55475 - Horizontal rule <hr> tags with noshade render oddly

Status: RESOLVED WORKSFORME

(Core :: Layout, defect)

Description

[John Hayward-Warburton]

On http://hadrian.ods.org/hr/ (a test case), the <hr noshade> element has a
strange, small, disjoined blob on the end of an otherwise decently rendered
horizontal rule.

Linux nightly build 2000100509.

Also, the style="color: red" directive doesn't work.

Comment 1

[Doron Rosenberg (IBM)]

html element

Comment 2

[John Hayward-Warburton]

Due to dead DNS, test case (just a single <hr noshade>) is at
http://212.56.123.232/hr/

Comment 3

[Randell Jesup [:jesup] (needinfo me)]

The calculations for the rounded endcaps on noshade HR's is off.  Whether you
have a gap between the center section seems to depend on whether the HR has an
even or odd width.  Also, if you make really big HR's (size=70 say), and magnify
them, you'll see the endcaps aren't aligned properly with the center section.

I say dump the rounded endcaps personally.  Or fix the calculations for them, I
don't care.

See nsHRFrame.cpp.

This was also discussed in passing in the resolution of the "HR's are broken by
-moz-bg-inset" bug recently fixed.

Confirming (I hadn't gotten around to reporting this myself).  FreeBSD 4.1
20001004xx

Comment 4

[Loco]

Ack, was putting in some info to confirm,  Lost my changes.  Oh, well.  What I 
saw on commercial branch bits:
Mac (10-05-13):
***/******
***/***
***/***
Linux (10-06-10):
***/***
***/*** *
***/*** *
Win NT (10-05-08):
 ***/***
****/****
 ***/***

So Win seems to be fine, but Mac and Linux as way off.

Comment 5

[Kevin McCluskey (gone)]

Dividing up claytons bugs to triage

Comment 6

[rods (gone)]

buster, I seem to remember you working on hr rendering - reassigning

Comment 7

[buster]

the rounded corners are what Nav4 does, so Kipp added them to HRs.  But IE
doesn't do them, and they don't effect layout at all, they're just a painting
phenomena.  Since we don't do them right (and neither does Nav4!), I'd suggest
we pull the code that does them.  It's 100% trivial.  Patch attached
momentarily.  This isn't a showstopper, but it cleans up some sloppiness that's
a bit embarassing.
Time to fix: 5 minutes.
Risk to fix: 0 (does not change layout at all, uses a well-tested code path in
all cases, rather than just some cases.)
Benefit: layout looks better on Mac and Linux.

Comment 8

[buster]

Attachment: trivial fix, remove buggy arc-drawing code

Comment 9

[buster]

per private conversation with jar, milestone -> Future.  This will not make NS6 RTM.

Comment 10

[Loco]

Low risk per buster's comment-can we put this into the trunk?

Comment 11

[Hixie (not reading bugmail)]

Reassigning QA Contact for all open and unverified bugs previously under Lorca's
care to Gerardo as per phone conversation this morning.

Comment 12

[bsharma]

qa contact updated.

Comment 13

[Christian Mattar]

This patch still looks good to me.
jst, roc+moz, can you give [s]r to this one-liner?

Comment 14

[Robert O'Callahan (:roc) (email my personal email if necessary)]

sr=roc+moz

Comment 15

[Hixie (not reading bugmail)]

Woah. I have a better fix in bug 38370 that doesn't get rid of the arcs, and 
has no blobs. I would recommend _not_ checking this in, and instead fixing the
bugs that block the checking in of the fix to bug 38370.

Comment 16

[Christopher Aillon (sabbatical, not receiving bugmail)]

*** Bug 97987 has been marked as a duplicate of this bug. ***

Comment 17

[Heikki Toivonen (remove -bugzilla when emailing directly)]

SPAM. HTML Element component is deprecated, changing to Layout component. See
bug 88132 for details.

Comment 18

[Heikki Toivonen (remove -bugzilla when emailing directly)]

Come on Bugzilla, you can do it...

Comment 19

[Bill McGonigle (not currently reading bugmail; please contact directly)]

An <hr noshade> tag used to (not so old Mozillas, NS4, etc) render as a black
line.  It's rendering as a gray line now (2001090605-macosX-trunk).

Comment 20

[Kevin McCluskey (gone)]

Build reassigning Buster's bugs to Marc.

Comment 21

[timeless]

*** Bug 106606 has been marked as a duplicate of this bug. ***

Comment 22

[Adam Hauner]

-> [e-mail]

Comment 23

[Bruce Davidson]

Attachment: Test case showing that <hr noshade="noshade"> renders okay

Comment 24

[Bruce Davidson]

Came across this bug whilst searching for dupes for bug 216084 (it isn't a
dupe). Realised this is probably fixed. The old test page isn't accessible any
more but I've created a test case using the info in the bug. That appears to
render fine with the noshade attribute specified - there are no "disjoined
blobs" anywhere in the latest nightly (2003081303).

Marking as works for me.

(Note that the styling of <hr>s is a whole separate issue...

Comment 25

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Probably fixed by bug 38370?