css3.
Thanks for the hint. ;-) I check my pages with http://validator.w3.org/check but it didn't complain about the non-compliant doctype.
IE11 doesn't have the quirk either, so we should just remove it.
Assignee: nobody → dbaron
Status: NEW → ASSIGNED
Created attachment 8552114 [details] [diff] [review] Remove quirk that text-decoration propagates to floats and absolutely positioned elements in quirks mode Note that the style element addition to underline-block-propagation-2-quirks.html and its reference is making a change already made in the -standards version, and the other change is reflecting a quirk that is not removed, and that is interoperable.
Attachment #8552114 - Flags: review?(dholbert)
Comment on attachment 8552114 [details] [diff] [review] Remove quirk that text-decoration propagates to floats and absolutely positioned elements in quirks mode IIUC, all the changes to underline-block-propagation-2* (mentioned in comment 12) are orthogonal to the actual code-change here, so it might be clearer to make those changes (and the reftest.list tweak for that testcase) in a separate "part 2" / "followup" commit, to avoid the implication that that test is changing to accommodate for the code-change here. r=me either way though
Attachment #8552114 - Flags: review?(dholbert) → review+
Status: ASSIGNED → RESOLVED
Last Resolved: 3 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla38
You need to log in before you can comment on or make changes to this bug.